Radiocarpal portals.
\\n\\n
IntechOpen was founded by scientists, for scientists, in order to make book publishing accessible around the globe. Over the last two decades, this has driven Open Access (OA) book publishing whilst levelling the playing field for global academics. Through our innovative publishing model and the support of the research community, we have now published over 5,700 Open Access books and are visited online by over three million academics every month. These researchers are increasingly working in broad technology-based subjects, driving multidisciplinary academic endeavours into human health, environment, and technology.
\\n\\nBy listening to our community, and in order to serve these rapidly growing areas which lie at the core of IntechOpen's expertise, we are launching a portfolio of Open Science journals:
\\n\\nAll three journals will publish under an Open Access model and embrace Open Science policies to help support the changing needs of academics in these fast-moving research areas. There will be direct links to preprint servers and data repositories, allowing full reproducibility and rapid dissemination of published papers to help accelerate the pace of research. Each journal has renowned Editors in Chief who will work alongside a global Editorial Board, delivering robust single-blind peer review. Supported by our internal editorial teams, this will ensure our authors will receive a quick, user-friendly, and personalised publishing experience.
\\n\\n"By launching our journals portfolio we are introducing new, dedicated homes for interdisciplinary technology-focused researchers to publish their work, whilst embracing Open Science and creating a unique global home for academics to disseminate their work. We are taking a leap toward Open Science continuing and expanding our fundamental commitment to openly sharing scientific research across the world, making it available for the benefit of all." Dr. Sara Uhac, IntechOpen CEO
\\n\\n"Our aim is to promote and create better science for a better world by increasing access to information and the latest scientific developments to all scientists, innovators, entrepreneurs and students and give them the opportunity to learn, observe and contribute to knowledge creation. Open Science promotes a swifter path from research to innovation to produce new products and services." Alex Lazinica, IntechOpen founder
\\n\\nIn conclusion, Natalia Reinic Babic, Head of Journal Publishing and Open Science at IntechOpen adds:
\\n\\n“On behalf of the journal team I’d like to thank all our Editors in Chief, Editorial Boards, internal supporting teams, and our scientific community for their continuous support in making this portfolio a reality - we couldn’t have done it without you! With your support in place, we are confident these journals will become as impactful and successful as our book publishing program and bring us closer to a more open (science) future.”
\\n\\nWe invite you to visit the journals homepage and learn more about the journal’s Editorial Boards, scope and vision as all three journals are now open for submissions.
\\n\\nFeel free to share this news on social media and help us mark this memorable moment!
\\n\\n\\n"}]',published:!0,mainMedia:{caption:"",originalUrl:"/media/original/237"}},components:[{type:"htmlEditorComponent",content:'
After years of being acknowledged as the world's leading publisher of Open Access books, today, we are proud to announce we’ve successfully launched a portfolio of Open Science journals covering rapidly expanding areas of interdisciplinary research.
\n\n\n\nIntechOpen was founded by scientists, for scientists, in order to make book publishing accessible around the globe. Over the last two decades, this has driven Open Access (OA) book publishing whilst levelling the playing field for global academics. Through our innovative publishing model and the support of the research community, we have now published over 5,700 Open Access books and are visited online by over three million academics every month. These researchers are increasingly working in broad technology-based subjects, driving multidisciplinary academic endeavours into human health, environment, and technology.
\n\nBy listening to our community, and in order to serve these rapidly growing areas which lie at the core of IntechOpen's expertise, we are launching a portfolio of Open Science journals:
\n\nAll three journals will publish under an Open Access model and embrace Open Science policies to help support the changing needs of academics in these fast-moving research areas. There will be direct links to preprint servers and data repositories, allowing full reproducibility and rapid dissemination of published papers to help accelerate the pace of research. Each journal has renowned Editors in Chief who will work alongside a global Editorial Board, delivering robust single-blind peer review. Supported by our internal editorial teams, this will ensure our authors will receive a quick, user-friendly, and personalised publishing experience.
\n\n"By launching our journals portfolio we are introducing new, dedicated homes for interdisciplinary technology-focused researchers to publish their work, whilst embracing Open Science and creating a unique global home for academics to disseminate their work. We are taking a leap toward Open Science continuing and expanding our fundamental commitment to openly sharing scientific research across the world, making it available for the benefit of all." Dr. Sara Uhac, IntechOpen CEO
\n\n"Our aim is to promote and create better science for a better world by increasing access to information and the latest scientific developments to all scientists, innovators, entrepreneurs and students and give them the opportunity to learn, observe and contribute to knowledge creation. Open Science promotes a swifter path from research to innovation to produce new products and services." Alex Lazinica, IntechOpen founder
\n\nIn conclusion, Natalia Reinic Babic, Head of Journal Publishing and Open Science at IntechOpen adds:
\n\n“On behalf of the journal team I’d like to thank all our Editors in Chief, Editorial Boards, internal supporting teams, and our scientific community for their continuous support in making this portfolio a reality - we couldn’t have done it without you! With your support in place, we are confident these journals will become as impactful and successful as our book publishing program and bring us closer to a more open (science) future.”
\n\nWe invite you to visit the journals homepage and learn more about the journal’s Editorial Boards, scope and vision as all three journals are now open for submissions.
\n\nFeel free to share this news on social media and help us mark this memorable moment!
\n\n\n'}],latestNews:[{slug:"step-in-the-right-direction-intechopen-launches-a-portfolio-of-open-science-journals-20220414",title:"Step in the Right Direction: IntechOpen Launches a Portfolio of Open Science Journals"},{slug:"let-s-meet-at-london-book-fair-5-7-april-2022-olympia-london-20220321",title:"Let’s meet at London Book Fair, 5-7 April 2022, Olympia London"},{slug:"50-books-published-as-part-of-intechopen-and-knowledge-unlatched-ku-collaboration-20220316",title:"50 Books published as part of IntechOpen and Knowledge Unlatched (KU) Collaboration"},{slug:"intechopen-joins-the-united-nations-sustainable-development-goals-publishers-compact-20221702",title:"IntechOpen joins the United Nations Sustainable Development Goals Publishers Compact"},{slug:"intechopen-signs-exclusive-representation-agreement-with-lsr-libros-servicios-y-representaciones-s-a-de-c-v-20211123",title:"IntechOpen Signs Exclusive Representation Agreement with LSR Libros Servicios y Representaciones S.A. de C.V"},{slug:"intechopen-expands-partnership-with-research4life-20211110",title:"IntechOpen Expands Partnership with Research4Life"},{slug:"introducing-intechopen-book-series-a-new-publishing-format-for-oa-books-20210915",title:"Introducing IntechOpen Book Series - A New Publishing Format for OA Books"},{slug:"intechopen-identified-as-one-of-the-most-significant-contributor-to-oa-book-growth-in-doab-20210809",title:"IntechOpen Identified as One of the Most Significant Contributors to OA Book Growth in DOAB"}]},book:{item:{type:"book",id:"2287",leadTitle:null,fullTitle:"Crude Oil Exploration in the World",title:"Crude Oil Exploration in the World",subtitle:null,reviewType:"peer-reviewed",abstract:'"Crude Oil Exploration in the World" contains multidisciplinary chapters in the fields of prospection and exploration of crude oils all over the world in addition to environmental impact assessments, oil spills and marketing of crude oils.',isbn:null,printIsbn:"978-953-51-0379-0",pdfIsbn:"978-953-51-6158-5",doi:"10.5772/2676",price:119,priceEur:129,priceUsd:155,slug:"crude-oil-exploration-in-the-world",numberOfPages:232,isOpenForSubmission:!1,isInWos:1,isInBkci:!1,hash:"929c0975182e0946cc7cc5ed77cfc137",bookSignature:"Mohamed Abdel-Aziz Younes",publishedDate:"March 16th 2012",coverURL:"https://cdn.intechopen.com/books/images_new/2287.jpg",numberOfDownloads:53235,numberOfWosCitations:23,numberOfCrossrefCitations:10,numberOfCrossrefCitationsByBook:3,numberOfDimensionsCitations:38,numberOfDimensionsCitationsByBook:3,hasAltmetrics:1,numberOfTotalCitations:71,isAvailableForWebshopOrdering:!0,dateEndFirstStepPublish:"May 4th 2011",dateEndSecondStepPublish:"June 1st 2011",dateEndThirdStepPublish:"October 6th 2011",dateEndFourthStepPublish:"November 5th 2011",dateEndFifthStepPublish:"March 4th 2012",currentStepOfPublishingProcess:5,indexedIn:"1,2,3,4,5,6,7",editedByType:"Edited by",kuFlag:!1,featuredMarkup:null,editors:[{id:"104550",title:"Prof.",name:"Mohamed",middleName:"Abdel-Aziz",surname:"Younes",slug:"mohamed-younes",fullName:"Mohamed Younes",profilePictureURL:"https://mts.intechopen.com/storage/users/104550/images/system/104550.jpg",biography:"Prof. Dr. Mohamed Abdel-Aziz Younes has recently rejoined the staff of Geology Department, Faculty of Science, Alexandria University, Egypt as a Professor of Petroleum Geology and Geochemistry in 2007.\nMohamed has B. Sc. Honors (1981) in Special Geology and M. Sc. (1985) in Petroleum Geology. He obtained his Ph. D. in Petroleum Geology and Geochemistry of the Gulf of Suez in (1991) from Alexandria University, Egypt.\nHe is an active member in the American Association of Petroleum Geologists (AAPG), European Association of Organic Geochemists (EAOG), the American Association for the Advancement of Science (AAAS) and the Egyptian Geological Survey (EGS). He obtained the Alexandria University Scientific Encouragement Prize for the academic year 2003. \nMohamed is the author of several international published research papers on the Petroleum Geology and Geochemistry of the Gulf of Suez, Western Desert, Nile Delta and Offshore Mediterranean Sea and the Upper Egypt.",institutionString:null,position:null,outsideEditionCount:0,totalCites:0,totalAuthoredChapters:"3",totalChapterViews:"0",totalEditedBooks:"1",institution:{name:"Alexandria University",institutionURL:null,country:{name:"Egypt"}}}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,coeditorOne:null,coeditorTwo:null,coeditorThree:null,coeditorFour:null,coeditorFive:null,topics:[{id:"790",title:"Petroleum Engineering",slug:"engineering-environmental-engineering-petroleum-engineering"}],chapters:[{id:"32402",title:"Crude Oil Geochemistry Dependent Biomarker Distributions in the Gulf of Suez, Egypt",doi:"10.5772/35463",slug:"crude-oil-geochemistry-dependent-biomarker-distribution-in-the-gulf-of-suez-egypt",totalDownloads:5376,totalCrossrefCites:0,totalDimensionsCites:0,hasAltmetrics:0,abstract:null,signatures:"M. A. Younes",downloadPdfUrl:"/chapter/pdf-download/32402",previewPdfUrl:"/chapter/pdf-preview/32402",authors:[{id:"104550",title:"Prof.",name:"Mohamed",surname:"Younes",slug:"mohamed-younes",fullName:"Mohamed Younes"}],corrections:null},{id:"32403",title:"Hydrocarbon Potentials in the Northern Western Desert of Egypt",doi:"10.5772/37028",slug:"hydrocarbon-potentials-in-the-northern-western-desert-of-egypt",totalDownloads:8533,totalCrossrefCites:1,totalDimensionsCites:7,hasAltmetrics:0,abstract:null,signatures:"M. A. Younes",downloadPdfUrl:"/chapter/pdf-download/32403",previewPdfUrl:"/chapter/pdf-preview/32403",authors:[{id:"104550",title:"Prof.",name:"Mohamed",surname:"Younes",slug:"mohamed-younes",fullName:"Mohamed Younes"}],corrections:null},{id:"32404",title:"Crude Oil and Fractional Spillages Resulting from Exploration and Exploitation in Niger-Delta Region of Nigeria: A Review About the Environmental and Public Health Impact",doi:"10.5772/36328",slug:"enviromental-impact-of-crude-oil-exloration-and-in-niger-delta-region-of-nigeria-a-review",totalDownloads:4069,totalCrossrefCites:0,totalDimensionsCites:1,hasAltmetrics:0,abstract:null,signatures:"John Kanayochukwu Nduka, Fabian Onyeka Obumselu and Ngozi Lilian Umedum",downloadPdfUrl:"/chapter/pdf-download/32404",previewPdfUrl:"/chapter/pdf-preview/32404",authors:[{id:"107866",title:"Dr.",name:"John Kanayochukwu",surname:"Nduka",slug:"john-kanayochukwu-nduka",fullName:"John Kanayochukwu Nduka"}],corrections:null},{id:"32405",title:"Magnetic Susceptibility of Petroleum Reservoir Crude Oils in Petroleum Engineering",doi:"10.5772/36927",slug:"magnetic-susceptibility-of-petroleum-reservoir-crude-oils-in-petroleum-engineering",totalDownloads:5183,totalCrossrefCites:0,totalDimensionsCites:5,hasAltmetrics:0,abstract:null,signatures:"Oleksandr P. Ivakhnenko",downloadPdfUrl:"/chapter/pdf-download/32405",previewPdfUrl:"/chapter/pdf-preview/32405",authors:[{id:"110489",title:"Prof.",name:"Aleksandr",surname:"Ivakhnenko",slug:"aleksandr-ivakhnenko",fullName:"Aleksandr Ivakhnenko"}],corrections:null},{id:"32406",title:"Environmental Bases on the Exploitation of Crude Oil in Mexico",doi:"10.5772/35706",slug:"environmental-bases-on-the-exploitation-of-crude-oil-in-mexico",totalDownloads:3716,totalCrossrefCites:1,totalDimensionsCites:2,hasAltmetrics:0,abstract:null,signatures:"Dinora Vázquez-Luna",downloadPdfUrl:"/chapter/pdf-download/32406",previewPdfUrl:"/chapter/pdf-preview/32406",authors:[{id:"105535",title:"Dr.",name:"Dinora",surname:"Vázquez-Luna",slug:"dinora-vazquez-luna",fullName:"Dinora Vázquez-Luna"}],corrections:null},{id:"32407",title:"Spreading and Retraction of Spilled Crude Oil on Sea Water",doi:"10.5772/37392",slug:"spreading-and-retraction-of-spilled-crude-oil-on-sea-water",totalDownloads:3442,totalCrossrefCites:2,totalDimensionsCites:11,hasAltmetrics:0,abstract:null,signatures:"Koichi Takamura, Nina Loahardjo, Winoto Winoto, Jill Buckley, Norman R. Morrow, Makoto Kunieda, Yunfeng Liang and Toshifumi Matsuoka",downloadPdfUrl:"/chapter/pdf-download/32407",previewPdfUrl:"/chapter/pdf-preview/32407",authors:[{id:"112525",title:"Dr",name:"Norman",surname:"Morrow",slug:"norman-morrow",fullName:"Norman Morrow"},{id:"112695",title:"Dr.",name:"Koichi",surname:"Takamura",slug:"koichi-takamura",fullName:"Koichi Takamura"},{id:"112889",title:"Dr.",name:"Nina",surname:"Loahardjo",slug:"nina-loahardjo",fullName:"Nina Loahardjo"},{id:"112890",title:"Dr.",name:"Winoto",surname:"Winoto",slug:"winoto-winoto",fullName:"Winoto Winoto"},{id:"112891",title:"Dr.",name:"Jill",surname:"Buckley",slug:"jill-buckley",fullName:"Jill Buckley"},{id:"114293",title:"Mr.",name:"Makoto",surname:"Kunieda",slug:"makoto-kunieda",fullName:"Makoto Kunieda"},{id:"114294",title:"Dr.",name:"Yunfeng",surname:"Liang",slug:"yunfeng-liang",fullName:"Yunfeng Liang"},{id:"114296",title:"Dr.",name:"Toshifumi",surname:"Matsuoka",slug:"toshifumi-matsuoka",fullName:"Toshifumi Matsuoka"}],corrections:null},{id:"32408",title:"Fate of Subsurface Migration of Crude Oil Spill: A Review",doi:"10.5772/36456",slug:"fate-of-subsurface-migration-of-crude-oil-spill-a-review-",totalDownloads:3996,totalCrossrefCites:1,totalDimensionsCites:2,hasAltmetrics:0,abstract:null,signatures:"P. O. Youdeowei",downloadPdfUrl:"/chapter/pdf-download/32408",previewPdfUrl:"/chapter/pdf-preview/32408",authors:[{id:"108400",title:"Dr.",name:"Patrick",surname:"Youdeowei",slug:"patrick-youdeowei",fullName:"Patrick Youdeowei"}],corrections:null},{id:"32409",title:"Crude Oil Transportation: Nigerian Niger Delta Waxy Crude",doi:"10.5772/35609",slug:"crude-oil-transportation-nigerian-niger-delta-waxy-crude-oil",totalDownloads:9796,totalCrossrefCites:5,totalDimensionsCites:9,hasAltmetrics:0,abstract:null,signatures:"Elijah Taiwo, John Otolorin and Tinuade Afolabi",downloadPdfUrl:"/chapter/pdf-download/32409",previewPdfUrl:"/chapter/pdf-preview/32409",authors:[{id:"105162",title:"Dr.",name:"Elijah",surname:"Taiwo",slug:"elijah-taiwo",fullName:"Elijah Taiwo"},{id:"111892",title:"Mr.",name:"John",surname:"Otolorin",slug:"john-otolorin",fullName:"John Otolorin"},{id:"111893",title:"Dr.",name:"Tinuade",surname:"Afolabi",slug:"tinuade-afolabi",fullName:"Tinuade Afolabi"}],corrections:null},{id:"32410",title:"Degradation of Petroleum Fractions in Soil Under Natural Environment: A Gravimetric and Gas Chromatographic Analysis",doi:"10.5772/35387",slug:"degradation-of-petroleum-fractions-in-soil-under-natural-environment-a-gravimetric-and-gas-chromatog",totalDownloads:1905,totalCrossrefCites:0,totalDimensionsCites:0,hasAltmetrics:0,abstract:null,signatures:"Prahash Chandra Sarma",downloadPdfUrl:"/chapter/pdf-download/32410",previewPdfUrl:"/chapter/pdf-preview/32410",authors:[{id:"104192",title:"Dr.",name:"Prahash",surname:"Sarma",slug:"prahash-sarma",fullName:"Prahash Sarma"}],corrections:null},{id:"32411",title:"To what Extent Do Oil Prices Depend on the Value of US Dollar: Theoretical Investigation and Empirical Evidence",doi:"10.5772/35432",slug:"to-what-extent-do-oil-prices-depend-on-the-value-of-us-dollar-theoretical-investigation-and-empirica",totalDownloads:5285,totalCrossrefCites:0,totalDimensionsCites:0,hasAltmetrics:0,abstract:null,signatures:"Saleh