Organoid Intelligence: Bridging Artificial Intelligence for Biological Computing and Neurological Insights

Abstract

Brain organoid implications have opened vast avenues in the realm of interdisciplinary research, particularly in the growing field of organoid intelligence (OI). A brain organoid is a three-dimensional (3D), lab-grown structure that mimics certain aspects of the human brain organization and function. The integration of organoid technology with computational methods to enhance the understanding of organoid behavior and to predict their responses to various stimuli is known as OI. The ability of brain organoids to adapt and memorize, is a key area of exploration. OI encapsulates the confluence of breakthroughs in stem cell technology, bioengineering, and artificial intelligence (AI). This chapter delves deep into the myriad potentials of OI, encompassing an enhanced understanding of human cognitive functions, and achieving significant biological computational proficiencies. Such advancements stand to offer a unique complementarity to conventional computing methods. The implications of brain organoids in the OI sphere signify a transformative stride towards a more intricate grasp of the human brain and its multifaceted intricacies. The intersection of biology and machine learning is a rapidly evolving field that is reshaping our understanding of life and health. This convergence is driving advancements in numerous areas, including genomics, drug discovery, personalized medicine, and synthetic biology.

1. Introduction

The human brain is a remarkably intricate and complex organ that remains one of the greatest mysteries in scientific exploration. Despite significant progress in neurobiology, many aspects of brain development and disorders remain elusive. Brain organoids represent a revolutionary advancement in neuroscience research that serves as an unparalleled tool to study the development of the human brain and provides insights into the fundamental processes underlying neurogenesis, cell migration, and axon guidance [1]. These three-dimensional (3D), self-organized mini-brains hold the potential to transform our understanding of the brain’s complexity, providing insights into early neurodevelopment, disease modeling, and drug testing [2]. Traditional two-dimensional (2D) cell cultures and animal models have limitations in fully capturing the complexity of human brain diseases. In contrast, brain organoids provide a more accurate model of the human brain’s cellular and molecular environment (Table 1) [3].

Characteristic	2D cell culture	3D brain organoid models	Organoid intelligence
Dimensionality	Flat, single layer of cells	3D structure mimicking tissue organization	3D structure integrated with AI components
Cell interactions	Limited cell-cell and cell-matrix interactions	Complex cell-cell and cell-matrix interactions More representative of in vivo conditions	Complex interactions with added AI-driven feedback and analysis
Complexity	Simple, often a single cell type	Multiple cell types, mimics tissue architecture	Combines biological complexity with computational analysis
Functionality	Basic cellular functions	Advanced functions like tissue-specific activities, formation of neural networks	Advanced functions with AI-driven enhancements, data analysis, and predictions
Applications	Basic research Drug screening	Disease modeling Drug testing Developmental biology	Advanced research Real-time analysis Predictive modeling
Limitations	Does not mimic in vivo conditions well Lacks tissue context	Still does not fully represent whole organs Limited lifespan	Technological challenges Ethical considerations
Cost	Relatively low	Higher due to complexity	Highest due to integration of advanced technologies
Ethical concerns	Few, mostly related to source of cells	Concerns about creating “conscious” tissue, source of cells	Amplified concerns about consciousness, AI ethics, and potential misuse

Table 1.

Comparison of the characteristics of 2D cell culture, 3D brain organoid models, and organoid intelligence.

In general, organoids are 3D structures grown from stem cells that mimic the architecture and functionality of real organs. They have been used to model various human tissues including the brain, liver, kidney, and gut. Organoids provide a more accurate representation of in vivo conditions compared to traditional 2D cell cultures, making them valuable tools for studying organ development, disease progression, and drug responses [4]. The advent of human stem cell-derived brain organoids has revolutionized the field of neuroscience, offering a promising avenue to replicate critical molecular and cellular aspects of learning, memory, and possibly aspects of cognition in vitro.

The term “organoid intelligence” (OI) has been coined to describe this emerging field, aiming to expand the definition of biocomputing towards brain-directed OI computing. The concept of organoid intelligence (OI) aims to harness the innate biological capabilities of brain organoids for biocomputing and synthetic intelligence by interfacing them with computer technology. The ability of brain organoids to adapt and memorize, is a key area of exploration. By leveraging the organoid’s brain-like functionality, we can harness its capacity to process complex inputs, study the physiology of learning, and generate responses to control peripheral output devices (Table 1). This exploration of OI is a multidisciplinary endeavor that necessitates the engagement of both the scientific community and the public [5].

Biological computing refers to harnessing the power of biologically derived molecules to perform storage, retrieval, and processing. Human neurons are capable of incredible information processing to the microscopic size of their densely packed computational units with trillions of hyperdense synapses and high metabolic efficiency [6]. Moreover, these cells have the remarkable ability to regenerate and repair themselves continuously, similar to a computer system that operates without the need for external maintenance. In a study, an 8000-pound supercomputer claims to exceed the computational speed of the human adult brain which can only do so by consuming a million times more energy [2]. However, with new advancements in organoids, future machine learning could harness the computational efficiency and power of the brain. Biocomputers also have the potential to be more energy-efficient than traditional computers. Biological systems are capable of performing complex tasks with minimal energy consumption. In this aspect, the human brain can only perform complex cognitive tasks with a power consumption of around 20 watts, far less than that of a typical computer [7].

OI research aims to explore how a 3D brain cell culture can be made more computer-like. The many possible applications of this work include a new generation of biological and hybrid (biological-electronic) computing technologies. OI, thus, advances in the understanding of physiology of cognition, learning, and memory, and the pathophysiological effects of developmental and degenerative diseases, intoxication, and infection. OI also has the potential to unlock new neuromimetic artificial intelligence (AI) algorithms and aid the development of new brain-computer-interface technology. In this chapter, we have discussed the establishment and culture methods of brain organoids with clinical applications emphasizing their potential role in biocomputing and OI. For this, we have discussed current progress, challenges, and future perspectives. Further, we delve into the brain-computer interface (BCI), aiming to comprehend the nuances of brain connectivity, bridging the gap between brain organoids and computers, the intriguing aspect of training brain organoids with artificial intelligence (AI). This chapter offers a comprehensive insight into the convergence of biology and technology, paving the way for groundbreaking innovations in understanding and harnessing the power of the human brain. It also aims to explore the significance of brain organoids in revolutionizing neuroscience research, enabling a deeper comprehension of brain development and providing new avenues for understanding and treating brain disorders. The potential of OI to revolutionize computing, neurological research, and drug development has been also discussed extensively.

2. Brain organoids: a brief overview

Brain organoids are three-dimensional (3D) structures that serve as in vitro models to mimic the early stages of human brain development. These organoids are derived from pluripotent stem cells (PSCs), which can be obtained from different sources, such as induced pluripotent stem cells (iPSCs) reprogrammed from adult cells, or embryonic stem cells (ESCs) derived from early-stage embryos [8]. PSCs have the potential to develop into any cell type in the body, including the various cell types found in the brain (neurons and glial cells). The process of differentiation from a PSC into a specific type of cell is guided by a complex interplay of genetic and environmental factors. Various transcription factors are also used to activate or deactivate certain genes, or the cells are exposed to specific signaling molecules to mimic the conditions in a particular tissue [9]. By carefully controlling these factors, stem cells can be coaxed to differentiate into a wide variety of cell types, from neurons to heart cells. One of the most significant advantages of brain organoids is their ability to model the development of the human brain. Traditional animal models, such as mice, have significantly different brain structures and developmental processes, making them less than ideal for studying human neurodevelopment. In contrast, brain organoids can recapitulate key aspects of human brain development, including cell differentiation, migration, and interaction. This makes them a valuable tool for studying neurodevelopmental disorders, such as autism and schizophrenia [10].

2.1 Process of brain organoid formation

The process of generating brain organoids involves a technique called cerebral organoid differentiation that mimics the natural processes occurring during early brain development in the human embryo [11]. Firstly, PSCs are isolated from donated human tissues (for iPSCs) or derived from early-stage embryos (for ESCs). Once PSCs are obtained, they are directed to differentiate into neural cells using specialized growth factors such as fibroblast growth factors (FGFs), epidermal growth factor (EGF), epidermal growth factor (EGF) and, insulin-like growth factor (IGF), and signaling molecules such as bone morphogenetic proteins (BMPs), noggin, Wnt proteins and neurotrophins, to coax the cells into adopting a neural fate. This process is akin to the early stages of embryonic development when neural tissue begins to form [12]. Through the action of specific signaling pathways, the differentiating PSCs organize themselves into a structure called neuroectoderm. Neuroectoderm is a transient embryonic tissue that gives rise to the central nervous system, including the brain and spinal cord. Within the neuroectoderm, neural stem cells are formed, which are self-renewing and capable of generating various neural cell types [13]. After the formation of neural stem cells, they begin to self-organize and differentiate further. This process is driven by complex genetic and biochemical interactions which result in the formation of distinct brain regions that are akin to early brain development in embryos. These interactions include gene expression regulation, extracellular matrix (ECM) interactions, cell fate determination and differentiation of neural progenitor cells into specific cell types, formation of specific brain structures, chemical signaling between neurons and other cells, influencing synapse formation and neuronal function, mediate cell–cell interactions and guide the migration of cells to appropriate locations and contributing to the formation of specific brain regions. These brain regions include the structures resembling cerebral cortex, hippocampus, and other brain regions, each with specific neuronal types and functions [14]. As the self-organization and morphogenesis continue, the brain organoids recapitulate critical events observed during early human brain development, such as neurogenesis, gliogenesis, and the establishment of neural circuits (Figure 1). The resulting brain organoids display a certain level of complexity and functionality resembling early stages of the human brain. To maintain the vitality and growth of these organoids, it’s essential to passage them periodically. Passaging involves carefully breaking down the larger organoids into smaller pieces, ensuring the core remains undamaged. These smaller fragments are then reseeded in fresh culture media, providing them with the necessary nutrients and space for further growth. This process not only ensures the health of the organoids but also allows for the expansion of the culture, paving the way for multiple experiments and prolonged study durations.