Mothana Obadi",downloadPdfUrl:"/chapter/pdf-download/32411",previewPdfUrl:"/chapter/pdf-preview/32411",authors:[{id:"104389",title:"Prof.",name:"Saleh Mothana",surname:"Obadi",slug:"saleh-mothana-obadi",fullName:"Saleh Mothana Obadi"}],corrections:null},{id:"32412",title:"A State-of-the-Art Review of Finance Research in Physical and Financial Trading Markets in Crude Oil",doi:"10.5772/35503",slug:"a-state-of-the-art-review-of-finance-research-in-physical-and-financial-crude-oil-markets",totalDownloads:1935,totalCrossrefCites:0,totalDimensionsCites:1,hasAltmetrics:0,abstract:null,signatures:"Andrew C. Worthington",downloadPdfUrl:"/chapter/pdf-download/32412",previewPdfUrl:"/chapter/pdf-preview/32412",authors:[{id:"104743",title:"Prof.",name:"Andrew",surname:"Worthington",slug:"andrew-worthington",fullName:"Andrew Worthington"}],corrections:null}],productType:{id:"1",title:"Edited Volume",chapterContentType:"chapter",authoredCaption:"Edited by"},subseries:null,tags:null},relatedBooks:[{type:"book",id:"1589",title:"Introduction to Enhanced Oil Recovery (EOR) Processes and Bioremediation of Oil-Contaminated Sites",subtitle:null,isOpenForSubmission:!1,hash:"239203117e3001beae3d650afe00ee19",slug:"introduction-to-enhanced-oil-recovery-eor-processes-and-bioremediation-of-oil-contaminated-sites",bookSignature:"Laura Romero-Zerón",coverURL:"https://cdn.intechopen.com/books/images_new/1589.jpg",editedByType:"Edited by",editors:[{id:"109465",title:"Dr.",name:"Laura",surname:"Romero-Zerón",slug:"laura-romero-zeron",fullName:"Laura Romero-Zerón"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"8229",title:"Oil and Gas Wells",subtitle:null,isOpenForSubmission:!1,hash:"72357a3bf0f9d65e56edbde1f8250f8f",slug:"oil-and-gas-wells",bookSignature:"Sid-Ali Ouadfeul and Leila Aliouane",coverURL:"https://cdn.intechopen.com/books/images_new/8229.jpg",editedByType:"Edited by",editors:[{id:"103826",title:"Dr.",name:"Sid-Ali",surname:"Ouadfeul",slug:"sid-ali-ouadfeul",fullName:"Sid-Ali Ouadfeul"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"1591",title:"Infrared Spectroscopy",subtitle:"Materials Science, Engineering and Technology",isOpenForSubmission:!1,hash:"99b4b7b71a8caeb693ed762b40b017f4",slug:"infrared-spectroscopy-materials-science-engineering-and-technology",bookSignature:"Theophile Theophanides",coverURL:"https://cdn.intechopen.com/books/images_new/1591.jpg",editedByType:"Edited by",editors:[{id:"37194",title:"Dr.",name:"Theophile",surname:"Theophanides",slug:"theophile-theophanides",fullName:"Theophile Theophanides"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"3161",title:"Frontiers in Guided Wave Optics and Optoelectronics",subtitle:null,isOpenForSubmission:!1,hash:"deb44e9c99f82bbce1083abea743146c",slug:"frontiers-in-guided-wave-optics-and-optoelectronics",bookSignature:"Bishnu Pal",coverURL:"https://cdn.intechopen.com/books/images_new/3161.jpg",editedByType:"Edited by",editors:[{id:"4782",title:"Prof.",name:"Bishnu",surname:"Pal",slug:"bishnu-pal",fullName:"Bishnu Pal"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"3092",title:"Anopheles mosquitoes",subtitle:"New insights into malaria vectors",isOpenForSubmission:!1,hash:"c9e622485316d5e296288bf24d2b0d64",slug:"anopheles-mosquitoes-new-insights-into-malaria-vectors",bookSignature:"Sylvie Manguin",coverURL:"https://cdn.intechopen.com/books/images_new/3092.jpg",editedByType:"Edited by",editors:[{id:"50017",title:"Prof.",name:"Sylvie",surname:"Manguin",slug:"sylvie-manguin",fullName:"Sylvie Manguin"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"371",title:"Abiotic Stress in Plants",subtitle:"Mechanisms and Adaptations",isOpenForSubmission:!1,hash:"588466f487e307619849d72389178a74",slug:"abiotic-stress-in-plants-mechanisms-and-adaptations",bookSignature:"Arun Shanker and B. Venkateswarlu",coverURL:"https://cdn.intechopen.com/books/images_new/371.jpg",editedByType:"Edited by",editors:[{id:"58592",title:"Dr.",name:"Arun",surname:"Shanker",slug:"arun-shanker",fullName:"Arun Shanker"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"72",title:"Ionic Liquids",subtitle:"Theory, Properties, New Approaches",isOpenForSubmission:!1,hash:"d94ffa3cfa10505e3b1d676d46fcd3f5",slug:"ionic-liquids-theory-properties-new-approaches",bookSignature:"Alexander Kokorin",coverURL:"https://cdn.intechopen.com/books/images_new/72.jpg",editedByType:"Edited by",editors:[{id:"19816",title:"Prof.",name:"Alexander",surname:"Kokorin",slug:"alexander-kokorin",fullName:"Alexander Kokorin"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"314",title:"Regenerative Medicine and Tissue Engineering",subtitle:"Cells and Biomaterials",isOpenForSubmission:!1,hash:"bb67e80e480c86bb8315458012d65686",slug:"regenerative-medicine-and-tissue-engineering-cells-and-biomaterials",bookSignature:"Daniel Eberli",coverURL:"https://cdn.intechopen.com/books/images_new/314.jpg",editedByType:"Edited by",editors:[{id:"6495",title:"Dr.",name:"Daniel",surname:"Eberli",slug:"daniel-eberli",fullName:"Daniel Eberli"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"57",title:"Physics and Applications of Graphene",subtitle:"Experiments",isOpenForSubmission:!1,hash:"0e6622a71cf4f02f45bfdd5691e1189a",slug:"physics-and-applications-of-graphene-experiments",bookSignature:"Sergey Mikhailov",coverURL:"https://cdn.intechopen.com/books/images_new/57.jpg",editedByType:"Edited by",editors:[{id:"16042",title:"Dr.",name:"Sergey",surname:"Mikhailov",slug:"sergey-mikhailov",fullName:"Sergey Mikhailov"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"1373",title:"Ionic Liquids",subtitle:"Applications and Perspectives",isOpenForSubmission:!1,hash:"5e9ae5ae9167cde4b344e499a792c41c",slug:"ionic-liquids-applications-and-perspectives",bookSignature:"Alexander Kokorin",coverURL:"https://cdn.intechopen.com/books/images_new/1373.jpg",editedByType:"Edited by",editors:[{id:"19816",title:"Prof.",name:"Alexander",surname:"Kokorin",slug:"alexander-kokorin",fullName:"Alexander Kokorin"}],equalEditorOne:null,equalEditorTwo:null,equalEditorThree:null,productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}}],ofsBooks:[]},correction:{item:{id:"78823",slug:"erratum-covid-19-transmission-in-children-implications-for",title:"Erratum: COVID-19 Transmission in Children: Implications for Schools",doi:null,correctionPDFUrl:"https://cdn.intechopen.com/pdfs/78823.pdf",downloadPdfUrl:"/chapter/pdf-download/78823",previewPdfUrl:"/chapter/pdf-preview/78823",totalDownloads:null,totalCrossrefCites:null,bibtexUrl:"/chapter/bibtex/78823",risUrl:"/chapter/ris/78823",chapter:{id:"77986",slug:"covid-19-transmission-in-children-implications-for-schools",signatures:"Evelyn Mendoza-Torres, Franklin Torres, Wendy Rosales-Rada, Liliana Encinales, Lil Avendaño, María Fernanda Pérez, Ivana Terán, David Vergara, Estefanie Osorio-Llanes, Paige Fierbaugh, Wendy Villamizar, Aileen Y. Chang and Jairo Castellar-Lopez",dateSubmitted:"June 15th 2021",dateReviewed:"July 12th 2021",datePrePublished:"September 13th 2021",datePublished:"March 16th 2022",book:{id:"10707",title:"Primary Health Care",subtitle:null,fullTitle:"Primary Health Care",slug:"primary-health-care",publishedDate:"March 16th 2022",bookSignature:"Ayşe Emel Önal",coverURL:"https://cdn.intechopen.com/books/images_new/10707.jpg",licenceType:"CC BY 3.0",editedByType:"Edited by",editors:[{id:"25840",title:"Prof.",name:"Ayse Emel",middleName:null,surname:"Onal",slug:"ayse-emel-onal",fullName:"Ayse Emel Onal"}],productType:{id:"1",title:"Edited Volume",chapterContentType:"chapter",authoredCaption:"Edited by"}},authors:[{id:"342716",title:"Assistant Prof.",name:"Aileen",middleName:null,surname:"Y. Chang",fullName:"Aileen Y. Chang",slug:"aileen-y.-chang",email:"chang@email.gwu.edu",position:null,institution:null},{id:"342718",title:"Dr.",name:"Evelyn",middleName:null,surname:"Mendoza-Torres",fullName:"Evelyn Mendoza-Torres",slug:"evelyn-mendoza-torres",email:"evelyn.mendozat@unilibre.edu.co",position:null,institution:null},{id:"427633",title:"Dr.",name:"Franklin",middleName:null,surname:"Torres",fullName:"Franklin Torres",slug:"franklin-torres",email:"dummy+427633@intechopen.com",position:null,institution:null},{id:"427634",title:"Dr.",name:"Wendy",middleName:null,surname:"Rosales-Rada",fullName:"Wendy Rosales-Rada",slug:"wendy-rosales-rada",email:"dummy+427634@intechopen.com",position:null,institution:null},{id:"427635",title:"Dr.",name:"Liliana",middleName:null,surname:"Encinales",fullName:"Liliana Encinales",slug:"liliana-encinales",email:"dummy+427635@intechopen.com",position:null,institution:null},{id:"427636",title:"Dr.",name:"Lil",middleName:null,surname:"Avendaño",fullName:"Lil Avendaño",slug:"lil-avendano",email:"dummy+427636@intechopen.com",position:null,institution:null},{id:"427637",title:"Dr.",name:"María Fernanda",middleName:null,surname:"Pérez",fullName:"María Fernanda Pérez",slug:"maria-fernanda-perez",email:"dummy+427637@intechopen.com",position:null,institution:null},{id:"427638",title:"Dr.",name:"Ivana",middleName:null,surname:"Terán",fullName:"Ivana Terán",slug:"ivana-teran",email:"dummy+427638@intechopen.com",position:null,institution:null},{id:"427639",title:"Dr.",name:"David",middleName:null,surname:"Vergara",fullName:"David Vergara",slug:"david-vergara",email:"dummy+427639@intechopen.com",position:null,institution:null},{id:"427640",title:"Dr.",name:"Estefanie",middleName:null,surname:"Osorio-Llanes",fullName:"Estefanie Osorio-Llanes",slug:"estefanie-osorio-llanes",email:"dummy+427640@intechopen.com",position:null,institution:null},{id:"427641",title:"Dr.",name:"Paige",middleName:null,surname:"Fierbaugh",fullName:"Paige Fierbaugh",slug:"paige-fierbaugh",email:"dummy+427641@intechopen.com",position:null,institution:null},{id:"427642",title:"Dr.",name:"Wendy",middleName:null,surname:"Villamizar",fullName:"Wendy Villamizar",slug:"wendy-villamizar",email:"dummy+427642@intechopen.com",position:null,institution:null},{id:"457495",title:"Dr.",name:"Jairo",middleName:null,surname:"Castellar-Lopez",fullName:"Jairo Castellar-Lopez",slug:"jairo-castellar-lopez",email:"dummy+427643@intechopen.com",position:null,institution:null}]}},chapter:{id:"77986",slug:"covid-19-transmission-in-children-implications-for-schools",signatures:"Evelyn Mendoza-Torres, Franklin Torres, Wendy Rosales-Rada, Liliana Encinales, Lil Avendaño, María Fernanda Pérez, Ivana Terán, David Vergara, Estefanie Osorio-Llanes, Paige Fierbaugh, Wendy Villamizar, Aileen Y. Chang and Jairo Castellar-Lopez",dateSubmitted:"June 15th 2021",dateReviewed:"July 12th 2021",datePrePublished:"September 13th 2021",datePublished:"March 16th 2022",book:{id:"10707",title:"Primary Health Care",subtitle:null,fullTitle:"Primary Health Care",slug:"primary-health-care",publishedDate:"March 16th 2022",bookSignature:"Ayşe Emel Önal",coverURL:"https://cdn.intechopen.com/books/images_new/10707.jpg",licenceType:"CC BY 3.0",editedByType:"Edited by",editors:[{id:"25840",title:"Prof.",name:"Ayse Emel",middleName:null,surname:"Onal",slug:"ayse-emel-onal",fullName:"Ayse Emel Onal"}],productType:{id:"1",title:"Edited Volume",chapterContentType:"chapter",authoredCaption:"Edited by"}},authors:[{id:"342716",title:"Assistant Prof.",name:"Aileen",middleName:null,surname:"Y. Chang",fullName:"Aileen Y. Chang",slug:"aileen-y.-chang",email:"chang@email.gwu.edu",position:null,institution:null},{id:"342718",title:"Dr.",name:"Evelyn",middleName:null,surname:"Mendoza-Torres",fullName:"Evelyn Mendoza-Torres",slug:"evelyn-mendoza-torres",email:"evelyn.mendozat@unilibre.edu.co",position:null,institution:null},{id:"427633",title:"Dr.",name:"Franklin",middleName:null,surname:"Torres",fullName:"Franklin Torres",slug:"franklin-torres",email:"dummy+427633@intechopen.com",position:null,institution:null},{id:"427634",title:"Dr.",name:"Wendy",middleName:null,surname:"Rosales-Rada",fullName:"Wendy Rosales-Rada",slug:"wendy-rosales-rada",email:"dummy+427634@intechopen.com",position:null,institution:null},{id:"427635",title:"Dr.",name:"Liliana",middleName:null,surname:"Encinales",fullName:"Liliana Encinales",slug:"liliana-encinales",email:"dummy+427635@intechopen.com",position:null,institution:null},{id:"427636",title:"Dr.",name:"Lil",middleName:null,surname:"Avendaño",fullName:"Lil Avendaño",slug:"lil-avendano",email:"dummy+427636@intechopen.com",position:null,institution:null},{id:"427637",title:"Dr.",name:"María Fernanda",middleName:null,surname:"Pérez",fullName:"María Fernanda Pérez",slug:"maria-fernanda-perez",email:"dummy+427637@intechopen.com",position:null,institution:null},{id:"427638",title:"Dr.",name:"Ivana",middleName:null,surname:"Terán",fullName:"Ivana Terán",slug:"ivana-teran",email:"dummy+427638@intechopen.com",position:null,institution:null},{id:"427639",title:"Dr.",name:"David",middleName:null,surname:"Vergara",fullName:"David Vergara",slug:"david-vergara",email:"dummy+427639@intechopen.com",position:null,institution:null},{id:"427640",title:"Dr.",name:"Estefanie",middleName:null,surname:"Osorio-Llanes",fullName:"Estefanie Osorio-Llanes",slug:"estefanie-osorio-llanes",email:"dummy+427640@intechopen.com",position:null,institution:null},{id:"427641",title:"Dr.",name:"Paige",middleName:null,surname:"Fierbaugh",fullName:"Paige Fierbaugh",slug:"paige-fierbaugh",email:"dummy+427641@intechopen.com",position:null,institution:null},{id:"427642",title:"Dr.",name:"Wendy",middleName:null,surname:"Villamizar",fullName:"Wendy Villamizar",slug:"wendy-villamizar",email:"dummy+427642@intechopen.com",position:null,institution:null},{id:"457495",title:"Dr.",name:"Jairo",middleName:null,surname:"Castellar-Lopez",fullName:"Jairo Castellar-Lopez",slug:"jairo-castellar-lopez",email:"dummy+427643@intechopen.com",position:null,institution:null}]},book:{id:"10707",title:"Primary Health Care",subtitle:null,fullTitle:"Primary Health Care",slug:"primary-health-care",publishedDate:"March 16th 2022",bookSignature:"Ayşe Emel Önal",coverURL:"https://cdn.intechopen.com/books/images_new/10707.jpg",licenceType:"CC BY 3.0",editedByType:"Edited by",editors:[{id:"25840",title:"Prof.",name:"Ayse Emel",middleName:null,surname:"Onal",slug:"ayse-emel-onal",fullName:"Ayse Emel Onal"}],productType:{id:"1",title:"Edited Volume",chapterContentType:"chapter",authoredCaption:"Edited by"}}},ofsBook:{item:{type:"book",id:"12100",leadTitle:null,title:"Nursing",subtitle:null,reviewType:"peer-reviewed",abstract:"This book will be a self-contained collection of scholarly papers targeting an audience of practicing researchers, academics, PhD students and other scientists. The contents of the book will be written by multiple authors and edited by experts in the field.",isbn:null,printIsbn:null,pdfIsbn:null,doi:null,price:0,priceEur:0,priceUsd:0,slug:null,numberOfPages:0,isOpenForSubmission:!0,isSalesforceBook:!1,hash:"e182e8721219e690a1be729935c41900",bookSignature:"",publishedDate:null,coverURL:"https://cdn.intechopen.com/books/images_new/12100.jpg",keywords:null,numberOfDownloads:null,numberOfWosCitations:0,numberOfCrossrefCitations:null,numberOfDimensionsCitations:null,numberOfTotalCitations:null,isAvailableForWebshopOrdering:!0,dateEndFirstStepPublish:"March 2nd 2022",dateEndSecondStepPublish:"March 23rd 2022",dateEndThirdStepPublish:"May 22nd 2022",dateEndFourthStepPublish:"August 10th 2022",dateEndFifthStepPublish:"October 9th 2022",remainingDaysToSecondStep:"2 months",secondStepPassed:!0,currentStepOfPublishingProcess:1,editedByType:null,kuFlag:!1,biosketch:null,coeditorOneBiosketch:null,coeditorTwoBiosketch:null,coeditorThreeBiosketch:null,coeditorFourBiosketch:null,coeditorFiveBiosketch:null,editors:null,coeditorOne:null,coeditorTwo:null,coeditorThree:null,coeditorFour:null,coeditorFive:null,topics:[{id:"16",title:"Medicine",slug:"medicine"}],chapters:null,productType:{id:"1",title:"Edited Volume",chapterContentType:"chapter",authoredCaption:"Edited by"},personalPublishingAssistant:null},relatedBooks:[{type:"book",id:"6550",title:"Cohort Studies in Health Sciences",subtitle:null,isOpenForSubmission:!1,hash:"01df5aba4fff1a84b37a2fdafa809660",slug:"cohort-studies-in-health-sciences",bookSignature:"R. Mauricio Barría",coverURL:"https://cdn.intechopen.com/books/images_new/6550.jpg",editedByType:"Edited by",editors:[{id:"88861",title:"Dr.",name:"R. Mauricio",surname:"Barría",slug:"r.