Figure 1.

Representation of process of generating brain organoids. PSCs isolated from iPSCs or ESCs, differentiate into an embryoid body. The differentiating PSCs organize themselves into a structure called neuroectoderm that subsequently matures into an organoid.

The generation of brain organoids through cerebral organoid differentiation is a fascinating process that enables researchers to create 3D models of early human brain development in a laboratory setting. Many studies have established 3D culture system that allowed for the self-organization of neural progenitor cells into structures resembling different brain regions. They showed that these organoids could recapitulate key aspects of human brain development and be used as models to study microcephaly, a neurodevelopmental disorder characterized by reduced brain size [15, 16]. While some studies have generated neuromuscular organoids containing spinal cord and muscle tissue. The researchers showcased the potential for generating complex multi-tissue organoids using cerebral organoid differentiation techniques. In an instance, a study developed a model of human neuromuscular system, which comprised of both the nervous system and the muscles. To develop this model, the researchers used PSCs to generate 3D organoids that contained both neural and muscle tissue, similar to the human trunk’s structure (from neck to abdomen). In these neuromuscular organoids, the neurons formed functional neuromuscular junctions with the muscle cells, closely mimicking the connections seen in the human body [17].

In contrast, there are reports highlighting the importance of cerebral organoids as a model system to study various infectious diseases. A study conducted by Cugola et al. focused on modeling Zika virus infection using cerebral organoids. The authors used human iPSCs to create cerebral organoids to model the early stages of brain development. They infected brain organoids with the Zika virus and observed the virus’s effects on neural stem cells and brain development. The Zika virus infection resulted in cell death and disruption of organoid growth, closely resembling the microcephaly observed in human fetuses infected with the virus in utero. They also found that Zika virus infection altered the expression of genes involved in cell cycle regulation, DNA repair, and cell death, contributing to the observed detrimental effects on brain development [18]. These organoids offered a more accurate representation of human brain tissue than traditional 2D cultures, making them an ideal tool to investigate the impact of the Zika virus on brain development.

2.2 Application of brain organoids

The applications of brain organoids span across various fields, including disease modeling, drug testing, personalized medicine, and understanding human brain development (Figure 2).

Figure 2.

Schematic flow chart illustrating the diverse applications and potential of organoid intelligence in the realm of neuroscience and medical research.

2.2.1 Disease modeling

The use of brain organoids in disease modeling has been a significant advancement in neuroscience. Brain organoids can be used to model various neurological and psychiatric diseases which are thought to arise during brain development. The advancements in methods for generating more complex brain organoids include the use of vascularized and mixed lineage tissue from PSCs. Moreover, synthetic biomaterials and microfluidic technology are boosting brain organoid development. The applications of brain organoids have also been seen in studying pre-term birth associated brain dysfunction, viral infections mediated neuroinflammation, neurodevelopmental, and neurodegenerative diseases [2]. Brain organoids can be used to model these early developmental processes, providing insights into how these disorders might originate. For example, researchers have used brain organoids to study the effects of genetic mutations associated with autism, revealing how these mutations can disrupt normal brain development. In another instance, brain organoids derived from patients with Alzheimer’s disease can recapitulate key features of the disease, including the accumulation of amyloid-beta plaques and neurofibrillary tangles [19].

Similarly, brain organoids can be used to study disorders like autism, schizophrenia, and epilepsy, providing insights into disease mechanisms and potential therapeutic targets [20]. A study by Robles et al. reported that the in vitro models of the human brain can facilitate efficient development and clinical translation of therapeutic treatments of mental disorders. In their study, they addressed the limitations of current models of neural circuitry, which include animal models and post-mortem brain tissue. These models have limitations in assessing the functional alterations in short-range and long-range network connectivity between brain regions implicated in many mental disorders, such as schizophrenia and autism spectrum disorders.

To address these limitations, the researchers developed an in vitro model of the human brain that models the in vivo cerebral tract environment. They combined microfabrication and stem cell differentiation techniques to develop an in vitro cerebral tract model that anchors human-induced PSC-derived cerebral organoids and provides a scaffold to promote the formation of a functional connecting neuronal tract. Morphological and functional analyses revealed the expression of key neuronal markers as well as functional activity and signal propagation along with cerebral tracts connecting cerebral organoid pairs. The reported in vitro models enabled the investigation of critical neural circuitry involved in neurodevelopmental processes and had the potential to help devise personalized and targeted therapeutic strategies [21].