-mauricio-barria",fullName:"R. Mauricio Barría"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"9500",title:"Recent Advances in Bone Tumours and Osteoarthritis",subtitle:null,isOpenForSubmission:!1,hash:"ea4ec0d6ee01b88e264178886e3210ed",slug:"recent-advances-in-bone-tumours-and-osteoarthritis",bookSignature:"Hiran Amarasekera",coverURL:"https://cdn.intechopen.com/books/images_new/9500.jpg",editedByType:"Edited by",editors:[{id:"67634",title:"Dr.",name:"Hiran",surname:"Amarasekera",slug:"hiran-amarasekera",fullName:"Hiran Amarasekera"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"1591",title:"Infrared Spectroscopy",subtitle:"Materials Science, Engineering and Technology",isOpenForSubmission:!1,hash:"99b4b7b71a8caeb693ed762b40b017f4",slug:"infrared-spectroscopy-materials-science-engineering-and-technology",bookSignature:"Theophile Theophanides",coverURL:"https://cdn.intechopen.com/books/images_new/1591.jpg",editedByType:"Edited by",editors:[{id:"37194",title:"Dr.",name:"Theophile",surname:"Theophanides",slug:"theophile-theophanides",fullName:"Theophile Theophanides"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"3161",title:"Frontiers in Guided Wave Optics and Optoelectronics",subtitle:null,isOpenForSubmission:!1,hash:"deb44e9c99f82bbce1083abea743146c",slug:"frontiers-in-guided-wave-optics-and-optoelectronics",bookSignature:"Bishnu Pal",coverURL:"https://cdn.intechopen.com/books/images_new/3161.jpg",editedByType:"Edited by",editors:[{id:"4782",title:"Prof.",name:"Bishnu",surname:"Pal",slug:"bishnu-pal",fullName:"Bishnu Pal"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"3092",title:"Anopheles mosquitoes",subtitle:"New insights into malaria vectors",isOpenForSubmission:!1,hash:"c9e622485316d5e296288bf24d2b0d64",slug:"anopheles-mosquitoes-new-insights-into-malaria-vectors",bookSignature:"Sylvie Manguin",coverURL:"https://cdn.intechopen.com/books/images_new/3092.jpg",editedByType:"Edited by",editors:[{id:"50017",title:"Prof.",name:"Sylvie",surname:"Manguin",slug:"sylvie-manguin",fullName:"Sylvie Manguin"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"371",title:"Abiotic Stress in Plants",subtitle:"Mechanisms and Adaptations",isOpenForSubmission:!1,hash:"588466f487e307619849d72389178a74",slug:"abiotic-stress-in-plants-mechanisms-and-adaptations",bookSignature:"Arun Shanker and B. Venkateswarlu",coverURL:"https://cdn.intechopen.com/books/images_new/371.jpg",editedByType:"Edited by",editors:[{id:"58592",title:"Dr.",name:"Arun",surname:"Shanker",slug:"arun-shanker",fullName:"Arun Shanker"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"72",title:"Ionic Liquids",subtitle:"Theory, Properties, New Approaches",isOpenForSubmission:!1,hash:"d94ffa3cfa10505e3b1d676d46fcd3f5",slug:"ionic-liquids-theory-properties-new-approaches",bookSignature:"Alexander Kokorin",coverURL:"https://cdn.intechopen.com/books/images_new/72.jpg",editedByType:"Edited by",editors:[{id:"19816",title:"Prof.",name:"Alexander",surname:"Kokorin",slug:"alexander-kokorin",fullName:"Alexander Kokorin"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"314",title:"Regenerative Medicine and Tissue Engineering",subtitle:"Cells and Biomaterials",isOpenForSubmission:!1,hash:"bb67e80e480c86bb8315458012d65686",slug:"regenerative-medicine-and-tissue-engineering-cells-and-biomaterials",bookSignature:"Daniel Eberli",coverURL:"https://cdn.intechopen.com/books/images_new/314.jpg",editedByType:"Edited by",editors:[{id:"6495",title:"Dr.",name:"Daniel",surname:"Eberli",slug:"daniel-eberli",fullName:"Daniel Eberli"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"57",title:"Physics and Applications of Graphene",subtitle:"Experiments",isOpenForSubmission:!1,hash:"0e6622a71cf4f02f45bfdd5691e1189a",slug:"physics-and-applications-of-graphene-experiments",bookSignature:"Sergey Mikhailov",coverURL:"https://cdn.intechopen.com/books/images_new/57.jpg",editedByType:"Edited by",editors:[{id:"16042",title:"Dr.",name:"Sergey",surname:"Mikhailov",slug:"sergey-mikhailov",fullName:"Sergey Mikhailov"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}},{type:"book",id:"1373",title:"Ionic Liquids",subtitle:"Applications and Perspectives",isOpenForSubmission:!1,hash:"5e9ae5ae9167cde4b344e499a792c41c",slug:"ionic-liquids-applications-and-perspectives",bookSignature:"Alexander Kokorin",coverURL:"https://cdn.intechopen.com/books/images_new/1373.jpg",editedByType:"Edited by",editors:[{id:"19816",title:"Prof.",name:"Alexander",surname:"Kokorin",slug:"alexander-kokorin",fullName:"Alexander Kokorin"}],productType:{id:"1",chapterContentType:"chapter",authoredCaption:"Edited by"}}]},chapter:{item:{type:"chapter",id:"79067",title:"Wrist Arthroscopy",doi:"10.5772/intechopen.99191",slug:"wrist-arthroscopy",body:'
Since wrist arthroscopies for diagnostical purposes were first reported and described in 1979, it has become an important diagnostic and therapeutic tool in the hands of trained specialists during the last decades. Nowadays it is widely used in the treatment of acute wrist injuries as well as different chronic conditions and degenerative diseases. Arthroscopy has assumed an important place in wrist surgery. The wide list of indications for wrist arthroscopy is continuously growing and requires specific operative skills and specialized training before entering the operating room for real surgery. At this point it’s necessary to highlight the huge investment of International Wrist Arthroscopy Society (IWAS) and Asia Pacific Wrist Association (APWA) in training programs, courses and workshops all around the world. Today arthroscopic methods are proposed in the treatment of almost all soft tissue and osseous problems of the wrist. They include synovitis, fibrosis, stiffness, triangular fibrocartilage complex (TFCC) problems, ganglion cysts, scapholunate- (SLIL) and lunotriquetral ligament (LTIL) tears, intra-articular distal radius fracture (DRF) and nonunion treatment, arthroscopic arthrolysis, treatment of scaphoid fractures and nonunions, arthroscopic treatment of Kienböck’s disease, arthroscopically assisted partial wrist fusions, etc. We can be grateful to the industry, that they have been able to listen to our wishes and create the tools needed for our surgeries. Since authors have been limited in the size of the manuscript as well as many well illustrated books on this topic have already been published, we will address only the most frequent problems of wrist arthroscopy in our practice.
Early arthroscopic explorations mostly focused on a knee joint, but M. Burman in the early thirties experimented with a use of arthroscopy in other joints, including the wrist joint. In 1970, the 1.7 mm No. 24 arthroscope was developed, allowing a wide angle of vision and clear focus utilizing a small diameter. M. Watanabe, who started to use arthroscopy in 1950s and developed the famous No. 21, used this scope to examine 21 wrists in 1970–1972. He developed dorsal approaches on the ulnar side of the extensor pollicis longus tendon to access the radiocarpal joint. Watanabe reported on 67 wrist arthroscopies, including visualization of the distal radioulnar joint. He also included 39 arthroscopies of the thumb carpometacarpal joint and metacarpophalangeal joints as well as 9 interphalangeal joint arthroscopies [1]. Y.C. Chen was an another enthusiast and pioneer of the wrist arthroscopy in 1970s. In 1979 he published an article about 90 arthroscopies of the wrist and finger joints in 34 clinical cases and 2 amputed arms with No.24 arthroscope. Eighty-four wrist and finger joints of four cadavers and two amputated arms were also dissected for macroscopic observation. This article was also illustrated with some color photographies taken with arthroscope [2].
In 1986 Terry Whipple published paper on wrist artrhroscopy technics and described the safe wrist artrhroscopy portals which are still used today [3]. The first wrist arthroscopy course was organized by Terry Whipple, Gary Poehling and James Roth in Winston-Salem, North Carolina, USA in 1986. The first textbook dedicated to the wrist arthroscopy was published in 1992 by T. Whipple. During the next decades the popularity of the wrist arthroscopy has grown, new indications and techniques have been developed. In 1997 P.C.Ho organized the 1st Hong Kong Wrist Arthroscopy course. With growing international interest in wrist arthroscopy, Christophe Mathoulin founded the European Wrist Arthroscopy Society (EWAS) in 2005 and the first EWAS cadaveric wrist arthroscopy course was organized in Strasbourg. In 2015 P.C. Ho and G. Bein developed the Asian Pacific Wrist Association (APWA). Both of them are wrist-specific international educational organizations with worldwide network of wrist arthroscopy courses and workshops (Figure 1). In 2019, EWAS evolved into the IWAS – International Wrist Arthroscopy Society.
Authors attending their first wrist arthroscopy course in Rotterdam in 2009 – Uldis Krustins (left), Janis Krustins (right).
Wrist arthroscopy requires standard arhroscopic equipment – arthroscopy column with monitor, video camera, video and photo recordingdevice, light source with fiber optic cable which nowadays can be integrated in one small box motorized shavers, radiofrequency ablators and X-Ray C arm and traction system. The patient is positioned supine on the operation table with the affected arm on a hand table. The arm is abducted 90° and the elbow flexed 90° allowing a vertical position of the forearm, wrist and hand. In this position the wrist is kept in neutral prono-supination. The surgeon is positioned at the head of the patient with the assistant beside or facing the surgeon. The arthroscopy column may be on the other side of the patient facing the surgeon (Figure 1). X-ray C arm is used when necessary and is not in the way of the staff all the time (Figures 2 and 3).
Standard setup and position of the staff during the wrist arthroscopy.
Use of C arm during the surgery.
Arthroscopic wrist procedures usually are performed under the regional block anesthesia with a pneumatic tourniquet placed on the upper arm, but there are surgeons who propose to do wrist arthroscopies under portal site local anesthesia (PSLA) without tourniquet [4]. Traditionally wrist arthroscopies were performed using irrigation, but it can be easily inspected and treated also in “dry technique” [5]. However, dry arthroscopy also has its limits. For example when radiofrequency ablators are used, water is necessary as milieu conductor and to prevent temperature peaks and possible joint damage. Also when using a burr the aspiration may be blocked by small cartilage and bone fragments and water facilitates the aspiration [6].
Arthroscopic manipulations in wrist require vertical traction to separate bones and create space for instruments and scope. The traction applied is usually 3.5 to 7 kg for wrist joint and 2 to 3 kg for the thumb [6, 7]. There are different types of wrist vertical traction towers – with Chinese finger traps or special traction hands for all fingers. Authors use K.L.Martin and Smith & Nephew wrist traction towers (Figure 4a, b).
Different finger traps and traction devices (4a - K.L.Martin; 4b -Smith & Nephew).
The most important instrument, of course, is the scope. In wrist small arthroscopes, between 1.9 mm and 2.7 mm, with camera angulated at 30°, are usually used Arthroscopes are shorter (60 to 80 mm), too. The second most important tool is a probe which helps to explore the joint and serves as an extension of the surgeon’s finger [8]. A variety of differently angled punches, baskets with or without the option of incorporating a suction mechanism and grasping forceps in various sizes are useful for specific manipulations (Figure 5). A motor is fitted with abrasive instruments, such as shavers or burrs of appropriate sizes: 2 to 3 mm in diameter and 6 to 8 cm long. Basic instruments also include a shaver for synovial resection and a burr for bony resection. A special electric bipolar diathermy machine is used for efficient tissue resection by vaporization. An irrigation system is used for joint cleaning. The equipment can be completed by different instruments or kits for specific arthroscopic procedures.
Set of manual instruments for wrist arthroscopy.
The map of safe wrist arthroscopy portals was first designed by Terry L. Whipple and co-authors in 1986 after anatomical studies of fresh cadaveric wrists which were arthroscoped an then tediously dissected to determine the relationship of each portal to the closest neurovascular and tendinous structures [3]. Seven dorsal wrist portals were identified - five portals for radio-carpal joint with relation to the six extensor compartments (1-2; 3-4; 4-5; 6R and 6 U), one for midcarpal joint – distally from the 3-4 portal and the seventh portal for DRUJ. Anatomical studies proved that 1-2,6 U and 6R portals are the most perilous due the proximity of the radial artery and dorsal radial and ulnar sensory nerve branches. The midcarpal, 3-4,4-5 and DRUJ portals are relatively safe, since neurovascular structures are usually remote [9]. Later additional portals for midcarpal and radio-carpal joint, DRUJ as well as portals and techniques for small joint arthroscopy were described [9, 10, 11, 12, 13, 14].
Localization of portals first has to be checked by palpation with fingertip, then standard intramuscular injection needle can be inserted to determine the exact orientation of the portal. Small and shallow horizontal incisions using no. 15 blade are recommended. Then skin, subcutaneous tissues and join capsule can be dissected using mosquito clip to push away any important structures without injuring them. It’s suggested to use a curved mosquito clip which can easily slip over the curve of the dorsal rim of the radius or proximal carpal bones.
The normal inclination of the dorsal radius and lunate must be taken into consideration when entering the joints with trocar and never use sharp trocars. The insertion angle usually is about 10° proximally to parallel of the dorsal joint axis, to match the distal articular curves of the bones (Figure 6).
Insertion angle of the instruments.
Volar portals can be used for visualization of the dorsal capsular structures like dorsal radiocarpal ligament (DRCL), palmar aspects of the carpal ligaments or as occasional accessory portals in arthroscopic assisted surgeries of the distal radius fractures or carpal injuries [13, 14, 15].
Localization, function of radiocarpal portals and structures at risk are presented in Table 1 and for midcarpal portals in Table 2.
Portal | Localization and function | Structures at risk |
---|---|---|
1-2 | Between APL and ECRB tendons at the dorsal aspect of the anatomical snuffbox. Used rarely, mostly to reach the radial styloid process and radial part of the joint, as well as for removal of the volar ganglion cysts or washout of the joint. | Radial artery, superficial sensory branch of the radial nerve |
3-4 | 1 cm distally from the Lister`s tubercle, between the EPL and EDC tendons. Always used as a primary portal for visualization of the joint. Almost a complete panoramic view of the volar radiocarpal joint. | EPL and EDC tendons |
4-5 | Axis of the 4th metacarpal, between the EDC and EDM tendons. Portal for instrumentation and visualization of the TFCC. | EDC and EDM tendons |
6R | At the radial aspect of the ECU tendon. Additional portal for instrumentation or visualization or reconstruction of the TFCC | Dorsal branch of the ulnar nerve |
6U | At the ulnar aspect of the ECU tendon. Additional portal for instrumentation, visualization of the reposition of DRF fragments or for reconstruction of the ulnar part of the TFCC | Dorsal branch of the ulnar nerve |
VR | 2 cm long incision parallel to FCR tendon. Used for evaluation of the DRCL and volar part of the SLIL or fixation of the DRF fragments | Palmar cutaneous branch of the radial nerve, radial artery and volar cutaneous branch of the median nerve |
VU | 2 cm long incision parallel to ulnar margin of the flexor tendons. Used for the evaluation of the volar part of the LTIL and dorsal ulnar capsule. | Ulnar artery, ulnar nerve and distal palmar branch of the ulnar nerve |
Radiocarpal portals.
APL –
Portals | Localization and function | Structures at risk |
---|---|---|
MCR | Localized 1 cm distally from the 3-4 portal, between ECRB and EDC tendons. Visualization of the scapho-lunate, scapho-capitate and scapho-trapezium-trapezoideum joints. | ECRB and EDC tendons |
MCU | Localized 1 cm distally from the 4-5 portal, on the axis of the 4th metacarpal bone, between EDC and EDM tendons. Visualization of the luno-triquetral, luno-capitate and. triquetro-hamate joints | EDC and EDM tendons |
STT | Localized on the axis of the 2nd metacarpal bone, ulnary from the EPL tendon, at the level of the scapho-trapezio-trapezoidal joint. Used for instrumentation and resection of the distal scaphoid in STT arthritis | Radial artery, EPL, ECRB and ECRL tendons, terminal branches of the sensory branch of the radial nerve |
STT-R | Localized radially to APL tendon at the same level as STT portal. Used for resection of the distal pole of the scaphoid | APL tendon, terminal branches of the sensory branch of the radial nerve |
Midcarpal portals.
APL –
The images below (Figures 7–18) illustrate the anatomical structures of the wrist that can be identified from different standard portals.
Standard dorsal portals of the wrist. MCR – midcarpal radial portal, MCU – midcarpal ulnar portal, STT - scapho-trapezio-trapezoidal portal, DDRU – dorsal distal radioulnar joint portal, PDRU – proximal distal radioulnar joint portal, ECU –
Standard volar radiocarpal portals. RVP – radial volar portal, UVP – ulnar volar portal.
Straight/radial view from 3 to 4 portal. SC – scaphoid, RSC – radioscaphocapitate ligament, LRL – long radiolunate ligament.
Straight view from 3 to 4 portal. SC – scaphoid, RLT – radiolunotriquetral ligament, Lu – lunate, LRL – long radiolunat ligament, SLIL – scapholunate interosseous ligament.
Straight/ulnar view from 3 to 4 portal - Lu – lunate, SC – scaphoid, UL – ulnolunate ligament, UT – ulnotriquetral ligament, SLIL - scapholunate interosseous ligament.
Ulnar view from 3 to 4 portal over the scaphoid ridge, radial part of TFCC accessible.
Ulnar view from 3 to 4 portal – TFCC, proximal part of Triquetrum and ulnar recess. 6 U – possible location of 6 U portal.
Degenerative central tear of TFCC in “ulna +” variation.
Rupture of the dorsal SLIL. View from 6R portal. SLIL - scapholunate interosseous ligament.
View of STT joint from MCR portal. STT – Scaphotrapeziotrapezoidal joint, MCR – midcarpal radial portal.
View of scapholunate joint from MCR portal.
View of lunotriquetral joint from MCR portal.
There are four described arthroscopy portals in the distal radio-ulnar joint. The anatomy of the DRUJ is complex because ulna articulates with both radius and proximal carpal row. Stability of the DRUJ is provided by TFCC with its volar and dorsal distal radioulnar ligaments, connecting at the fovea of the ulnar head. Even in normal wrists DRUJ is a quite narrow place for visualization and instrumentation, therefore it’s suggested to use 1.9 mm scope, reduce the traction of the arm and introduce the scope in the DRUJ when the wrist is fully supinated [16, 17, 18]. Localization of the DRUJ portals, their functions and structures of the risk are described in Table 3.