2.2.2 Drug testing

Traditional drug testing methods often rely on 2D cell cultures and animal models. Although these methods have provided valuable insights, they have limitations. 2D cell cultures lack the complex cellular interactions and architecture of the human brain. While animal models, although are more complex and do not fully replicate human physiology. In contrast, brain organoids offer a more accurate and physiologically relevant model of the human brain, making them an attractive platform for drug testing [22]. Researchers are now able to test the effects of potential drugs on these organoids by recreating the disease conditions and providing insights into the drug’s efficacy and potential side-effects [23]. Brain organoids can also be used for personalized medicine. Many studies have created patient-specific organoids that carry the same genetic information as the patient [24, 25]. This allowed the testing of drugs on these patient-specific organoids and potentially predicted the patient’s response to the drug. This approach has led to more personalized treatment strategies that improve patient outcomes. Rajan and colleagues illustrated an integrative genomic, in vitro, and in vivo functional treatment paradigm for high-grade glioma (GBM) using patient-derived organoids (PDOs). Their study suggested that rapidly implemented individualized drug response prediction models provide actionable information for the physician to combat recurrences or treatment resistances in GBM [24]. Similarly, another study suggested that veteran-derived human cerebral organoids not only can be used as an innovative human model to uncover the cellular responses to Gulf War toxicants but can also serve as a platform for developing personalized medicine approaches for the veterans [25]. Furthermore, brain organoids are used for high-throughput drug screening. By generating large numbers of organoids, researchers have tested the effects of many different drugs or drug combinations simultaneously. This accelerated the drug discovery process, potentially leading to the development of new treatments more quickly.

2.2.3 Understanding brain development

Brain organoids can be used to study the processes of human brain development and to understand how disruptions in these processes can lead to neurological and psychiatric disorders [18, 19]. Similarly, organoids derived from patients with autism have been shown to exhibit abnormal patterns of neuronal activity, providing insights into the underlying mechanisms of the disorder [26]. Human brain development is a complex process that involves the coordinated proliferation, migration, and differentiation of neural progenitor cells into a variety of specialized cell types, including neurons and glial cells. This process is tightly regulated by a network of genes and signaling pathways. Disruptions in these processes can lead to a range of neurological and psychiatric disorders, including autism, schizophrenia, and epilepsy [27]. Brain organoids provide a unique platform to study these processes in a controlled laboratory setting. One of the key advantages of using brain organoids to study brain development is their ability to recapitulate the 3D architecture of the brain. Unlike traditional 2D cell cultures, organoids can form complex structures that resemble those found in the developing human brain. This includes the formation of distinct brain regions, such as the cerebral cortex, and the development of intricate networks of neurons that are capable of transmitting electrical signals [26].

2.2.4 Neurodegeneration

Neurodegenerative diseases, including Alzheimer’s, Parkinson’s, Huntington’s, and amyotrophic lateral sclerosis, are characterized by the progressive loss of structure or function of neurons. These diseases pose significant challenges to healthcare systems due to their chronic nature, lack of effective treatments, and the aging global population [28]. Brain organoids have emerged as a promising tool in the study of neurodegenerative diseases. They can be used to study the progression of these diseases and to test potential treatments. This is particularly important as these diseases often have a complex etiology involving both genetic and environmental factors. Brain organoids can be genetically manipulated to carry disease-specific mutations, allowing researchers to study the effects of these mutations on brain development and function [29]. Besides, various studies have also demonstrated the use of microglia-containing human brain organoids for studying brain development and pathology. They provide a new avenue to model brain development and pathology [30]. Microglia also play critical roles in neural development, synaptic formation and plasticity, and neural network maturation. Studies conducted over the past decades have implicated microglia dysfunction in multiple brain disorders, ranging from psychiatric disorders to neurodegenerative diseases [31].

3. Biocomputers: the intersection of biology and computing

Biocomputing involves the development and application of data storage, retrieval, and processing systems that use biological components. The concept is rooted in the idea that biological systems are inherently computational and can be harnessed to perform computing tasks. Biocomputers leverage the capabilities of biological molecules, particularly DNA and proteins, to perform computational calculations [32]. DNA has been used to create logic gates, the fundamental building blocks of digital circuits. These DNA-based logic gates can perform simple binary operations, such as AND, OR, and NOT functions, similar to their silicon-based counterparts in traditional computers [33]. One of the most significant advantages of biocomputers is their potential for parallel processing. Unlike traditional computers, which process information sequentially, biocomputers can perform many operations simultaneously. This is because biological reactions often occur in parallel, with many molecules interacting at the same time. This parallelism allows biocomputers to solve complex problems more efficiently than traditional computers [34]. Another advantage of biocomputers is their potential for miniaturization. Biological molecules are incredibly small, allowing for the development of ultra-compact computing systems. This is particularly relevant in the context of medical applications, where biocomputers could be used to develop smart drugs that can perform computations inside the body to diagnose or treat diseases [35].

3.1 History of biocomputers

The concept of biocomputing or using biological systems to perform computational tasks, has its roots in the fundamental understanding of life processes. The history of biocomputers is a fascinating journey that intertwines the fields of biology, computer science, and engineering (Figure 3). The idea of biological computation can be traced back to the mid-twentieth century when researchers began to understand the genetic code and the process of protein synthesis. The realization that DNA carries information and the cell translates this information into proteins, a process akin to reading and executing a program, laid the groundwork for the concept of biological computation [36]. However, the field of biocomputing truly began to take shape in the 1990s, with the advent of DNA computing. The pioneer of this field was Leonard Adleman, a computer scientist from the University of Southern California [37]. In 1994, Adleman published a groundbreaking paper in which he demonstrated that DNA could be used to solve a well-known mathematical problem, the seven-point Hamiltonian path problem, also known as the “traveling salesman” problem. Adleman’s work showed that DNA molecules, with their ability to store information and undergo specific reactions, could be used to perform computations [38]. Following Adleman’s work, the field of DNA computing rapidly expanded. Researchers developed methods to use DNA to perform logical operations and to solve more complex computational problems. DNA was used to create logic gates, the fundamental building blocks of digital circuits, and to build DNA-based calculators and other devices [39]. In the 2000s, the field of biocomputing took another leap forward with the development of biological computers—living cells that are engineered to perform computational tasks. These biological computers use genetic circuits to process information and make decisions based on their inputs. Researchers have engineered bacteria that can “count” by flipping a genetic switch each time they divide [40].

Figure 3.

Timeline for the development of biocomputers. A summary of key landmark studies leading to the fundamental understanding of biocomputers.

3.2 Role of human brain cells in biocomputing

Biocomputing, at its core, is the use of biological materials or biological architectures for computational purposes. It’s a field that merges biology, computer science, and information technology. Biocomputing has opened up exciting new possibilities in the field of neuroscience. One of the most intriguing developments is the use of human brain cells, or neurons, in biocomputing. Neurons have the potential to revolutionize the way we think about computing with their unique ability to process and transmit information. Neurons are the fundamental units of the brain, responsible for receiving sensory input from the external world, processing this information, and directing the body’s response. They communicate with each other through specialized connections known as synapses, where electrical signals are converted into chemical signals and vice versa. This complex network of neurons and synapses forms the basis of the brain’s computational power [41]. Just like transistors in an electronic circuit, neurons can switch on and off in response to certain inputs. However, unlike electronic transistors, neurons can process a wide range of inputs and can adapt their responses based on past experiences, a property known as plasticity [42]. Neurons can also be used to create biological neural networks, which can be trained to perform complex tasks. For example, researchers have grown networks of neurons on microelectrode arrays and trained these networks to control robotic devices. These biological neural networks have the potential to outperform traditional artificial neural networks in certain tasks, due to their inherent plasticity and adaptability [43].

One of the key applications of neurons in biocomputing is in the development of neuromorphic systems. Neuromorphic systems are designed to mimic the structure and function of the brain, with the aim of achieving the brain’s computational efficiency and adaptability. These systems use artificial neurons and synapses to process information in a parallel and distributed manner, similar to the brain [41]. Biocomputing also allows for the creation of detailed computational models of neural systems. These models can simulate the behavior of neurons and neural networks, providing insights into how the brain processes information. This can help researchers understand normal brain function and the mechanisms underlying neurological disorders [44]. Besides, neuroscience research often generates large amounts of complex data. Biocomputing techniques, such as machine learning and data mining, can help analyze this data, uncovering patterns and relationships that might not be apparent through traditional analysis methods. Biocomputing plays a crucial role in the development of Brain-Computer Interfaces (BCIs), which are systems that enable direct communication between the brain and an external device. BCIs can be used to restore or augment lost neurological functions, such as movement in individuals with paralysis or communication in individuals with severe speech impairments [45]. By decoding the electrical activity of neurons, BCIs can allow individuals to control prosthetic limbs, computer cursors, or even drones with their thoughts. Another subfield of biocomputing is neuromorphic computing that involves designing computer systems based on the architecture of the brain. Neuromorphic computers mimic the brain’s structure and function and could potentially perform complex tasks more efficiently than traditional computers [44].

3.3 The potential of biocomputers in medical research and treatment

One of the most promising applications of biocomputing is in disease diagnosis and monitoring. Researchers have developed biocomputers that can operate inside the human body, detect disease markers and even deliver targeted treatments. These biocomputers are designed to respond to specific biological signals, such as the presence of a particular protein associated with a disease [40]. In addition, biocomputing enhances the capabilities of medical imaging. By applying machine learning algorithms to imaging data, many researchers created models that can predict disease progression or response to treatment. This led to earlier and more accurate diagnoses, as well as more effective monitoring of disease progression [46]. Biocomputing techniques, such as machine learning and artificial intelligence (AI), can be used to analyze and interpret medical images. These techniques can identify patterns and features in images that may be difficult for humans to detect, leading to more accurate diagnoses. AI algorithms can be trained to identify tumors in medical images with a high degree of accuracy. Biocomputing algorithms can be used to reconstruct medical images, improving their quality and making them easier to interpret. This is particularly useful in modalities like computed tomography (CT) scan and magnetic resonance imaging (MRI), where raw data must be transformed into a visual format [47]. In medical imaging, segmentation is the process of identifying and delineating specific structures or regions of interest in an image. Biocomputing can automate this process, saving time and improving consistency. On the other hand, radiomics which is a field that extracts a large number of quantitative features from medical images using data-characterization algorithms, can be used to create predictive models for diagnosis, prognosis, or therapy response, which is particularly useful in personalized medicine. Biocomputing can also automate and improve the accuracy of the process image registration in which different sets of data are transformed into one coordinate system and is often necessary when multiple imaging modalities are used [48]. Biocomputing can be used to create 3D visualizations of medical images, providing a more comprehensive view of anatomical structures. This can be particularly useful for surgical planning and patient education.