Portals | Localization and function | Structures at risk |
---|---|---|
D-DRUJ | Localized 5-8 mm proximally from the 6R portal, between EDM and ECU tendons. Visualization of the ulnar head, undersurface of the TFCC, sigmoid notch, insertion of the volar and dorsal radioulnar ligaments in ulnar fovea | TFCC, articular cartilages of the sigmoid notch and ulnar head |
P-DRUJ | Localized 1 cm proximal to the D-DRUJ portal, between EDM and ECU tendons. Visualization of the sigmoid notch, head of ulna and volar capsule, | TFCC, articular cartilages of the sigmoid notch and ulnar head |
V-DRUJ | 2 cm long incision parallel to ulnar margin of the flexor tendons – the same as for VU portal or can be created by pushing the blunt trocar from 6U portal towards the anterior ulnocarpal capsule between UL and UT ligaments, exiting ulnar to the flexor tendons, where small skin incision can be made. Visualization of the dorsal radioulnar ligament, possibility to proceed an arthroscopic wafer procedure when TFCC is intact | Ulnar artery, ulnar nerve and distal palmar branch of the ulnar nerve, flexor tendons |
DF | Localized 1 cm proximally to the 6U portal between ECU tendon and ulnar styloid dorsally and FCU tendon volarly. Forearm must be fully supinated. Visualization of the ulnar fovea region, this portal is used as a dedicated working portal for fixation of the TFCC to the ulnar fovea in proximal TFCC lesions | Dorsal branch of the ulnar nerve, ECU and FCU tendons |
DRUJ portals.
ECU –
There mostly are two standard portals for STT, first carpometacarpal joint (CMC), metacarpophalangeal (MCP), proximal interphalangeal joint (PIP) and distal interphalangeal joint (DIP). Arthroscopical access to pisotriquetral (PT) [19, 20] and fourth or fifth CMC joints also are described while usefulness of these procedures is limited or not yet established [10].
First CMC portals are localized approximately 1 cm distally from STT portals on both sides of the first dorsal compartment. Accessory dorsal portal (the dorsal ulnar portal) can be used as necessary for better viewing the radial side of the joint by placing a trocar into the volar portal, across the CMC or the STT joint and out the dorsum of the hand (Figure 19) [21].
CMC and STT portals. a – localization on the skin, b – verification with fluoroscope.
There are two portals – radial and ulnar for MCP, PIP and DIP joint arthroscopies, the naming of them is related to relationship with extensor tendons and they were established by Chen [2]. MCP joint arthroscopies can be successfully used in the rheumatoid conditions when synovectomy and thermal shrinkage can be performed [22, 23, 24], or in traumatic cases such as gamekeepers injury [10, 25], collateral ligament ruptures and reduction of the intraarticular metacarpal head fractures as well as in cases of complex MCP joint dislocations [26, 27]. Indications of the MCP joint arthroscopy include also chronic cases of instability, removal of the loose bodies as well as in cases of joint stiffness caused by intraarticular fibrosis or even cases of septic arthritis [18, 28].
Arthroscopy of the PIP and DIP joints has not been widely accepted because of the limited indications and technical limitations. The main indications are inflammatory or septic arthritis as well as removal of foreign bodies. Several authors suggest horizontal placement of the hand instead of using a traction tower, as it is important to be able to flex the joint freely [29, 30]. Cobb reported several cases of the DIP arthroscopic arthrodesis [10]. Since authors have no personal experience in finger joint arthroscopy, further discussion on this topic will not be continued in this chapter.
Many years the use of intra-articular fluid for wrist arthroscopy was considered essential. Francisco del Piñal
Ganglion cysts are the most common benign soft-tissue tumors of the wrist. Dorsal cysts are more common than volar and surgical treatment is indicated for painful ganglions or large ones for cosmetic purposes. These ganglions usually appear in the dorsal scapholunate region which consists of three anatomical structures – the dorsal segment of scapholunate (SL) ligament, the dorsal intercarpal ligament (DICL) and the dorsal capsuloscapholunate septum (DCSS) [31]. The extra-articular part of the cyst can vary in size and location as well as in relation to dorsal ligaments. Surgical treatment of the so called “occult” ganglion cysts, who are small, intracapsular and can be very painful, is challenging by conventional methods. Arthroscopic treatment of such ganglion cysts is a method of a choice.
There are two different arthroscopical techniques for resection of the dorsal ganglia. The one is an access via radiocarpal joint and the other is through the ganglion and via the midcarpal joint [32, 33]. Some authors describe the necessity to combine radiocarpal and midcarpal portals, thus enabling a complete resection [34].
In the 2nd edition of Wrist Arthroscopy Techniques by C. Mathoulin different techniques of the dorsal ganglion artrhroscopic resections using only midcarpal portals are described and well-illustrated [35]. In our hands the midcarpal approach works perfect in most cases, except if ganglions are located very proximally (Figure 20). It provides also a good cosmetic result with only two almost invisible scars on the dorsal aspect of the wrist, which is important especially in younger females.
Dorsal ganglion, view from MCU portal, a – visualization of ganglion cyst after synovectomy, b – ganglion removed, clean extensor tendons visible.
Aftertreatment – patients have to be encouraged to start early movements. In some cases, if patients have low pain malaise, short arm cast or wrist splint is recommended for first week after surgery. Recurrence rate for dorsal wrist ganglions treated arthroscopically is from 3 to 12% [34, 36, 37, 38]. Complications are rare and they are less common than in open surgeries. Most common complications reported are some stiffness (less than with open surgery), neurapraxia, extensor tenosynovitis and complex regional pain syndrom [39]. In meta-analysis presented by Head et al. in 2015, mean complication rate for arthroscopic surgical excision was 4%, and recurrence rate 6% [40]. Complication rate reported for open surgeries was 14% and recurrence rate 21%.
Volar wrist ganglions are less common than dorsal ganglions (about 20%) and they mainly occur in the radiocarpal joint, especially in the radial corner of the volar aspect. Volar ganglions in the midcarpal joint are very rare ant mostly they occur as a result of STT arthritis. The most common appearance is below FCR and FPL tendons. The technique of the arthroscopic volar ganglion resection was first described by P.C.Ho et al. in 2003 [41]. The origin and stalk of the ganglion usually locates between radioscaphocapitate (RSC) and long radiolunate (LRL) ligaments. It becomes visible by gently pushing with finger on the ganglion while scope is positioned in 3-4 portal. Shaver is inserted in the 1-2 portal and ganglion has to be removed gently to avoid injuries of the neurovascular structures and flexor tendons (Figure 21) [39, 42].
Volar ganglion of the wrist, a – ganglion detected in volar radial corner of the wrist, b – ganglion removed.
Aftertreatment is similar to that one for dorsal ganglions. In a systemic review presented by Fernandes et al. in 2014 mean complication rate for arthroscopic volar ganglion surgeries was 6% and recurrence rate 6.9% [43]. Reported complications are increased cyst site volume during the immediate postoperative period, radial artery injuries, neuropraxia of the dorsal radial nerve, partial lesions of the median nerve [39, 43, 44].
Intraosseus ganglions (IOG) can affect all carpal bones but mostly they affect the lunate, capitate and scaphoid [45]. In patients who have dorsal wrist ganglions, the prevalence of IOGs is reported to be almost 50% [46]. Most of them are asymptomatic and can be found during the routine radiographs or CT scans because of the different complains. Surgical treatment is recommended for the symptomatic IOGs and include the curettage of the damaged bone and bone grafting.
Arthroscopically assisted treatment of the intraosseus ganglions of the lunate was first described by Ashwood and Bain in 2003 with the aim of reducing the morbidity that has been seen with open techniques [47].
Surgeries can be performed via routine radiocarpal or midcarpal portals – depending of the localization of the ganglion cyst. Usually the ganglion cyst cannot be visualized by arthroscope, because they still remain covered by the articular cartilage. The location of the drill hole has to be determined by the preoperative radiographic investigations. Once the ganglion is removed with the arthroscopic cutter and the hole is debrided with curette and shaver, it can be filled with bone grafts from the distal radius or iliac crest, which can be harvested via small incision and then delivered into the bone through a trocar under the arthroscopic visualization (Figure 22).
Arthroscopic debridement of the scaphoid cyst, a – defect of the bone after debridement, b – defect closed with autologous bone graft (ABG).
Aftetreatment includes immobilization for 10 to 14 days and patients are advised not to return to light duties until 6 weeks after the surgery, and heavy manual labor is avoided for a minimum of three months [18].
Scapholunate interosseus ligament (SLIL) should be considered as a key stone of the intercarpal stability. It is U shaped in the sagittal plane and has three components – dorsal, volar and proximal [48]. The dorsal segment is the strongest portion with a tensile strength of 260 - 300 N and approximate thickness of 3 mm [32]. The proximal component is the weakest and avascular, The volar part has a tensile strength of 120- 150 N and approximate thickness of 1 mm. The palmar and dorsal segments work collectively to prevent rotational translation between scaphoid and lunate, whereas the intermediate segment has little role in stability [49, 50, 51].
Scapholunate stability is effectively ensured by a complex associating the dorsal and volar portions of the SLIL, the dorsal intercarpal (DIC) ligament, the dorsal radiocarpal (DRC) ligament, the radioscaphocapitate (RSC) ligament, the scaphotrapezial (ST) ligament, and the dorsal capsulo-scapholunate septum (DCSS). The integrity of these various stabilizers is taken into account while determining the arthroscopic classification of “predynamic” scapholunate instability [52].
When the SLIL is injured, the scaphoid tends to move into volarflexion, while the lunate, which is still fixed to the triquetrum, is forced, due to carpal kinematics, to follow the triquetrum into dorsal extension. The opposite happens with time when the LT interosseous ligament (LTIL) is injured. This static instability is often referred to radiologically as dorsal intercalated segment instability (DISI), following an SLIL injury and volar intercalated segment instability (VISI) following a LTIL injury [53].
The first arthroscopic classification of SLIL tears was presented in 1996 by Geissler et al. [54] using a 4-stage grading system (Table 4) (Figures 23 and 24).
Grade | Description |
---|---|
I | Attenuation or hemorrhage of interosseous ligament as seen from radiocarpal space. No incongruency of carpal alignment in mid-carpal space. |
II | Attenuation or hemorrhage of interosseous ligament as seen from radiocarpal space. Incongruency or step-off of carpal space. There may be slight gap (less than width of probe) between carpal bones. |
III | Incongruency or step-off of carpal alignment as seen from both radiocarpal and mid-carpal space. Probe may be passed through gap between carpal bones. |
IV | Incongruency or step-off of carpal alignment as seen from both radiocarpal and mid-carpal space. There is gross instability with manipulation. 2.7-millimeter arthroscope may be passed through gap between carpal bones (“drive thru sign”). |
Geissler’s arthroscopic classification of SLIL tears.
Geissler grade II tear – attenuation an hemorrhage (*) of SLIL.
Geissler grade IV tear – arrow shows gap between scaphoid and lunate. Scope easily passes between bones and slides into midcarpal joint.
In 2013 Messina et al. published the European Wrist Arthroscopy Society (EWAS) Classification of Scapholunate tears which was based on anatomical arthroscopic study and is an evolution of Geissler’s classification (Table 5) [55].
Arthroscopic stage (EWAS) | Arthroscopic testing of SLIOL from MC joint | AP findings |
---|---|---|
I | No passage of the probe | Not found in cadaver specimens |
II lesion of membranous SLIL | Passage of the tip of the probe in the SL space without widening (stable) | Lesion of proximal/membranous part of SLIL |
III A partial lesion involving the volar SLIL | Volar widening on dynamic testing from MC joint (anterior laxity) | Lesion of anterior and proximal part of SLIL with or without lesion of RSC- LRL |
III B partial lesion involving the dorsal SLIL | Dorsal SL widening on dynamic testing (posterior laxity) | Lesion of proximal and posterior part of SLIL with partial lesion of DIC |
III C complete SLIL tear, joint is reducible | Complete widening of SL space on dynamic testing, reducible with removal of probe | Complete lesion of SLIL (anterior, proximal, posterior), complete lesion of one extrinsic ligament (DIC lesion or RSC/ LRL) |
IV complete SLILwith SL gap | SL gap with passage of the arthroscope from MC to RC joint No radiographic abnormalities | Complete lesion of SLIL (anterior, proximal, posterior), lesion of extrinsic ligaments ( DIC, and RSC/ LRL) |
V | Wide SL gap with passage of the arthroscope through SL joint Frequent X Ray abnormalities such as an increased SL gap, DISI deformity | Complete lesion of SLIL, DIC, LRL, RSC, involvement of one or more other ligaments (TH, ST, DRC) |
Arthroscopic EWAS (European Wrist Arthroscopy Society) Classification and corresponding anatomo-pathological (AP) findings in cadaver specimens.
SLIL: scapholunate interosseous ligament. MC: midcarpal. RC: radiocarpal. RSC: radio-scapho-capitate. LRL: long radiolunate. DIC: dorsal intercarpal ligament. SL: scapholunate. TH: triquetro-hamate. ST: scaphotrapezial. DRC: dorso radiocarpal. DISI: dorsal intercalated segmental instability.
The existing classifications, however, describe the dynamic instability of the scapholunate joint but do not distinguish the site of ligament attenuation or tear, particularly in its volar portion.
The choice of the procedure for SLIL injuries in the absence of arthritis depends on the extent of the lesion, quality of the ligament remnants and reducibility of the joint [53].
Garcia-Elias et al. [56] developed a set of 6 questions that provide a useful framework for developing stage-based treatment algorithms. By answering these questions in terms of yes or no, each case can be placed into one of seven categories (Table 6). As expected, the increasing number of negative answers indicates a progression of the dysfunction from minimal (Stage 1) to maximal (Stage 7). In general, all instabilities from the same stage will be treated similarly.
Stage | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
---|---|---|---|---|---|---|---|
Is the dorsal SL ligament intact? | NO | NO | NO | NO | NO | NO | |
If repaired, has it good chances of healing? | NO | NO | NO | NO | NO | ||
Is the radioscaphoid angle normal? | NO | NO | NO | NO | |||
Is the lunate uncovering index normal? | NO | NO | NO | ||||
Is the misalignment easily reducible? | NO | NO | |||||
Are the joint cartilages normal all over the wrist? | NO |
Six questions by Marc Garsia-Elias.
Detailed description of indications and treatment methods depending on the time after injury, the stage of SLIL disruption and stability or instability of the carpus is presented in the report of the IFSSH Committee On Carpal Instability in 2016 (part 2: Management of scapho-lunate dissociation [57].
Since this book is oriented to the arthroscopic methods of treatment further discussion on open surgical procedures will not be proceeded.
In acute injuries, arthroscopy can be used to determine the extent of scapholunate interosseous ligament injury. Partial tears may be treated by percutaneous pinning of the scaphoid and lunate, thus allowing for the possibility of primary healing or fibrosis.
Predynamic or occult SL injury results from an incomplete tear of the SL ligament. In selected cases with reducible scapholunate instability (Garcia-Elias stages 2, 3 and 4) where the ligament is partially (Figure 25) or completely ruptured, and where the scaphoid is well aligned or can be reduced, Mathoulin et al. proposed the arthroscopic dorsal capsuloplasty, which may be combined with K-wire fixation of the scapholunate and the scaphocapitate joints [58, 59].
Partial, reparable rupture of SLIL, a- view from MCU portal, yellow arrow shows the step off between articular surfaces od scaphoid and lunate, red arrow points detachment of SLIL from lunate, b- The same patient – normal SLIL in RC joint, view from 3 to 4 portal.
This technique can be performed only in cases when ligament stumps remain attached to the scaphoid and lunate. This technique includes synovectomy of midcarpal and radiocarpal joints. Then the scope is introduced into the 6R portal to inspect the gap between the lunate and the dorsal capsule (Figure 26).
View from 6R portal with detached dorsal capsule and injured dorsal capsuloligamentous scapholunate septum (DCSS).
An absorbable monofilament suture is passed through a needle. This needle is inserted through the skin via the 3–4 portal, then shifted slightly distally so as to cross the joint capsule (Figure 27).
Needle inserted via 3-4 portal and shifted distally to enter the midcarpal joint.
The needle is located inside the joint through the scope and then pushed through the SLIL stump on the scaphoid side. The needle is oriented dorsal to volar and angled proximal to distal, allowing it to enter the midcarpal joint (Figure 28). A second needle and suture are then inserted parallel to the first into the SLIL stump attached to the lunate. The scope is returned to the MCU portal. The two needles are located inside the midcarpal joint, after they have passed between the scaphoid and lunate. Both sutures are externalized via MCR portal with hemostat and the knot is tied outside the joint. Then the knot is pulled back into the midcarpal joint (Figure 29).
View from the UMP - needle in the midcarpal joint and suture is knotted by twisting of the needle.
The inner knot of both sutures at the level of SL joint.
At this point the traction of the wrist is released to reduce the gap between scaphoid and lunate. Transfixation of the scapholunate and scaphocapitate joints with K-wires can be performed if reduction is insufficient. The final knot is tied after the wrist is released from traction and positioned in slight extension [52].
After treatment includes 8 weeks of immobilization and an adequate rehabilitation.
There are several other, more complicated arthroscopic SLIL repair procedures described, but the indications of these techniques are limited to predynamic and dynamic SL instability.
P.C.Ho et al. in 2002 designed an arthroscopic assisted box reconstruction of scapholunate ligament with palmaris longus (PL) tendon graft [60]. It enables simultaneous reconstruction of the dorsal and palmar SL ligaments anatomically with the use tendon graft in a boxlike structure. With the assistance of arthroscopy and intraoperative imaging as a guide, a combined limited dorsal and volar incision exposed the dorsal and palmar SL interval without violating the wrist joint capsule. Bone tunnels of 2.4 mm are made on the proximal scaphoid and lunate. A palmaris longus tendon graft is delivered through the wrist capsule and the bone tunnels (Figure 30) to reduce and connect the two bones in a boxlike fashion.
A palmaris longus tendon graft deliverd through the lunate and both capsules.
Once the joint diastasis is reduced and any DISI malrotation corrected, the tendon graft is knotted and sutured on the dorsal surface of the SL joint extra-capsularly in a shoe-lacing manner (Figure 31a
Tightened tendon graft, view from volar (a) and dorsla (b) side.
Corella et al. in 2011 published a novel all arthroscopic technique for scapholunate instability [61]. They developed the BTT ligamentoplasty - Bone (base of second metacarpal bone), Tendon (flexor carpi radialis graft), Tenodesis (in the scaphoid and lunate). This technique aims to reproduce the tripple tenodesis technique proposed by Garcia-Elias et al. in 2006 [56], but with an arthroscopic approach reducing soft tissue trauma. It reconstructs both the dorsal and volar portion of the SL ligament with a 3-mm graft of the FCR tendon, which is fixed to the scaphoid and lunate tunnels with interference screws. Graft resistance and strength can be increased with the use of a 1.3 mm SutureTape. The SutureTape is passed and fixed with the screws along with the tendon graft. After the graft is fixed to the scaphoid bone with the anchor, the volar portion of the SutureTape that exits from the lunate tunnel is sutured to the portion that exits from the scaphoid tunnel. It’s recommended to start early postoperative wrist mobilization with this technique – dart-throwing motion from the 3rd week and flexion/extension movements from the 5th week after surgery [62].