Biocomputing also has significant potential in the field of drug discovery. Computational models of biological systems are being used to predict how potential drugs will interact with their targets in the body. This has speeded up the drug discovery process and reduced the need for animal testing [49]. Biocomputing allows for in silico screening of large databases of chemical compounds to identify potential drug candidates. These techniques help to predict how a compound will interact with a particular biological target based on its structure and properties. This has significantly speeded up the initial stages of drug discovery by reducing the number of compounds that need to be tested in the lab [50]. In addition, biocomputing is used to create predictive models of biological systems. These models simulate the interactions between potential drugs and their targets, predicting the efficacy and potential side effects of a drug before it is tested in the lab or in clinical trials. Once potential drug candidates are identified, biocomputing can be used to optimize these compounds. This can involve tweaking the structure of the compound to improve its efficacy, reduce side effects, or improve other properties such as solubility or stability [49]. Moreover, biocomputing has contributed to the development of personalized medicine. By using computational models to analyze a patient’s genetic data, doctors could predict how the patient will respond to different treatments. This has led to more effective and personalized treatment plans [50].

4. Organoid intelligence (OI): the next frontier in biocomputing

OI refers to the integration of organoid technology with computational methods to enhance our understanding of organoid behavior and to predict their responses to various stimuli. High-throughput techniques such as single-cell RNA sequencing, proteomics, and imaging are used to generate comprehensive datasets from organoids. These datasets can provide detailed information about the organoid’s cellular composition, structural organization, and functional dynamics [51]. Advanced computational methods, including machine learning and AI, are used to analyze the complex datasets generated from organoids. These analyses reveal patterns and relationships that are not readily apparent, providing insights into the organoid’s behavior. The data and insights obtained from organoids are used to create predictive models. These models then simulate the organoid’s response to various stimuli, such as drugs or genetic modifications. Thus, this guides the experimental design and accelerate the pace of research [52].

4.1 The concept of OI

The advent of organoid technology has revolutionized the field of biological research, particularly in the realm of neuroscience. Brain organoids provide a unique platform for studying neurodevelopmental processes, disease modeling, and drug testing. However, the intersection of organoid technology with AI has given rise to a new concept called organoid intelligence or OI. It is an emerging field that leverages the extraordinary biological processing power of brain organoids and the computational prowess of AI [53]. This fusion aims to create a new frontier in biocomputing and ‘intelligence-in-a-dish’, providing exceptional opportunities for understanding the human brain and developing novel therapeutic strategies. Table 2 summarizes the advantages and disadvantages of OI. Brain organoids are generated through a process that mimics the stages of human brain development. Starting from PSCs which have the potential to differentiate into any cell type in the body, researchers guide these cells to become neural progenitor cells. These progenitors then self-organize and differentiate further into various types of brain cells, forming organoids with structures similar to those found in the human brain [54]. The resulting organoids contain multiple types of neurons and supportive glial cells, and they exhibit complex activities such as neuron firing and network formation.

Aspect	Advantages	Disadvantages
Research potential	Enables real-time analysis and feedback on organoid behavior. Provides a platform for advanced research and predictive modeling.	Still in its infancy, so the full potential and applications are not yet fully understood.
Data analysis	AI can process and analyze vast amounts of data from organoids quickly. Can identify patterns and insights that might be missed by human researchers.	Risk of over-reliance on AI, potentially missing nuanced biological insights.
Personalized medicine	Potential for real-time adaptation and personalized therapeutic strategies based on individual organoid responses.	Ethical concerns about creating personalized brain models and potential misuse.
Technological integration	Combines the strengths of biological and computational research. Can lead to the development of novel technologies and methodologies.	Technological challenges in seamlessly integrating organoids with AI systems.
Ethical concerns	Can provide insights without the need for animal or human testing in some cases.	Amplified concerns about creating “conscious” tissue. Ethical dilemmas related to AI ethics, data privacy, and potential misuse.
Cost and accessibility	Potential for scalable solutions in research and medicine with the integration of AI.	High initial costs due to the integration of advanced technologies. May not be accessible to all researchers or institutions.

Table 2.

Advantages and disadvantages of OI.

The concept of OI takes this a step further. By integrating AI methodologies, researchers analyzed and interpreted the complex patterns of activity within brain organoids. This provided insights into how neurons communicate and form networks, how these networks change over time, and how they respond to various stimuli. Furthermore, AI helped to optimize the process of organoid generation, improving the reproducibility and scalability of organoid production [52]. Recent research has highlighted more on the formation of OI using brain organoids. One of the study presented a novel approach in creating biologically relevant environment for the growth and development of human embryonic stem cell (hESC)-derived brain organoids. The authors used a decellularized adult porcine brain extracellular matrix (B-ECM) as a scaffold for the organoids [55]. The decellularization process involved removing cells from the ECM using chemical, physical, and enzymatic methods. This process resulted in a cell-free ECM scaffold with minimal interference from cellular antigens and metabolites. The B-ECM was then processed into a hydrogel, which was used as a scaffold for the organoids. The hydrogel formation process involved solubilizing the ECM proteins with acids and enzymes, specifically pepsin, which cleaves the non-helical protein regions outside of the triple helix protein structure of collagen. After neutralization to physiological pH, the hydrogel formation followed a collagen-based self-assembly process. The authors found that hESCs cultured in B-ECM hydrogels showed gene expression and differentiation outcomes similar to those grown in Matrigel [55]. This indicated that B-ECM hydrogels can be used as an alternative scaffold for human cerebral organoid formation. The use of B-ECM hydrogels as a scaffold for organoid formation is significant because it provides a more biologically relevant environment for the organoids, which could potentially enhance their development and functionality. This could have important implications for the study of neurodevelopment and disease, as well as for the development of OI (Figure 4) [56].

Figure 4.

Schematic representation of the development process of OI. The process starts with PSCs and progresses through differentiation into neural progenitor cells, self-organization into brain organoids, AI analysis of organoid activities, optimization of organoid production with AI, integration with biologically relevant scaffolds like B-ECM hydrogels, and culminates in the creation of ‘intelligence-in-a-dish’.

4.2 Current research and advancements in the field

OI is an emerging interdisciplinary field that combines the biological processing capabilities of organoids with the computational power of AI. One of the significant advancements in the field of OI is the development of more biologically relevant environments for organoid growth. An exciting development is the integration of AI with organoid and organ-on-a-chip technologies. One of the studies discussed the convergence of AI and microfluidics in biomedicine, with a focus on organoids-on-chips [57]. This integration allowed for the rapid analysis of complex organoid data, potentially leading to the development of novel therapeutic strategies. Furthermore, advancements in the efficient culture of organoids derived from adult human stem cells have laid the foundation for PDO culture. This advancement was particularly significant for personalized medicine, as it allowed for the testing of drug responses on organoids derived from a patient’s own cells [58].

Currently, researchers synthesize enhanced organoids that have the potential to mimic the computational and cognitive capabilities of an in vivo brain. A laboratory-grown brain organoid is characterized by high levels of myelination, a greater abundance of support cells, and a capacity for spontaneous neural activity, representing a substantial advancement from earlier iterations that lacked such physiological complexity [59]. With these synthesized mini-brains, few studies proposed using available 3D microelectrode arrays to perform external electrophysiological recordings of brain organoids to perform tasks such as learning, memory, and other cognitive functions [52]. Other studies have shown latent diffusion models that can reconstruct regional brain activation under fMRI, thereby decoding visual stimuli into high-fidelity images. Just as AI can be trained, OI could be taught to recognize and respond to specific environmental stimuli, potentially surpassing the performance of any modern computer. With the help of sensory units from organoids, OI may even be able to display decoded neural outputs on a digital screen, offering a novel way to understand complex biological processes. With the combined development of AI and OI, technology that leverages both systems could surpass the human brain as a superior information processing device [52]. Previous clinical trials for neurologic conditions such as Alzheimer’s disease, autism, and schizophrenia have exhibited poor success rates, possibly attributed to the poor translation of research findings from animal models to human pathophysiology. Brain organoids created by reprogramming a small sample of a patient’s skin cells, offer a non-invasive method to test brain tissue without the need for extracting tissue from living specimens [53]. Access to these recapitulated brains may allow for novel biomarker discovery using “omics” during initial disease stages, which could lead to innovative and personalized therapeutics.

4.3 The potential applications and implications of OI

OI marks the efficiency of the human brain in processing complex information compared to machines. While machines can perform tasks faster, the human brain is more efficient in terms of power consumption and data efficiency. This has led to high expectations for biological, brain-directed computing as an alternative to silicon-based computing, with the potential for the advances in computing speed, processing power, data efficiency, and storage capabilities, all with lower energy needs [51]. For instance, Smirnova et al. proposed biofeedback to systematically train organoids with increasingly complex sensory inputs and output opportunities. They envision interfacing the brain organoids with computers, sensors, and machine interfaces to facilitate supervised and unsupervised learning. The authors use the term “OI” for this approach to stress its complementarity to AI, where computers aim to perform tasks done by brains, often by modeling our understanding of learning [56]. Glioblastoma, a grade IV astrocytoma, is the most aggressive primary brain tumor, with an overall median survival of 16.0 months following standard treatment. Despite intensive treatment, the tumor almost invariably recurs. This poor prognosis is attributed to the initiation, propagation, and differentiation of cancer stem cells. Current in vitro models are limited at preserving the inter-and intra-tumoural heterogeneity of primary tumors. Therefore, brain cancer research models aims to recapitulate glioblastoma stem cell function, while remaining amenable for analysis [60]. Cerebral organoids are emerging as cutting-edge tools in glioblastoma research. The opportunity to generate cerebral organoids via iPSCs, and to perform co-cultures with patient-derived cancer stem cells (GLICO model), has enabled the analysis of cancer development in a context that better mimics brain tissue architecture.