Hand surgeons began applying wrist arthroscopy to the surgical treatment of the DRF in the late 90’s of the last century. Arthroscopic reduction of intraarticular fragments, as opposed to conventional methods, may improve outcomes regardless of the method of fixation, volar locking plates or external fixator and K-wires [63, 64, 65, 66, 67, 68, 69]. Failure to reduce intra-articular fractures of the distal radius predisposes to pain, restricted movement and degenerative arthritis. The functional results of treatment for articular DRF’s are determined by alignment of fragments of extraarticular fracture, by restoring bone shape, length and fold, anatomical reposition of joint surface, prevention of additional damage to soft tissue, as well as potential post-operative complications [70, 71, 72, 73, 74]. Fluoroscopy alone provides an image that has poorer resolution than that of the magnified camera used for direct arthroscopic visualization, whereas even a small degree of displacement is obvious arthroscopically [75]. It is obvious that optical visualization of the articular space gives an opportunity to detect a greater number of soft tissue lesions more often, than only fluoroscopic and clinical evaluation or surgeon’s mistrust about the possibilities of such injuries [76]. Wrist arthroscopy is currently recommended for the treatment of any articular distal radius fracture (Figure 32a
Dorsally displaced articular fracture of the distal radius.
There are two controversial fracture fixation techniques. In cases of volar plating, standard flexor carpi radialis approach can be used. Once the fracture is preliminarily fixed with the volar locking plate (VLP) (Figure 33), the wrist joint is assessed arthroscopically using the 3-4 and 4-5, 6R, 6 U or 1-2 portals to remove blood clots, small articular fragments or to make an additional reposition and fixation with K-wires. Distal screws can be inserted only after arthroscopic inspection of the radiocarpal joint and a fluoroscopic confirmation of the correct position for the screws (Figures 34a
Preliminary fixation of the volar plate before arthroscopic part of the surgery.
Additional fragment reposition and fixation.
Final fluoroscopic image of the surgery.
After treatment includes 2 weeks in short plaster cast and early mobilization can be allowed as volar locking plate provides rigid fixation.
In cases of comminuted fractures when fixation with VLP is impossible, arthroscopically assisted fracture reposition and fixation with K-wires and external fixator is recommended (Figures 36a–c and 37a
Position of the monolateral external fixator over the wrist joint. a and b - Comminuted volar, distal, articular fracture of the radius.
Final x-ray after the application of external fixation and K-wires.
The associated soft tissue lesions can be found from 30 to 66% of DRFs, but not all of them require surgical treatment [79, 80, 81]. In cases of associated soft tissue injuries like SLIL and LTIL acute ruptures or TFCC lesions, arthroscopically guided, debridement of the injured ligaments or TFCC is advised, as well as trans-articular fixation of the scapholunate and/or lunotriquetral joints with K-wires, or application of peripheral sutures for TFCC tears.
Authors have never experienced severe complications as tendon ruptures or infection but we have found that the more extensive use of K-wires in reduction and/or fixation during external fixation and K-wire fixation is more likely to result in nerve damage. Furthermore, the complication of subsequent loss of position of fragments also occurred in patients treated with K-wires and external fixator [81].
In last two decades several minimally invasive plate osteosynthesis (MIPO) techniques using volar locking plates on DRF are presented [82, 83, 84]. In cases of comminuted articular DRF’s this technique can be supplemented with an arthroscopic visualization. After all, two major lines of MIPO techniques evolved and got new promoters: single longitudinal incision and double transverse incision, leading to the creation of new special volar plates setups adapted to each technique’s pitfalls and benefits [85, 86]. Unfortunately authors do not have any personal experience with MIPO technique.
Arthroscopic wrist arthrolysis is indicated in situations of posttraumatic wrist rigidity. It can be performed in radiocarpal joint, midcarpal joint and even in DRUJ. The most frequent clinical pathological conditions are adhesive capsulitis and arthrofibrosis of the wrist. Capsulitis is due to ligament and/or capsule contractures, and wrist arthrofibrosis is usually due to osseous band fibrosis of the radius and/or first row carpal bone(s) from a radius articular fracture. These two conditions can be associated in the same case [87]. The technique of the arthroscopic arthrolysis of the wrist was presented by R. Luchetti et al. in 2006. In radiocarpal joint almost all possible portals, including volar portals must be used during this surgery. It could be difficult to orient in the joint and to triangulate instruments because of the fibrotic adhesions inside the joint. Once they are removed (Figure 38a
Intraarticular adhesions (a) after removal (b).
Complications – in cases when osteochondral lesions of various severity are present during the procedure of the arthrolysis, it is quite common for fibrotic bridges to reform in a few months and provoke partial or complete radiocarpal ankylosis. The use of articular instruments and motorized instruments can cause unwanted osteoarticular lesions (chondral scuffing, ligament injuries etc.) that can manifest themselves postoperatively in the form of pain or wrist instability [87, 89].
The incidence of acute scaphoid fractures is about 70% of all carpal fractures and 11% of all wrist fractures. Young men in the 2nd and 3rd decade of life are the main target population of this injury. Two-thirds of all scaphoid fractures occur in the waist area and 60 – 85% are non-dislocated fractures. Distal third is affected in 25% of cases, but proximal third in 5-10% of fractures [90]. Two morphological bone types are identified: type I or full scaphoids and type II or slender scaphoids. Type I possess more robust internal vascular network than type II scaphoids which may prove to be related to development of nonunion, avascular necrosis or Preiser disease [91]. Indications for surgical treatment are: displacement greater than 1 mm, commination, open fracture, scaphoid fracture with perilunate dislocation, associated carpal instability – scapholunate angle greater than 60°, radiolunate angle greater than 15°, as well as angulation of the scaphoid – intrascaphoid angle greater than 35° and height to length ratio greater than 0.65 [92].
In cases when volar approach with retrograde screw insertion is chosen, arthroscopic treatment of scaphoid fracture has to be started with a small, anterior volar incision through which a 1-mm K-wire is inserted into the scaphoid under fluoroscopy control. This step can be the most difficult one of the entire procedure. If a rolled-up drape is placed under the wrist to extend it to 60°, the K-wire will be about 45° to horizontal. The K-wire is angled from the distal tubercle toward the middle of the carpus. The second stage includes an arthroscopic evaluation when the wrist is placed in vertical traction. Usually midcarpal portals (MCU) are used to visualize the fracture site. If the additional reposition is required, the K-wire can be removed from the proximal pole and manual maneuvers as well as hook probe can be used to achieve the correct position. Then cannulated headless compression screw can be inserted when wrist is released from traction. After the compression of the fracture fragments it’s recommended to make a final arthroscopic evaluation of the midcarpal and radiocarpal joints, to verify the compression and length of the screw [93, 94]. The alternative is a dorsal approach. It provides direct unobstructed access to the proximal scaphoid pole permitting the placement of a central axis guide-wire for antegrade screw implantation [95, 96].
Active wrist motion exercises are initiated immediately or within 10 days after surgery. Strengthening exercises were delayed until healing was well established on X-rays of the scaphoid, usually 3 to 4 months after surgery [93, 97].
Acute scaphoid fractures are often missed and patients return with pain when delayed union or nonunion manifests (Figure 39a).
(a) X-ray of scaphoid nonunion, (b) shaver in the nonunion site, (c) defect of the scaphoid after removal of debris, (d) fixation of the scaphoid with K-wires, (e) defect filled with bone graft, (f) final x-ray after the surgery.
The natural history of untreated scaphoid nonunions is progression to carpal collapse resulting in wrist arthritis and chronic painful disability [98, 99]. Osteoarthritis at the scaphoid-radial styloid joint is significantly associated with dorsiflexed intercalated segment instability (DISI) deformity. An overall incidence of DISI deformity of the wrist is about 56%, and the frequency of DISI pattern increased with longer duration of non-union [100]. Arthroscopic management of scaphoid nonunions without severe deformities or arthritis is effective [101]. This simplifies postoperative recovery, reduces complications, and preserves the wrist’s capsule-ligament complex—and, thus, the scaphoid’s precarious vascularization [102]. Arthroscopic management of scaphoid nonunion is based on the following ideas: that scaphoid vascularity can be preserved because of the minimally invasive nature of arthroscopic surgeries; and that direct visualization with magnification can facilitate accurate debridement of the nonunion site, identify fibrous union and punctate bleeding from fracture site and aid perfect reduction [103].
Principles of the arthroscopic treatment of the scaphoid nonunions are the same as with other surgical techniques: excision of pseudrthrosis, correction of humpback deformity, restoration of the length of the bone, bone grafting and a stable fixation.
Surgical technique includes inspection of the radiocarpal joint via standard portals, synovectomy and arthroscopically guided styloidectomy, if necessary. Arthroscopic treatment of the nonunion is performed via midcarpal portals. The scope is inserted in MCU portal and instruments in MCR, accessory portal (close to the nonunion) or STT portal. If a frank bony defect is encountered, it is curetted with a fine-angled curette or motorized shaver (Figure 39b), until all fibrotic tissue and sclerotic bone are removed.
If the tourniquet is used, at this point it has to be released, to assess the vascularity of the bones. Any humpback and DISI deformity should be identified and corrected. Once the deformity of the scaphoid is corrected, fragments have to be transfixed with K-wires from the tubercle of the scaphoid to the proximal pole for provisional scaphoid fixation (Figure 39c and d).
This process is controlled under arthroscopic and fluoroscopic guidance. The bone graft is taken from the ipsilateral distal radius or iliac crest depending on the amount needed for filling the defect. The bone graft is inserted into a trocar and then the end of the trocar is placed at the nonunion site. The graft is pushed into the trocar with a blunt guide wire until the nonunion site is filled (Figure 39e and f).
Some surgeons recommend to add fibrin glue to protect the graft but others claim that once the scaphoid is fixed and the traction released, the capitate’s native anatomical position will provide sufficient graft stabilization [102, 104]. The fragments are stabilized with screw(s) and/or K-wires. Recorded union rate with this procedure is 86 – 100% [105, 106, 107]. Arthroscopically treated patients achieve faster healing despite shorter time to surgery in the percutaneous group. Local bone grafting is considered as the main reason for this outcome [108].
Thumb CMC joint pain, instability and progressive arthritis is a common problem affecting many patients, especially middle-aged women. Once present, the symptoms of thumb CMC osteoarthritis are typically progressive and may lead to significant functional disability. There are two classifications for thumb CMC osteoarthritis: Eaton – Litter classification which is based on radiological changes [109] (Table 7) and arthroscopic classification developed by [110] (Table 8). He also presented an algorithm for management of the CMC osteoarthritis incorporating arthroscopical stages into radiological classification and subsequent treatment decision-making. Treatment methods depend on the stage of the radiologic and arthroscopic findings and can contain detriment, thermal shrinkage, correctional osteotomy of the 1st metacarpal base as well as arthroscopic resection with different interposition arthroplasties or suspensionplasties.
Class | Description |
---|---|
Synovitis phase. Subtle carpometacarpal joint space widening | |
Slight carpometacarpal joint space narrowing, sclerosis, and cystic changes with osteophytes or loose bodies < 2 mm | |
Advanced carpometacarpal joint space narrowing, sclerosis, and cystic changes with osteophytes or loose bodies > 2 mm | |
Arthritic changes in the carpometacarpal joint as in Stage III with scaphotrapezial arthritis |
Eaton-Litter radiological classification of 1st CMC arthritis.
Stage | Arthroscopic changes |
---|---|
I | Intact articular cartilage Disruption of the dorsoradial ligament and diffuse synovial hypertrophy. Inconsistent attenuation of the anterior oblique ligament (AOL) |
II | Frank eburnation of the articular cartilage on the ulnar third of the base of first metacarpal and central third of the distal surface of the trapezium. Disruption of the dorsoradial ligament + more intense synovial hypertrophy. Constant attenuation of the AOL |
III | Widespread, full-thickness cartilage loss with or without a peripheral rim on both articular surfaces. Less severe synovitis. Frayed volar ligaments with laxity |
With recent advances in arthroscopic techniques, partial trapezectomy with or without different soft tissue or implant interposition has been reported with good results [111, 112, 113, 114]. Theoretical advantages over open procedures include a decreased risk of neurovascular injuries, smaller incisions decreased postoperative pain and shorter overall recovery time. On the other side, this technique has several disadvantages, including increased setup and operative procedure time, increased surgical training, increased equipment cost and additional x-ray fluoroscopy time [115].
There is growing evidence that techniques involving use of no interposition result in a high rate of satisfactory outcomes [116, 117]. Cobb et al. in 2015 compared outcomes of patients treated with or without tendon interposition and found no difference in outcomes.
Another promising technique is an arthroscopic hemitrapeziectomy and suture button suspensionplasty [118, 119].
Authors have their own small experience (6 patients) with arthroscopic hemitrapezectomy and interposition arthroplasty with RegJoint™ implant (Figure 40a–e). The follow up is 12 to 36 months without any severe complications. Marcuzzi et al. in 2020 published long-term results with open technique [120]. They found dissapointing radiological results with an almost complete collapse of the metacarpal base on the distal pole of scaphoid in more than 80% of patients. However the results did not correspond with clinical outcomes that were very satisfactory. We hope that arthroscopical technique preserving the dorsal capsule will improve the outcomes, but further investigations are necessary.
(a) preopertive x-ray of 1st CMC arthritis, (b) CMC portals connected with skin incison, (c) Regjoint Scaffold sutured before insertion via elongated CMC portal, (d) Implant pulled in and positioned in the site, (e) transfixation of the bones and implant with K-wire.
Complication rate with arthroscopic CMC arthroplasties is about 4% and include as follows: CRPS, ulnar or radial sensory nerve neuropathy, transitory numbness near the portal, prolonged hematoma, FPL tendon rupture and superficial skin necrosis [121].
The development of our understanding and management of TFCC tears has been developed with important contributions including Palmer’s classification of TFCC tears (Table 9), G. Poehling’s inside-out suture technique, F. Del Piñal’s all inside suture technique [122], A. Atzei’s and R. Luchetti’s TFCC tear classification (Figure 41), T. Nakamura’s anatomical and clinical studies [123] and J.R. Haugstvedt’s developed techniques for the TFCC tears and lunotriquetral tears as well as studies about DRUJ functional anatomy and pathomechanics [1, 124].
Class or subclass | Description |
---|---|
1A | Central slit |
1B | Ulnar avulsion with or without distal ulnar fracture |
1C | Distal avulsion(carpal attachement) |
1D | Radial avulsion with or without sigmoid notch fracture |
2A | TFCC wear |
2B | TFCC wear, lunate or ulnar chondromalacia |
2C | TFCC perforation, lunate or ulnar chondromalacia |
2D | TFCC perforation, lunate or ulnar chondromalacia, lunotriquetral ligament tear |
2E | TFCC perforation, lunate or ulnar chondromalacia, lunotriquetral ligament tear, ulnocarpal osteoarthritis |
Palmer’s classification of TFCC tears.
A. Atzei’s and R. Luchetti’s classification of TFCC tears (reprinted by permission).
TFCC – is one of the instrinsic ligaments of the wrist, with load bearing function between the lunate, triquetrum and ulnar head. TFCC acts as stabilizer for the ulnar aspect of the wrist joint [125].
TFCC consists of five parts: fibrocartilaginous disc and the meniscal homolog, volar ulnocarpal ligaments (ulnolunate and ulno-triquetral), dorsal and volar radioulnar ligaments (each with a superficial and deep part), ulnar collateral ligament as well as the floor of the fibrous 5th and 6th extensor compartments [125, 126].
Palmer had a two-dimensional view of the TFCC [127]. Nakamura described it as a three dimensional structure, and separated TFCC in three components: the distal component which acted like a hammock to suspend the carpus, the triangular ligament as the proximal component which stabilized the radius to the ulna, and the UCL as the ulnar component which stabilized the carpus to the ulna [128, 129]. Atzei and Luchetti updated previous “hammock” concept to the novel “iceberg” concept [130]. In analogy with the iceberg, during arthroscopy of the radiocarpal joint (RCJ) the TFCC shows its “emerging” tip. The tip of the iceberg represents that part of the TFCC that functions as a shock absorber. This part is much more smaller than “submerged” part which can be seen only in case of the DRUJ arthroscopy. The submerged TFCC represents the foveal insertions of the TFCC and functions as the stabilizer of the DRUJ and of the ulnar carpus. The larger size of the submerged portion of the iceberg corresponds to its greater functional importance.
TFCC biomechanics:
TFCC stabilizes DRUJ and ulnocarpal joint
TFCC allows the transmission and distribution of forces from wrist onto ulna and provides a gliding surface for the carpus during complex movements of the wrist.
The central disc works as the distribution mechanism for the mechanical stress onto proximal triquetrum and the lunate
Clinical assessment of TFCC tears:
The ulnar fovea sign is the most reliable clinical sign [131], where the patient has the point of tenderness over the ulnar capsule in the area between extensor carpi ulnaris (ECU) and flexor carpi ulnaris (FCU) tendons.
The ballotment test evaluates DRUJ stability. This is a simple and reliable to determine DRUJ laxity [132].
Imaging assessment of TFCC tears:
Radiographs – of limited value for TFCC injury diagnostics, but very important for acute and chronic wrist pain. The presence of ulnar styloid fracture alone or with distal radius fracture is of some importance for the diagnosis of the TFCC tear [133]. The Galeazzi fracture-subluxation is a particular condition that is associated with a TFCC tear [134].
MRI and MRI arthrogram. MRI is more useful to exclude associated pathologies of the ulnar compartment. Comparing specificity and sensitivity of MRI, MRI arthrography and artroscopy for diagnosis of the TFCC tear, confirm the arthroscopy as the gold standard for diagnosis [135, 136].
Arthroscopic examination of TFCC. Three arthroscopic tests are used to check the type of TFCC injury:
The “trampoline sign” – the loss of elasticity of the TFCC – seen in complete avulsion injuries of the proximal and distal portions of the TFCC
The “hook sign”– positive in complete tears of the TFCC and negative in other cases. The hook test is more accurate than the trampoline test to detect foveal tears of the TFCC of the wrist [137]
The “ghost sign” – reverse ““trampoline sign”. This indicates an avulsion of the deep fibers of the TFCC. The sign is negative in distal lesions and positive in isolated proximal lesions.
Atzei’s/Luchetti’s classification also shows the stability/instability of the DRUJ joint and possible surgical treatment to corresponding TFCC tear.
An algorithm of treatment according to Atzei’s/Luchetti’s classification:
CLASS 0 – isolated styloid fracture without TFCC tear. Frequently associated with distal radial fractures. DRUJ is stable. If isolated treatment is wrist splinting for 3 weeks.
CLASS 1 – periferal tear of the TFCC distal component, the DRUJ may be slightly lax. Hook test negative. Small tear requires 4 weeks of wrist immobilization followed by two weeks splinting. A larger tears requires arthroscopic TFCC suture.
CLASS 3 – periferal tear of the TFCC proximal component. Mild to severe laxity of the DRUJ joint. Hook test is positive. TFCC foveal reattachment is required by transosseus sutures or a suture anchor.
CLASS 4 – nonrepairable peripheral TFCC tear due to the massive defect or poor healing potential. This condition requires reconstruction with tendon graft.
CLASS 5 – DRUJ arthritis following peripheral TFCC tear. Arthroscopy shows significant degenerative or traumatic cartilage defect. Suggested treatment – arthroplasty or prosthetic replacement.
In cases of peripheral repairable TFCC tears, authors use debridement and synovectomy to detect and refresh the site of the rupture. Usually 6R portal is used for shaver and 3-4 portal for visualization. Occasionally 6 U portal can be used if tears are localized more volarly. Once the size of tear is recognized, portal can be elongated to vizualise extensor tendons by transillumination of the capsule. Needle with suture loop is passed a little bit proximally from the margin to the TFCC to capture capsule together with the TFCC. Once recognized in the joint, suture is captured with mosquito forceps and one part of it passed via the portal or, in cases if several sutures necessary, via extra holes in the capsule. Location of the extensor tendons is evaluated to avoid capture of them in the suture and knots are tightened extra-articulary (Figure 42a–c). The reattachment can be performed with an inside-out, outside-in, or all-inside technique, providing good to excellent results, which tend to persist over time, in 60–90% of cases [138].
Periferal, reconstructable TFCC tear. (a) tear after the debridement, (b) sutures passed the TFCC and capsule, (c) sutures tightened extra-articulary via 6R portal.
In cases of proximal reparable TFCC foveal detachment, we prefer to use the transosseus refixation of the TFCC described by T. Nakamura [139]. We use the original Arthrex target device through 6R portal and an approximately 1 cm longitudinal incision on the ulnar side of the ulnar cortex, 10–15 mm proximal from the tip of the ulnar styloid. Then target device is set on the TFCC and two parallel channels with original 1.6 mm K-wires are made from the ulnar cortex through the head of ulna and TFCC. Then follows a manipulation with needles, suture loops and main suture, where different techniques of the suture insertion are possible (Figure 43a–c).
Proximal reparable detachement of TFCC. (a) Arthrex targeting device over the foveal area, view from 4-5 portal, (b) sutures passed through the TFCC and head of ulna, (c) sutures tightened and TFCC reattached to foveal region.
After the main suture is passed through the bone channels to make outside-in pullout suture of the TFCC to the fovea and tensed with knot over the cortex. Another option is to hide the knot inside the ulna and tense with a push-lock anchor.
After treament includes 2-3 weeks in long arm plaster cast, following 3 weeks in short cast with following rehabilitation after the cast is removed.
In cases of unrepearable TFCC injuries or degenerative tears, an anatomic reconstruction with free tendon graft is recommended. The arthroscopic reconstruction is a mini invasive option of the Adams-Berger procedure [140], but it requires an experience in arthroscopic surgery. Nowadays tendon grafts can be fixed in the bone channel with interference screw, instead of the original procedure where tendons were wrapped around the bone and sutured together. Nevertheless, when well done, this technique provides good stabilization of the DRUJ, while maintaining good mobility of the wrist in all directions [141].
A systemic review by Liu et al. about the surgical repair of TFCC tears confirms that arthroscopic techniques achieve overall better outcomes compared with open repair technique. For foveal tears, transosseous sutures achieve overall better functional outcomes compared with suture anchors. Current evidence demonstrates that TFCC repair achieves good clinical outcomes, with low complication rates [142].
During the last 4 decades wrist arthroscopy has turned from the diagnostic tool of some enthusiasts to the widely used therapeutical complex for treatment of different wrist pathologies. Evolution of the wrist arthroscopy equipment as well as skills of the surgeons has allowed us to improve our knowledge of the wrist anatomy and biomechanics. Wrist arthroscopy is especially valuable for evaluation of intra-articular soft tissue pathologies. Furthermore - arthroscopic classification systems have been described for TFCC, SLIL and LTIL lesions, Kienböck disease, 1st CMC joint, etc.
Wrist arthroscopy techniques have proved superiority over the open techniques with lower complication rates and recurrence rates. For example in wrist ganglion surgery open surgical excision had a mean recurrence of 21%, compared with a recurrence rate of 59% for aspiration. The lowest rate was observed with arthroscopic excision, with a recurrence of 6% across all studies [40].
Arthroscopic scapho-lunate ligamentous repair is now considered the less damaging and denervating than open repair [143]. Although several arthroscopic SLIL reconstruction methods as well as arthroscopic reconstruction technique for LTIL tears have been described, these surgeries are challenging, therefore different modalities and variations of open procedures are still actual and used. Some arthroscopic techniques require a long learning curve and years of practice.
A systematic review about arthroscopic vs. open TFCC surgeries shows comparable results between open and arthroscopic procedures, in terms of DRUJ re-instability and functional outcome scores. There is insufficient evidence to recommend one technique over the other in clinical practice [144]. However arthroscopic procedures are less aggressive and may allow quicker recovery, especially in athletes [145]. In combination with a TFCC procedure, the ulnar variance can readily be assessed. Ulnar abutment or impingement can be directly visualized through dynamic assessment. Whilst ulnar shortening is an extra-articular procedure, the arthroscopic wafer procedure allows for intra-articular treatment without the need for hardware. This overcomes the issues of hardware prominence and circumvents non-union rates of about 10%, while also allowing for a quicker return to work [145, 146].
Wrist arthroscopy is beneficial also in the treatment of distal radius articular fractures, because it helps to visualize articular gaps and step-offs unrecognized with the fluoroscope alone. Although arthroscopically assisted DRF surgeries have superior long-term outcomes in several parameters [76], the advantage of this procedure, however, is the recognition of associated soft tissue lesions which can be prevented if recognized.
The next aspect is professional training and experience of the surgeon. Leclercq et al. in the multicenter study organized by EWAS found that surgeons who perform less than 25 wrist arthroscopies per year have a complication rate of 12.06%, whereas among the surgeons who perform more than 75 wrist arthroscopies per year, the complication rate is 3.95%. Surgeon with less than 5 years of practice in wrist arthroscopy have complication rate 13.6%, whereas surgeons who had 15 or more years of practice complication rate is only 2.3%. Surgeons with longer practice and greater amount of wrist artrhroscopies performed per year, more often are doing therapeutical arthroscopies. This ratio is up to 87% of procedures comparing to less experienced colleagues who perform therapeutical procedures in about 60.5% of cases [147].
Arthroscopy has assumed an important place in wrist surgery. It requires specific operative skills, training, technical equipment and patience, because these surgeries sometimes take more time than expected, even if you think, you are trained enough (my personal experience). Minimally invasive surgery is a trend of our century and arthroscopic treatment of wrist pathologies has already demonstrated promising outcomes and it’s superiority over open surgical procedures.
The great majority of current Augmented Reality (AR) applications are built using general purpose processors as development platforms where the processing tasks are executed in software. However, software execution is not always the best solution for the high intensive requirements of the many processing tasks involved in AR, and it inevitably constrains frame rate and latency, which compromises real time operation, and magnifies size and power consumption, hindering mobility. These limitations make the spread of AR applications more difficult. This is particularly remarkable in the case of mobile real time applications.
\n\t\t\tTo overcome the aforementioned constraints in the design of embedded AR systems, this chapter presents a hardware/software co-design strategy based on Field Programmable Gate Array (FPGA) devices and Electronic System-Level (ESL) description tools as an alternative to the traditional software-based approach. Modern FPGAs feature millions of gates of programmable logic, with dedicated hardware resources and with the widest range of connectivity solutions. FPGA internal structure makes itself perfectly suitable for exploiting parallelism at several levels. Moreover, because of its flexibility, it is possible to implement not only specific algorithms, but also AD/DA interfaces, controllers, and even several microprocessors, what makes it feasible to build more complex and powerful Systems on a Chip (SoC) with improved performance and reduced costs, size and power consumption. FPGA (re)programmability is also a key factor, which provides not just reduced time to market and design flexibility, but also in-the-field upgradability and intellectual property protection. Thanks to these characteristics, FPGAs are giving rise to a new paradigm in computation named Reconfigurable Computing. ESL, on the other hand, is an emerging electronic design methodology that focuses on building models of the entire system with a high-level language such as C, C++, or MATLAB, which are later used by improved electronic design tools to generate an automated and correct-by-construction implementation of the system. ESL codesign tools allow for developers with little or no prior hardware design skills to implement complex systems composed of mixed software and application-specific hardware modules.
\n\t\t\tThe objective of this chapter is to provide a clear vision of the possibilities of FPGA devices and the new development methodologies for embedded AR systems. To do it so, the authors explain the FPGAs key features which make them suitable for the implementation of AR applications. The design flow and tools for hardware description and hardware/software co-design from low to the highest level are described. A survey of the most noteworthy FPGA-based works in image processing, computer vision, computer graphics, multimedia, communications and wearable computing is presented. Finally, the chapter is completed with an example which illustrates the advantages of the FPGA-based approach as platform for developing AR applications: a portable real time system for helping visually impaired people. This system enhances the patient’s knowledge of the environment with additional video information using a see-through head mounted display. The description of its main processing cores for video acquisition and processing, for hand recognition, for the user interface, etc. and the evaluation of their performances highlight the advantages of the FPGA-based design and reveal the key topics for the implementation of AR systems.
\n\t\tAfter a successful decade of exploration and consolidation of fields and applications, it is time for AR to break the border of the research domain and reach the common people domain. For it, the user needs to feel AR as a part of his own body, not as an external an uncomfortable artefact. Inevitably, this entails ubiquity and mobility. For such a qualitative jump, one of the major challenges that AR has to face is the hardware under applications and the development of new platforms for interaction (Veas & Kruijff, 2008). It is the key to find the optimum solution to the complicated trade-off between quality, speed, power and size. Most AR research published to date relies on the use of general purpose computing hardware to perform computations, render computer graphics and provide video overlay functionality. Systems that rely on general purpose computing hardware are larger in size and consume more power than those which have devices customised for specific tasks. When working in indoor environments, such issues are rarely considered since the systems are not normally required to be mobile, but the challenge of ubiquity and mobility raises a new scenario. In it, doubtlessly, computer processing hardware is one of the research topics where an extra development effort must be done in order to provide a more effective AR experience (Zhou et al, 2008).
\n\t\t\tOver the last years, several hardware platforms have been introduced to support mobile AR in two different directions: the head-mounted display direction using laptops in backpack systems, and the handheld direction using lightweight portable devices. Both directions share a design feature: the great majority of AR applications consist of software being executed on general purpose processors. However, not all applications can be solved through the use of the traditional software-based approach since it imposes limitations that force the designer to choose between robustness/compactness, power consumption and processing power, what, in last term, difficult the spread of AR.
\n\t\t\tIn spite of their importance, only a minority of papers describing AR mobile systems report on features such as timing performance, frame rates, power consumption, size, weight, etc., which, in the last instance, determine the viability of the AR application. When these data are provided, they show important weak points, like in the case of the ArcheoGuide application described in (Dähne & Kariginannis, 2002), where the incapability of the hardware platform of the system of parallelizing processing tasks causes that authors rule out a gesture recognizer based on interaction due to its interference with the video tracking performance. This is also the case of the video see-through AR system on cellphone described in (Möhring et al, 2004), which hardly can manage four 160×120 12-bit colour frames per second.
\n\t\t\tSuch limitations lead to approaches where the AR device tends to be ‘dumb’, acting simply as a viewport of the AR application with the largest part of the processing taking place in remote servers. This is the philosophy behind the AR-phone (Assad et al, 2003). In it, the system performance relies on the wireless networking. Without reporting frame rates, the authors admit the convenience of moving some parts of the processing into the user interface module with the aim of avoiding the overload in the communication that ruins the performance. The mobile phone-based AR application for assembly described in (Billinghurst et al, 2008) is also a client/server architecture where the complex processing is executed on a PC. It works with still images instead of video and the virtual model is quite simple, but the data transmission between the phone and the PC over the WLAN rises the time to send and receive an image up to 3.4 seconds. A similar client/server implementation is presented in (Passman & Woodward, 2003) for running AR on a PDA device. It uses a compression algorithm developed by the authors to relieve the communication of the rendering data overhead, but the best case refresh rate hardly reaches two frames per second. Another interesting approach can be found in (Wagner et al, 2005), where the authors present a system architecture for interactive, infrastructure-independent multi-user AR applications which runs on a PDA (personal digital assistance).
\n\t\t\tDue to these limitations, some applications resign to video processing on behalf of usability. AR on-demand (Riess & Stricker, 2006) presents a practicable solution on low-end handheld devices where images of the real scene are taken only when needed, superposing real-time virtual animations on that single still image. The information is not blended in the field of view of the user, but can be easily watched over the display of a PDA or a head worn display beside the eye. The goal of this approach is to develop a system which does not dominate the user, but offers support when required.
\n\t\t\tThe use of image sensors is probably the most common way to capture the real environment and enhance it with virtual objects, i.e. to create augmented reality. For this reason, it can serve us well to understand the role of FPGAs and custom-made hardware platforms for AR applications. In the cases where designs must be prepared to meet frame rate, image resolution, algorithm performance, power consumption and size requirements, traditional platforms based on desktop computers and their corresponding slight variations based on laptops are unsuitable. Some examples of these works can be found in (Piekarski et al., 2001; Feiner et al, 1997; Hoellerer et al, 1999). Although these works may be interesting for validating their respective systems for their corresponding intended applications, the proposed devices appear to be complex, heavy, expensive and fragile. Developments made in standard microprocessors-based units are not efficient or inconvenient for the purpose of unconstrained mobility. The authors of (Piekarski et al., 2004) noticed this fact, and proposed modifications in backpack designs which aim to improve its size, weight and power consumption. The authors conclude that FPGAs and specialized video overlay units result beneficial to minimize power and to accelerate computation. In (Johnston et al., 2003), the authors point to FPGAs and developments in hardware for embedded systems as the most likely alternative platform for real-time performance of AR devices. Indeed, parallelism and concurrency can be exploited for image processing while keeping power consumption low, bringing a solution to the challenge of embedded image processing.
\n\t\t\tA good example of exploiting FPGAs with an image sensor can be found in (Matsushita et al, 2003), where a FGPA device controls a fast CMOS image sensor for ID recognition. In (Foxlin & Harrington, 2000), a self-reference head and hand tracker is presented and its applicability in a wearable computer is shown. Smith and colleagues in (Smith et al, 2005) improved the work presented in (Piekarski et al, 2001) by migrating the proposed hand-tracking algorithm from the laptop to an FPGA. Indeed, this new development allows a further miniaturization of the system and minimizes power consumption. In (Luk et al, 1998; Luk et al., 1999), some good initial examples of the use of reconfigurable computing for AR are shown. In concrete, video mixing, image extraction and object tracking are run on an FPGA-based platform. In this line, our group presented in (Toledo et al, 2005) a fully FPGA based AR application for visual impaired individuals affected by tunnel vision. A Cellular Neural Network extracts the contour information and superimposes it on the patient’s view. Finally, the authors of (Guimaraes et al, 2007) present an interesting platform which aims to help developers on the construction of embedded AR applications. The infrastructure is based on FPGA and enables the creation of hardware based AR systems.
\n\t\t\tThis short review is enough to come to the conclusion that mobile AR applications are severely constrained by the up to date usual software-based approach and that many real-time applications require algorithmic speedup that only dedicated hardware can provide. Hardware implementation of algorithms and processing tasks can considerably improve the performance and by the hence the utility/viability of the AR system.
\n\t\tThe introduction of the Field-Programmable Gate Array (FPGA) devices about 25 years ago gave rise to the reconfigurable computing concept. Early FPGA generations were quite limited in their capacities. Nowadays, the most advanced manufacturing technologies are used to feature devices with millions of gates of programmable logic, with dedicated hardware resources, with the widest range of system connectivity solutions, enabling more complex and powerful systems on a single chip.
\n\t\t\tReconfigurable hardware offers a trade-off between the very specialized solution of application specific integrated circuits (ASIC) and the inefficiency of general purpose processor (GPP), combining the desirable high performance of ASICs with the characteristics of system flexibility of GPPs. Like ASICs, it involves hardware implementation and consequently parallelism and high performance. Like GPPs, it provides reconfigurability, and hence, flexibility and rapid prototyping. The key of reconfigurable hardware lies on that the flexibility is provided by the hardware design rather than by software-programmable hardware. With clock frequencies an order of magnitude lower than that of typical microprocessors, FPGAs can provide greater performance when executing real-time video or image processing algorithms as they take advantage of their fine-grained parallelism.
\n\t\t\tThere are several companies that produce diverse flavours of FPGAs: Xilinx, Altera, Atmel, Lattice Semiconductor, Actel, SiliconBlue Technologies, Achronix or QuickLogic are the main competitors. At an architectural level, however, all of them share some common characteristics, which define an FPGA as a regular structure of programmable modules, including three types of resources:
\n\t\t\tLogic Blocks (LB), used to implement small parts of the circuit logical functions.
Input/Output Blocks (IOBs) that connect internal resources with the device pins, and usually adapt the electrical characteristics of the signals to interface the output world.
Routing channels with programmable Interconnect Blocks (IBs), which allow the connection between LBs or between these and the IOBs.
In an FPGA, both the connection of the wire resources and the functionality of the logic blocks can be programmed by the user. Logic blocks and routing architectures differ from vendor to vendor, but their regular distribution make all of them well suited for highly parallelizable algorithms, composed of bit-level operations that can be distributed in regular structures. Nevertheless, they are ill suited to high precision arithmetic operations, such as large operand multiplication or floating-point calculations. The presence of on-chip memory is also limited, thus many applications require the existence of an external RAM chip. The data transfer increases circuit delays, and increases power consumption and board area.
\n\t\t\tTo overcome these difficulties, researchers and manufacturers proposed new architectures with specific purpose resources and advanced configuration characteristics. Up to Several hundred dedicated MACs (Multiply and Accumulate) modules are included in the larger devices for fast DSP calculations. To increase data storage capabilities, RAM blocks are distributed over the circuit, providing several Mbits of dedicated memory. Recently some devices come with dedicated interface controllers, such as Ethernet MAC, USB or PCI bridges. The list includes also AD/DA converters, PLLs for highly customizable clock generators, or hard-core microprocessors (with much better performance than their soft-core counterparts). As an example, Xilinx’s latest Virtex 5 chips include one or two embedded PowerPC™ 405 processors. These 32-bit RISC processors from IBM Corporation are included as hard-cores, making them run at around 450MHz, without decreasing the number of user (LB) resources. Several real-time operating systems (RTOS), included different Linux porting are available for this processor, allowing the designer to centre his efforts in the application-specific parts of the project.
\n\t\t\tDevices can be classified with respect to their
IOBs can be configured as input, output or bidirectional ports. They frequently include on-chip pull-up/down resistors and tri-state buffers. Connectivity is guarantied by supporting main standards, including 3GIO, Infiniband, Gigabit Ethernet, RapidIO, HyperTransport, PCI, Fibre Channel, Flexbus 4, between others. Some chips, like the Virtex 6 family, accommodate several multi-gigabit transceivers to perform serial I/O from 3.125, up to 11Gbps.
\n\t\t\tMost devices can be programmed by downloading a single-bit stream into the configuration memory. The device’s configuration is typically memory-based, using any of the SRAM, EPROM, EEPROM or the most recent flash technologies. Other OTP (one-time programmable) devices use fuse or antifuse technologies (for a detailed description of the different technologies and chip characteristics, the interested reader is referred to companies’ web pages).
\n\t\t\tSRAM devices are the dominant technology nowadays. However, SRAM cells are volatile, meaning that the stored information is lost when power is not applied. These devices require and external “boot”, and are typically programmed from a host processor or a non-volatile memory after chip reset. Memory-based FPGAs have the advantage of been re- and in-system programmable. Devices can be soldered directly to the PCB, without using special sockets. If the design changes, there is no need to remove the device from the board, but it can be simply re-programmed in-system, using typically a JTAG interface. On the other hand, the non-volatile EEPROM and flash devices are better protected against unauthorized use and reverse-engineering, because the programming information is not downloaded to the device at power up.
\n\t\t\tThe re- and in-system programmability of the memory based FPGAs has opened a broad area of new application scenarios. The term (Re-) Configurable Computing refers to computers that can modify their hardware circuits during system execution. The key for this new computing paradigm has been the development of new FPGAs with extremely quickly configuration rates. While first devices required several seconds to get programmed, in newer FPGAs the configuration download can be done in about one millisecond, and devices with configuration times of about 100 s are expected in the next years.
\n\t\t\t\tDynamic reconfiguration can be used in a number of ways. The least demanding technique consists of switching between several different configurations that are prepared beforehand, what enables to perform more computational algorithms than those permitted by the physical hardware resources. This could be seen as the hardware equivalent of quitting one program and running another. When faster programming times are available, reconfiguration can be done in a kind of context swapping: the FPGA reconfigures itself time-sharing the execution of different tasks, making the illusion that it is performing all its functions at once (multi-tasking). An example of this techniques was used in (Villasenor et al., 1996) to build a single-chip video transmission system that reconfigures itself four times per video frame, requiring just a quarter of the hardware needed by an equivalent fixed ASIC.
\n\t\t\t\tThe most challenging approach to dynamic reconfiguration, and surely the most powerful, involves chips that reconfigure themselves on the fly, as a function of requirements emerged during algorithm execution. In this computing system, if a functional unit is missed, it is recovered from the resources store and placed on the chip, in some cases replacing the space occupied by another not-in-use circuit. The problem here is double: loading the proper configuration
Partial reconfiguration enables the remote upgrade of hardware across a network, by delivering new
With their re-configurability characteristics and the introduction of new special-purpose resources blocks, FPGAs offer a number of advantages over classical design methodologies based on general purpose processors or even the more specialized Digital Signal Processors (DSP), such as unmatched parallelism, versatility, and short time to market. Moreover, Re-configurable Computing has been presented as a promising paradigm that will compete against the traditional von Newmann architecture and its parallel enhancements (Hartenstein, 2002), (Hartenstein, 2004). Augmented Reality applications and, particularly, embedded ones, can take huge benefits from these emerging so called config-ware technologies.
\n\t\t\tFrom a historical point of view, the first FPGA design tools where traditional schematic based editors coupled with a physical design software that performed the
Been textual, HDLs are more manageable than schematics. One the other hand, both of them support structural descriptions, however, only HDLs allow for higher-level “behavioural” descriptions. Finally, HDLs have specific constructions for modular design and component re-usability. This last characteristic has become crucial, as the size of the standard design has grown from ten or cents of
Another important characteristic is that high-level HDL descriptions, contrary to schematic descriptions, are technology independent, what allows the designer to synthesize his project over a number of FPGA devices with minor changes at the description level. Different FPGA vendors and architectures can be benchmarked until an optimal implementation is met, with little impact on the design time.
\n\t\t\t\tDespite of the use of HDLs, the current design tools and methodologies have became inadequate to effectively manage the tens of millions gates that the silicon technology allows gather together in a single chip, furthermore when the pressure to reduce the design cycle increases continuously.
\n\t\t\t\tThe tendency has been towards the use of pre-designed and pre-verified cores trying to bridge the gap between available gate-count and designer productivity. Instead of developing a system from scratch, designers are looking at effective methodologies for creating well-verified reusable modules that can be incorporated in a “mix & match” style to the application-specific blocks. These reusable hardware modules are called
The design process based on HDLs like VHDL, Verilog, etc., is not exempt of difficulties, as these traditional methodologies still require deep hardware skills from the designer. Moreover, the high integration levels of current chips have transformed the concept of System On a Chip (SoC) into reality, increasing design complexity up to an unprecedented level. A typical SoC consists of one or several microprocessors/microcontrollers, multiple SRAM/DRAM, CAM or flash memory blocks, PLL, ADC/DCA interfaces, function-specific cores, such as DSP or 2D/3D graphics, and interface cores such as PCI, USB and UART (Figure 1).
\n\t\t\t\tThe intensive use of predesigned IP cores can just mitigate the problem, but in a SoC project, designers can not describe the hardware-specific modules at the Register Transfer Level (RTL), as the HDL methodologies propose, and then wait for a hardware prototype before interacting with the software team to put the design together. New EDA (Electronic Design Automation) tools must incorporate system-level modelling capabilities, such that the whole system, software and hardware, can be verified against its specifications right from the beginning of the design process. This requires an integrated hardware-software co-simulation platform that permits to confer hardware engineers the same level of productivity of software engineers.
\n\t\t\t\tTo meet the challenges posed by the SoC complexity, new languages, tools and methodologies with greater modelling capabilities are been developed. In the area of design languages, there has been a lot of discussion about the role and applicability area of the various existing and new languages. SystemC, SytemVerilog, Verilog 2005, Analogue and Mixed-Signal versions of Verilog and VHDL, or Vera are some of the new proposals.
\n\t\t\t\tStructure of core-based system on a chip
One of the most promising alternatives is been developed by the Open SystemC Initiative (OSCI), a collaborative effort launched in 1999 among a broad range of companies to establish the new standard for system-level design. SystemC (OSCI, 2006) (Black & Donovan, 2004) is a modelling platform consisting of C++ class libraries and a simulation kernel for design at the system and register transfer levels. Been a C-based language, SystemC can bring the gap between software engineers used to work with C and C++, and hardware engineers, that use other languages such as Verilog or VHDL. Furthermore, a common specification language would favour the creation of design tools that allow designers to easily migrate functions from hardware into software and vice versa.
\n\t\t\t\tFollowing this approach, a new generation of tools for highly complex circuit design is been developed. This new methodology, known as ESL (Electronic System Level), aims to target the problem of hardware-software co-design from system-level untimed descriptions, using different flavours of High Level Languages (HLLs), such as C, C++ or Matlab. The main difference between tools is the projection methodology used to implement a given algorithm, that is, the approach used to partition and accelerate that algorithm; some tools used a fixed processor based architecture that can be expanded with custom application-specific coprocessors; other are best tailored to create just custom hardware IP modules that could be later integrated in larger systems; some provide a flexible processor architecture whose instruction set can be expanded with application specific instructions supported by custom ALUs/coprocessors; finally, some of them are intended to provide a complete SoC design environment, giving support for custom hardware modules design, standard microprocessors, application software and the necessary hw-to-hw and hw-to-sw interfaces. A detailed review of these tools is beyond the scope of this chapter, so we will just summarize some of them in no particular order, in Table 1. The interested reader can find an exhaustive taxonomy of the design methodologies and ESL design environments commercially or educationally available in (Densmore & Passerone, 2006).
\n\t\t\t\tCompany | \n\t\t\t\t\t\t\tWeb page | \n\t\t\t\t\t\t\tProduct | \n\t\t\t\t\t\t
Bluespec | \n\t\t\t\t\t\t\twww.bluespec.com | \n\t\t\t\t\t\t\tBluespec Development Workstation | \n\t\t\t\t\t\t
CriticalBlue | \n\t\t\t\t\t\t\twww.criticalblue.com | \n\t\t\t\t\t\t\tCascade | \n\t\t\t\t\t\t
Codetronix | \n\t\t\t\t\t\t\twww.codetronix.com | \n\t\t\t\t\t\t\tMobius, XPSupdate | \n\t\t\t\t\t\t
Impulse Accelerated Tech. | \n\t\t\t\t\t\t\twww.impulseaccelerated.com | \n\t\t\t\t\t\t\tCoDeveloper | \n\t\t\t\t\t\t
Mitrionics | \n\t\t\t\t\t\t\twww.mitrionics.com | \n\t\t\t\t\t\t\tMitrion Software Development Kit | \n\t\t\t\t\t\t
Nallatech | \n\t\t\t\t\t\t\twww.nallatech.com | \n\t\t\t\t\t\t\tDIME-C | \n\t\t\t\t\t\t
Poseidon Design Systems | \n\t\t\t\t\t\t\twww.poseidon-systems.com | \n\t\t\t\t\t\t\tTriton Builder | \n\t\t\t\t\t\t
System Crafter | \n\t\t\t\t\t\t\twww.systemcrafter.com | \n\t\t\t\t\t\t\t\n\t\t\t\t\t\t |
ARC International | \n\t\t\t\t\t\t\twww.teja.com | \n\t\t\t\t\t\t\tARChitect, others | \n\t\t\t\t\t\t
Xilinx Inc. | \n\t\t\t\t\t\t\twww.xilinx.com | \n\t\t\t\t\t\t\tAccelDSP, System Generator | \n\t\t\t\t\t\t
Graphics | \n\t\t\t\t\t\t\twww.mentor.com | \n\t\t\t\t\t\t\tCatapult C | \n\t\t\t\t\t\t
Cadence Design System | \n\t\t\t\t\t\t\twww.cadence.com | \n\t\t\t\t\t\t\tC-to-Silicon Compiler | \n\t\t\t\t\t\t
Some companies providing ESL tools.
In this section, we analyze the main features and workflow of CoDeveloper™, an ESL tool from Impulse Accelerated Technologies, Inc. (Impulse, 2009) used for hardware-software co-design, to evaluate its suitability for the non-hardware specialist scientist in general, as in the case of most AR researches. From our experience, we then provide some keys to get better results with this tool, which may be easily generalized for similar tools, with the aim of making the reconfigurable hardware approach for embedded AR solutions a bit closer for a broaden number of researchers.
\n\t\t\t\tImpulseC compiler uses the communicating sequential process (CSP) model. An algorithm is described using ANSI C code and a library of specific functions. Communication between processes is performed mainly by data streams or shared memories. Some signals can be transferred also to other processes like flags, for non continuous communication. The API provided contains the necessary functions to express process parallelization and communication, as standard C language does not support concurrent programming.
\n\t\t\t\tOnce the algorithm has been coded, it can be compiled using any standard C compiler. Each of the processes defined is translated to a software thread if the operating system supports them (other tools do not have this key characteristic, and can only compile to hardware).
\n\t\t\t\tThe entire application can then be executed and tested for correctness. Debugging and profiling the algorithm is thus straightforward, using standard tools. Then, computing intensive processes can be selected for hardware synthesis, and the included compiler will generate the appropriate VHDL or Verilog code for them, but also for the communication channels and synchronization mechanisms. The code can be generic of optimized for a growing number of commercially available platforms. Several
The versatility of their model allows for different uses of the tool. Let us consider a simple example, with 3 processes working in a dataflow scheme, as shown in Figure 2. In this case, Producer and Consumer processes undertake just the tasks of extracting the data, send them to be processed, receive the results and store them. The computing intensive part resides in the central process, which applies a given image processing algorithm. A first use of the tool would consist in generating application specific hardware for the filtering process that would be used as a primitive of a larger hardware system. The Producer and Consumer would then be “disposable”, and used just as a testbench to check, first, the correct behaviour of the filtering algorithm, and second, the filtering hardware once generated.
\n\t\t\t\tTypical CoDeveloper model
A different way of using the tool could consist in generating an embedded CPU accelerated by specific hardware. In this case, Producer and Consumer would be used during the normal operation of the system, and reside in an embedded microprocessor. The filter would work as its coprocessor, accelerating the kernel of the algorithm. CoDeveloper generates the hardware, and resolves the software-to-software and hardware-to-hardware, communication mechanisms, but also the software-to-hardware and hardware-to-software interfaces, for a number of platforms and standard buses. This is a great help for the designer that gets free of dealing with the time-consuming task of interface design and synchronization.
\n\t\t\t\tFinally, the objective can be accelerating an external CPU by means of a FPGA board. In this case, the software processes would reside on the host microprocessors, which would communicate to the application specific hardware on the board by means of a high performance bus (HyperTransport, PCI, Gigabit Ethernet, etc.). As in the previous case, software, hardware and proper interfaces between them (in the form of hardware synchronization modules and software control drivers) are automatically generated for several third party vendors.
\n\t\t\tThe results obtained in our experiments with different applications shown that AR-like algorithms can benefit from custom hardware coprocessors for accelerating execution, as well as for rapid prototyping from C-to-hardware compilers. However, to obtain any advantage, both, an algorithm profiling and a careful design are mandatory. These are the key aspects we have found to be useful:
\n\t\t\t\tThe algorithm should make an intensive use of data in different processing flows, to make up for the time spent in the transfer to/from the accelerator.
The algorithm should make use of several data flows, taking advantage of the massive bandwidth provided by the several hundred o I/O bits that FPGA devices include.
The working data set should be limited to 1-2MB, so that it may be stored in the internal FPGA memory, minimizing access to external memory.
The algorithm should use integer or fixed point arithmetic when possible, minimizing the inference of floating point units that reduce the processing speed and devour FPGA resources.
The algorithm must be profiled to identify and isolate the computational intensive processes. All parallelizing opportunities must be identified and explicitly marked for concurrent execution. Isolation of hardware processes means identifying the process boundaries that maximize concurrency and minimize data dependencies between processes, to optimize the use of onchip memory.
Maximize the data-flow working mode. Insert FIFO buffers if necessary to adjust clock speeds and/or data widths. This makes automatic pipelining easier for the tools, resulting in dramatic performance improvement.
Array partitioning and scalarizing. Array variables usually translate to typical sequential access memories in hardware, thus if the algorithm should use several data in parallel, they must be allocated in different C variables, to grant the concurrent availability of data in the same clock cycle.
Avoiding excessive nested loops. This could difficult or avoid correct pipelining of the process. Instead, try partitioning the algorithm in a greater number of flattened processes.
The nature of image processing demands the execution of intensive tasks that in many cases (as it is AR) must meet the requirement of high frame rate. This encourages the use of specific hardware in order to improve the performance of the intended applications. Indeed, the correct choice of hardware can raise dramatically the system performance. Current systems offer different benefits and limitations depending on the type of processing performed and its implementation. In this sense, general-purpose CPUs are the best alternative for sequential tasks where it is necessary to perform complex analysis of images and to run highly branched algorithms. However, in applications of 3D image processing and fast rendering scenes, the Graphics Proccesor Units (GPU) are more suitable because they have specific processing architectures designed to perform vector operations on large amounts of data. FPGAs are especially suitable for performing pre-processing tasks like colour format conversion, image filtering, convolution, and more in general, any repetitive operation that does not require highly complex algorithms, even in those cases when these algorithms are parallelizable and can benefit from the use of unconventional specific architectures. The large number of memory blocks available on FPGAs provides parallel processing support and enables very fast access to data stored in these caches. Developers can leverage the high bandwidth I/O on these devices and thus increase the speed of the functions and data rate that traverse the FPGA on GPUs or CPUs. Thanks to the versatility to develop dedicated circuits and the high degree of parallelism, FPGAs can achieve performances similar to some other hardware alternatives that run at higher frequencies of operation. These reasons explain why the implementation of many algorithms is the focus of a wide number of works since the last decade, and why from early stages of the evolution of reconfigurable hardware, several FPGA-based custom computing machines have been designed to execute image processing or computer vision algorithms (Arnold et al., 1993; Drayer at al., 1995).
\n\t\t\t\tOf high interest for AR applications is the implementation of object tracking algorithms, where different approaches have been followed. For example, the authors in (Dellaert & Tarip, 2005) present an application where a multiple camera environment is used for real time tracking with the aim of assisting visually impaired persons by providing them an auditory interface to their environment through sonification. For this purpose an octagonal board can support up to 4 CMOS cameras, an Xscale processor and a FPGA which handles the feature detection in parallel for all cameras. Another FPGA-based application for counting people using a method to detect different size heads appears in (Vicente et al., 2009). More examples of FPGA-based approaches to object tracking can be found in the literature: for colour segmentation (Garcia et al., 1996; Johnston et al., 2005); for implementing an artificial neural network for specifically hand tracking (Krips et al., 2002; Krips et al., 2003); for recognizing hand gestures (In et al., 2008); and for increasing pixel rate to improve real-time objects tracking by means of a compression sensor together with an FPGA (Takayuki et al., 2002).
\n\t\t\t\tSimilarly, the human exploration in virtual environments requires technology that can accurately measure the position and the orientation of one or several users as they move and interact in the environment. For this purpose a passive vision FPGA-based system has been proposed by Johnston et al. (Johnston et al., 2005). The aim of this system is to produce a generalised AR system in the sense that accurate estimation of a mobile user’s position relative to a set of indoor targets is accomplished in real time. FPGA-based systems are also used to develop a system for tracking multiple users in controlled environments (Tanase et al., 2008).
\n\t\t\t\tVision-based algorithms for motion estimation, optical flow, detection of features like lines or edges, etc. are also widely used in AR. Recently, several motion estimation alternatives have been proposed to be implemented on a FPGA platform. Some of them are compared in (Olivares et al., 2006). Some other good examples interest of the recent literature on this issue can be found in (Yu et al., 2004), where mobile real-time video applications with good trade-off between the quality of motion estimation and the computational complexity are presented, and (Akin et al., 2009), that focuses on the reduction of the computational complexity of a full search algorithm from 8 bits pixel resolution to one. Optical flow can also be used to detect independent moving objects in the presence of camera motion. Although many flow-computation methods are complex and currently inapplicable in real-time, different reliable FPGA-based Real-time Optical-flow approaches have been proposed in the recent years (Martin et al., 2005; Diaz et al., 2006). Furthermore, other processing image techniques like super resolution, atmospheric compensation and compressive sampling may be useful in enhancing the images and to reconstruct important aspect of the scenes. These techniques are highly complex and the use of FPGAs is encouraged to achieve the necessary acceleration. These topics are covered in detail in several articles dedicated to reconfigurable computing (Bowen et al., 2008; Bodnar et al., 2009; Ortiz et al., 2007).
\n\t\t\tComputer graphics is another field that can benefit from the flexibility of software programmable devices. This explains the increasing attention paid to FGPAs in the last years for the purpose of graphics acceleration, traditionally assigned to GPUs or graphics cards.
\n\t\t\t\tThe suitability of FPGAs for the implementation of graphic algorithms has been analysed since the mid 90s. Singh and Bellec (Singh & Bellec, 1994) introduce the notion of
Radiosity high computational cost has been improved using FPGA devices by Styles et al. (Styles & Luk., 2002). Ye and Lewis (Ye & Lewis, 1999) proposed a new architecture for a 3D computer graphic rendering system which synthesizes 3D procedural textures in an FPGA device, enhancing the visual realism of computer rendered images, while achieving high pixel rate and small hardware cost. In order to improve the efficiency of 3D geometric models represented by a triangle mesh, some mesh compression/decompression algorithms were developed. Mitra and Chiueh (Mitra & Chiueh, 2002) proposed the BFT algorithm and presented a novel FPGA-based mesh decompressor.
\n\t\t\t\t\n\t\t\t\t\tStyles and Luk. (Styles & Luk., 2000) analyzed the customization of architectures for graphics applications for both general and specific purposes, and prototyping them using FPGAs. Based on their results, the authors remark the suitability of FPGAs. In the same work an API that allows the execution of OpenGL graphics applications on their reconfigurable architecture is also presented.
\n\t\t\tComputer graphics is another field that can benefit from the flexibility of software programmable devices. This explains the increasing attention paid to FGPAs in the last years for the purpose of graphics acceleration, traditionally assigned to GPUs or graphics cards.
\n\t\t\t\tThe suitability of FPGAs for the implementation of graphic algorithms has been analysed since the mid 90s. Singh and Bellec (Singh & Bellec, 1994) introduce the notion of
Radiosity high computational cost has been improved using FPGA devices by Styles et al. (Styles at al., 2002). Ye and Lewis (Ye & Lewis, 1999) proposed a new architecture for a 3D computer graphic rendering system which synthesizes 3D procedural textures in an FPGA device, enhancing the visual realism of computer rendered images, while achieving high pixel rate and small hardware cost. In order to improve the efficiency of 3D geometric models represented by a triangle mesh, some mesh compression/decompression algorithms were developed. Mitra and Chiueh (Mitra & Chiueh, 2002) proposed an algorithm and presented a novel FPGA-based mesh decompressor. Styles et al. (Styles at al., 2000) analyzed the customization of architectures for graphics applications for both general and specific purposes, and prototyping them using FPGAs. Based on their results, the authors remark the suitability of FPGAs. In the same work an API that allows the execution of OpenGL graphics applications on their reconfigurable architecture is also presented.
\n\t\t\tFPGA technology has appeared to be also very useful for communication systems. Two important factors encourage its expansion in this field: the falling prices of the devices and the inclusion of DSP capabilities. Typical communication problems such as data formatting, serial to parallel conversion, timing and synchronization can be faced naturally in a FGPA device thanks to its specific features. Furthermore, FPGAs are convenient for the development of the necessary glue logic for the interconnection of processors, modems, receivers, etc. Several examples can be found in the literature of the field. In (Ligocki et al, 2004) the authors describe the prototype development of a flexible communication system based on a FPGA. The main focus of this work is on software concerns, considering that FPGA technologies are the core of the project. Other authors exploited the spatial/parallel computation style of FPGAs for wireless communications. Due to the computational complexity of WLAN (Wireless local area network), and taking into account the capabilities of modern microprocessors, an implementation based exclusively on microprocessors is not convenient, requiring a large number of components. Parallel computation allows improving the efficiency of the implementation of the discrete components, and makes it possible to accelerate some complex parts of WLANs (Masselos & Voros, 2007).
\n\t\t\t\tOf special interest results the benefits of FPGAs for embedded software radio devices. In (Hosking, 2008), it is shown how its inherent flexibility makes of FGPA devices an excellent choice for coping with the increasing diverse array of commercial, industrial, and military electronic systems. Additionally, the large number of available IP cores offer optimized algorithms, interfaces and protocols which can shorten significantly the time-to-market.
\n\t\t\tIt is a common understanding that the concept of wearable computing comes from the tools developed to intensify the experience of seeing, what led to augmented reality. However, it can be noticed that in the literature different authors understand the concept of wearable systems in very different manners. Some authors claim that a system is wearable as long as it can be transported by a human. According to this, some authors present wearable solutions based on small variations of desktop applications in a back-up with a laptop (Feiner et al, 1997; Hoellerer et al, 1999). However, let us focus in this chapter on only wearable devices which do not constraint the mobility of the user. For this purpose, the use of FPGAs results of the highest interest. In wearable systems, the problem of combining simultaneously high performance and low power consumption requirements in small dimensions can be overcome with FPGAs. Contrary to ASICs (application specific integrated circuit), reconfigurable logic offers more flexibility to adapt dynamically the processes and the possibility of integrating different processing units in only one device. This way, it is possible to reduce the number of chips in a system, which can be an important advantage.
\n\t\t\t\tAn interesting study on the improvements regarding energy saving when implementing critical software processes on reconfigurable logic can be found in (Stitt et al, 2002). The LART board, presented in (Bakker et al, 2001) combines a low-power embedded StrongARM CPU with an FPGA device, which offers a better power/MIPS ratio, pursuing power consumption reduction. The FPGA is used for dedicated data processing and for interconnecting several LARTs working in parallel. In a step forward, the authors in (Enzler et al, 2001) analyze the applications of dynamically reconfigurable processors in handheld and wearable computing. They consider a novel benchmark which includes applications from multimedia, cryptography and communications. Based on that work, the authors of (Plessl et al, 2003) presented the concept of an autonomous wearable unit with reconfigurable modules (WURM), which constitutes the basic node of a body area computing system. The WURM hardware architecture includes reconfigurable hardware and a CPU. In the prototype, the implementation is done on only one FPGA, including the CPU as a soft core.
\n\t\t\t\tFinally, let us remark the importance of networking for wearable computing. Indeed, the constraints in power consumption, size and weight of wearable computers increase the need for network capabilities to communicate with external units. A study about the interest of FPGA approaches for network on chip implementations across various applications and network loads is presented in (Schelle & Grunwald, 2008). Recently, several authors have followed FPGA-based approaches in their solutions. In (Munteanu & Williamson, 2005), an FPGA is exploited to provide consistent throughput performance to carry out IP packet compression for a network processor. (Wee et al, 2005) presents a network architecture that processes in parallel cipher block chaining capable 3DES cores by using about the 10% of the resources of an FPGA Xilinx Virtex II 1000-4. Within the frame of the European Diadem Firewall Project, IBM suggests the use of standalone FPGA-based firewalls in order to achieve an accelerated network architecture (Thomas, 2006).
\n\t\t\tWe have proposed a platform for developing fully FPGA-based embedded systems aimed for image and video processing applications. It is a hardware/software system created for speeding up and facilitating the development of embedded applications. As the survey of works in previous sections highlights, FPGA devices are very suitable for the implementation of the processing tasks involved in AR. The adoption of an FPGA-based approach allows executing different tasks and algorithms in parallel, which ensures the best performance and the optimum power consumption. The platform acquires video in standard analogue formats, digitizes, and stores it in external memory devices. In order to confer versatility to the embedded system, the platform includes as a key component an interface which allows for user interaction. This interface makes it possible to display text and, by means of hand pose recognition or voice recognition, to choose options and configure parameters. Thanks to it, the user can customize the functionality of the hardware at run-time.
\n\t\t\tVideo is a primary input to AR systems and can be typically used to develop video see-through systems, to execute vision-based tracking algorithms or as input to a user interface. In order to process video we have developed a frame grabber which accepts standard analogue video signal, converts it into digital and stores it in memory.
\n\t\t\t\tThe frame grabber is based on the SAA7113 video input processor, from Philips Semiconductors, which is able to decode PAL, SECAM and NTSC from different sources (CVBS, S-video) into ITU-R BT 601. It is configured and controlled through I2C bus, so an I2C controller module must be included to properly manage the SAA7113. The SAA7113 presents video data at an 8 bit digital video port output, with 720 active pixels per line in YUV 4:2:2 format and a number of lines in a frame depending on the video standard (PAL, NTSC). The 4:2:2 output format implies that there is a luminance Y value for each pixel, but only a chrominance pair UV for two pixels. The YUV colorspace is used by the PAL and NTSC colour video standards. However, RGB is the most prevalent choice for computer graphics. Therefore, we have included a converter from YUV to RGB in the design. It just implements the corresponding linear equations, which can be found, e.g., in (Jack, 2005).
\n\t\t\t\tThe image from the SAA7113 must be stored in a frame buffer. In our platform it is made of external asynchronous SRAM memory devices. A memory interface for generating the memory control signals and read /write operations was implemented on the FPGA.
\n\t\t\t\tColour data stored in the memory is in YUV format since it optimizes the space and the access to memory. While in RGB format a pixel is defined with 24 bits, the YUV format from the video codec uses 32 bits to define two pixels, with exactly the same colour information in both RGB and YUV colorspaces. As physically each frame buffer consists of a 32 bit 256 KB SRAM, it is more efficient to store colour information in YUV colorspace, since it is possible to store two pixels in just one address, and so halving the number of access to the memory.
\n\t\t\tUnlike PC-based solutions, where visualization of text information is completely usual, this is not so natural in hardware-based solutions. Most of the present FPGA-based embedded systems do not offer an interface to the user. Sometimes they just consider a UART to connect with a computer and transfer some information. However, the use of a PC simply for running the software that manages the communications and the interface is a poor, very low efficient solution, even unfeasible in embedded systems. With the aim of overcoming this drawback of FPGA-based embedded systems, we have designed a hardware core which facilitates the addition of a user interface to FPGA-based systems. The core is based on MicroBlaze, a standard 32 bit RISC Harvard-style soft processor developed for Xilinx FPGA devices. This gives flexibility and versatility, and ensures fast re-design of the hardware architecture for enhanced or new applications. The core is made up of hardware components and software functions in different levels. Thanks to the core it is possible to present text information in a VGA monitor to the user, who can navigate through menus, select options and configure parameters by means of a pointer device. So, it provides the flexibility of adapting the systems to the user requirements or preferences. Next, its basic modules are described.
\n\t\t\t\t\n\t\t\t\t\t
In order to display text, all the bitmaps associated to the font characters were previously defined and stored in a ROM memory using FPGA internal logic resources. The ROM works as a MicroBlaze peripheral, using a dedicated Fast Simplex Link (FSL) channel.
\n\t\t\t\tTo display text, we have considered a text screen, which manages the visualization of the strings. In the default mode, it is a 640×480 array, whose elements correspond to the pixels in the VGA output. A 64 colours palette has been considered, which implies 6 bits for each pixel. Due to its size, the array is stored in external SRAM.
\n\t\t\t\tThe text screen is defined as a peripheral and connected to the MicroBlaze microprocessor by means of the On-chip Peripheral Bus (OPB). The hardware of this peripheral includes the SRAM memory interface to control write and read operations and the logic to interpret the data from MicroBlaze into address and data buses values. It carries out two different tasks:
\n\t\t\t\tit receives data from MicroBlaze, and manages the write operations in the SRAM memory just when a modification in the text information displayed is done.
it reads data from the memory to show the text screen in the VGA monitor. These data are sent to the VGA Generator module. This process is independent on MicroBlaze.
In order to create the interface presented to the user, the function mb_OutTextXY has been prototyped to be instantiated in the software application running in MicroBlaze. It is similar to the equivalent standard C function, and it allows to define a text string and to specify its colour and position in the screen. When the mb_OutTextXY function is executed, the writing instruction of the text screen peripheral is called to write in the SRAM the colour values of the pixels which correspond to each element of the string, according to its position and its colour.
\n\t\t\t\tOnce the strings are stored in the text screen, the basic user application waits for an interrupt from pointer device. When it happens, an interrupt handler classifies the interrupt and reads the coordinates of the pointer position. Since the position of each text string is known, it is possible to determine in the software application which one has been selected by the user, and then to reply with the desired actions.
\n\t\t\t\t\n\t\t\t\t\t
A pointing device is required to interact within the user interface. With this aim, we have proposed a hand-based interface designed for mobile applications. It detects the user hand with a pointing gesture in images from a camera, and it returns the position in the image where the tip of the index finger is pointing at. In an augmented reality application the camera will be placed on a head mounted display worn by the user. A similar system is proposed in (Piekarski et al, 2004), but our approach is based on skin colour, without the need of glove or coloured marks. Our hand-based interface is aimed for performing pointing and command selection in the platform for developing FPGA-based embedded video processing systems herein described.
\n\t\t\t\tVision-based algorithms have been used to build the hand and the pointing gesture recognizers. The image from the camera, once acquired, digitalized and stored by the previously described frame grabber, is segmented using an ad-hoc human skin colour classifier. Human skin colour has proven to be useful in applications related to face and hands detection and tracking. With this colour skin approach we try to generalize the use by eliminating external accessories, what reduces costs. The skin colour classifier is made of sub-classifiers, each one defined as a rule-based algorithm built from histograms in a colourspace. The rules in each colourspace define a closed region in the corresponding histogram: a pixel of an image is classified as skin if its colour components values satisfy the constraints established by the rules. Classifiers in the YIQ, YUV and YCbCr colorspaces have been considered, so three sub-classifiers have been implemented. To generate a unique output image, their outputs are merged using logical functions. The use of different colorspaces is aimed at achieving invariance to skin tones and lighting conditions. Further details can be found in (Toledo et al, 2006).
\n\t\t\t\tOnce the image has been segmented the next processing task is to look for the pointing gesture. The solution adopted consists of convoluting the binary image from the skin classifier with three different templates: one representing the forefinger, other the thumb and the third the palm (Toledo et al, 2007). This modularity makes easier the addition of new functionality to the system through the recognition of more gestures. Due to the size of the hand and the templates, an optimized solution for the FPGA-based implementation of large convolution modules has been specifically developed. It can convolve binary images with a three-value template in one clock cycle independently of the template size. It is based on distributed arithmetic and has been designed using specific resources available in Xilinx FPGAs. The maximum size of the template depends on the FPGA device. In this application images are convoluted with 70×70 templates. Each convolution module sends to the MicroBlaze soft processor its maximum value and its coordinates on the image. A software algorithm running on MicroBlaze decides that a hand with the wanted gesture is present when the maximum of each convolution reaches a threshold and their relative positions satisfy some constraints derived from training data. Then, the algorithm returns the position of the forefinger. Otherwise, it reports that no pointing hand is detected.
\n\t\t\t\tThe software application on MicroBlaze also includes an algorithm for dynamically adapting the skin classification and the parameters for hand recognition. Taking into account the number of pixels classified as skin in the image, the maximum value and the coordinates of each convolution and the detection or not of the pointing hand pose, it tunes each skin classifier and the merging of their binary output images in order to achieve the optimum classification rates, and it also tunes the values of the different constraints to their right values in order to find the desired hand posture. The FPGA implementation of these tasks allows taking advantage of parallelism in each processing stage at different levels. For example, the three classifiers for skin recognition are executed at the same time on an input pixel. Since the constraints of a classifier are all evaluated at the same time, the time required to classify a pixel is just the maximum delay associated to a constraint, three clock cycles in our case. Besides, the three convolutions that look for the hand position are performed in parallel, and the operations involved in each convolution are all executed at the same time in only one clock cycle. Meanwhile, the software application in MicroBlaze is using the information extracted by the hardware modules as input parameters to the algorithm which estimates the presence and position of the hand. Thanks to exploiting the parallelism inherent to FPGA devices, the hand detection algorithm can process 640×480 pixel images at more than 190 frames per second with a latency of one frame. It also makes it feasible that new processing cores can be added to the system with small performance penalty.
\n\t\t\t\tIn addition to the hand-based interface, a controller for a generic PS/2 mouse has been implemented and added to the MicroBlaze system as an OPB peripheral.
\n\t\t\tThe platform can generate signals for displaying video on monitors and analog screens. The generation of the synchronization and RGB signals for a VGA monitor is carried out by the VGA generator module, which can be configured to generate different resolutions. The platform also includes the SAA7121, an integrated circuit from Philips which encodes digital video into composite and S-video signals. The video generator module also deals with the mixing of the video from the different sources included in the platform.
\n\t\t\tWe have validated the usefulness of the described platform in an application for people affected by a visual disorder known as tunnel vision. It consists in the loss of the peripheral vision, while retaining clear and high resolution central vision. As shown in Figure 3, it is like looking through a keyhole or a ring in the mid-periphery. Tunnel vision is associated to several eyes diseases, mainly glaucoma and retinitis pigmentosa, and reduces considerably the patient’s ability to localize objects, which inevitably affects the patient’s relationship with people and the environment.
\n\t\t\tSimulation of patient affected by tunnel vision view. A residual 10º field of view has been considered to simulate the tunnel vision effect. The severe reduction of the visual field (right) can be observed comparing with the normal vision (left).
To aid affected people, it is necessary to increase the patient’s field of view without reducing the usefulness of the high resolution central vision. With this aim, (Vargas-Martín & Peli, 2001) proposed an augmented view system where the contour information obtained from the image of a camera is superimposed on the user’s own view. In their work, contours are generated by an edge detection algorithm performed by a four-pixel neighborhood gradient filter and a threshold function, running on a laptop PC (Vargas-Martín & Peli, 2002). They draw the conclusion that, although patients consider the system useful for navigating and obstacle avoiding, a specifically designed system to perform image processing and increase frame rate is necessary. Obviously, an effective improvement of the user’s environment perception requires real time processing. To achieve it, we have used our FPGA-based hardware platform, which ensures the video frame rate and the low latency that the mobility of the application required.
\n\t\t\tSimulation of patient’s view through the HMD for outdoor and indoor environments. A residual 10º field of view has been considered to simulate the tunnel vision effect.
Overall system schematic showing the main modules and the output view presented to the user.
In our system, the image acquired with the frame grabber is processed to extract contour information and it is used to enhance the user’s perception of the environment by the superimposition on his own view of the entourage seen with a see-through head mounted display. To carry out the required processing we proposed the use of a Cellular Neural Network (CNN), which can be tuned to produce customized results and allows increasing the versatility of the system through the possibility of using different templates. The difficulties that rise when designing digital hardware implementation of CNNs are addressed in (Martínez et al, 2007; Martínez et al, 2008), where a novel approach is also proposed. It has been later optimized in (Martínez et al, 2009). After processed, the image from the camera must be properly zoomed out in order to be shown to the user in his residual central vision. A digital zooming algorithm has been included in the design with this purpose. It has been designed to minimize the number of access to the external memory where the original input data are stored.
\n\t\t\tThe image resulting from the processing with the CNN, suitably minified, is sent to the VGA output available in the hardware platform. Figure 4 shows some examples of the system output.
\n\t\t\tDue to the special characteristics of the target application, a vision-based interface such as the described in the previous section is not a suitable approach for interacting with this system. Instead, we have developed a voice recognition system which, as the whole of the herein described system, is implemented in FPGA. Since the aid device is conceived as a personal system, an easy and high-reliable speaker dependent recognition algorithm has been designed. In a simple and fast initial step, the user records a customized set of keywords. They are stored in external non-volatile flash memory, so the user only has to do it for the first time use. Later, in operating mode, the algorithm detects when the user says a word and convolves it with all the previously recorded ones. The latency response depends on the number of key words, but it is typically in the order of the milliseconds, fast enough to not appreciating any significant delay.
\n\t\t\tThis user interface makes it feasible to adapt the functionality to the user preferences through, for example, basic modifications in the CNN processing, the minification factor or the colour and intensity of the contour information superimposed in his view.
\n\t\t\tThe great amount of resources available in FPGA devices and their inherent parallelism make it possible to fulfill the requirements of the application without compromising the performance in speed, size and power consumption.
\n\t\t\tA simplified diagram of the whole system is shown in Figure 5. The current prototype has been built with boards from Avnet, which populates Xilinx devices and the additional integrated circuits mentioned. The prototype also uses a Sony Glasstron PLMS700E as head mounted display to superimpose the video output on the user view. At the present moment, the system is under test and validation by visually impaired people, offering very successful initial results and improving the patients’ ability to localize objects, orientate and navigate. Once passed the tests, a commercial device will be manufactured and packed into a small shoulder bag or belt bag.
\n\t\t