Bioprinting technology is an advanced technique that can print various cells and biomaterials at desired locations, allowing for the creation of complex 3D biological structures. Evidence suggest that this technology can help standardize organoids and automate the fabrication process [61]. This technology has been increasingly used for the creation of brain organoids, which are miniaturized and simplified versions of an organ produced in vitro in 3D that show realistic micro-anatomy. In addition, the Blood-Brain Barrier (BBB) is a highly selective semipermeable border of endothelial cells that prevents solutes in the circulating blood from non-selectively crossing into the extracellular fluid of the central nervous system where neurons reside. The BBB is crucial for protecting the brain from harmful substances and maintaining homeostasis, but it also poses a significant challenge for drug delivery to the brain. Tan et al. developed 3D cellular and organoid culture for BBB in neuropharmacology. The authors used nanorobotics and AI to design, control, and target the BBB. Nanorobotics involved the use of microscopic robots for medical purposes, while AI was used to optimize these processes [58].

The implications of OI are vast, spanning from advancements in understanding brain development and diseases, to ethical considerations, and the potential for novel biocomputing models. Brain organoids offer unique opportunities for studying human brain development and neurological disorders. They allow scientists to model the complex interactions and processes that occur during brain development, which can lead to a better understanding of various neurological conditions. For instance, the use of brain organoids has already provided insights into the early stages of neurodevelopmental disorders, which are difficult to study in humans [62]. As we venture into creating entities that closely resemble human brains, ethical questions inevitably arise. One of the primary concerns is the potential for these organoids to develop some form of consciousness. If brain organoids were to reach a level of complexity where they could experience sensations or emotions, this would raise serious ethical questions about their use and treatment. There are also concerns about the potential misuse of this technology, such as in creating human-animal chimeras or the potential for unauthorized use in creating enhanced beings or biological weapons. The concept of OI introduces the possibility of using brain organoids as a form of biological computing. This could involve connecting brain organoids with real-world sensors and output devices, and even with each other, to create complex, networked interfaces. Such systems could allow for faster decision-making, continuous learning during tasks, and greater energy and data efficiency. This could revolutionize fields such as artificial intelligence, robotics, and data processing, opening up new avenues for exploration and innovation. Moreover, the implications of OI are profound, offers exciting possibilities for scientific discovery and technological innovation, while they also pose significant ethical challenges that need to be carefully navigated.

5. Brain-computer Interface (BCI): understanding brain connectivity

BCI is a rapidly evolving technology which revolutionizes our understanding of brain connectivity and has the potential to drastically alter the landscape of neuroscience, medicine, and human interaction with technology. BCIs enable direct communication between the brain and an external device, bypassing the traditional route of peripheral nerves and muscles. This technology has profound implications for understanding brain connectivity, with potential applications ranging from assisting individuals with neurodegenerative disorders to enhancing human cognition [61]. Understanding brain connectivity is a fundamental aspect of neuroscience. The human brain is a complex network of billions of neurons interconnected through trillions of synapses. These connections form intricate networks that facilitate the transmission and processing of information, leading to cognition, behavior, and consciousness. BCIs provide a unique window into these networks, allowing scientists to monitor and interpret the electrical activity of the brain in real-time. The intricate connections between different brain regions are essential for cognitive functions [63]. Brain organoids provide a unique opportunity to study neural connectivity and synapse formation. This enables researchers to investigate the impact of genetic mutations or environmental factors on neural network development and function, elucidating how disruptions in connectivity may contribute to neurodevelopmental disorders.

BCIs work by detecting changes in brain activity by typically using electroencephalography (EEG) or functional magnetic resonance imaging (fMRI). These changes are then translated into commands that control the external devices. For instance, a person might think about moving their right hand, which would generate a specific pattern of brain activity. The BCI can detect this pattern and translate it into a command to move a robotic arm [45]. One of the most promising applications of BCIs is in the field of neurorehabilitation. For individuals with neurodegenerative disorders or those who have suffered a stroke, BCIs can provide a means of communication and control that is independent of muscle control [64]. This can significantly improve the quality of life for these individuals, allowing them to interact with their environment and communicate with others despite their physical limitations. In addition to their therapeutic applications, BCIs also offer exciting possibilities for enhancing human cognition. By interfacing directly with the brain, BCIs potentially augment any cognitive abilities which can improve memory, attention, and even sensory perception. This could have far-reaching implications, from improving performance in everyday tasks to enhancing human capabilities in fields like science, engineering, and art [45].

6. Probes to connect brain organoids and computer

The integration of brain organoids and computers represents a cutting-edge frontier in the field of neuroscience and biotechnology. This integration is made possible through the use of organoids with computer systems. These probes serve as a bridge, enabling the transmission of information between the biological and digital realms [61]. The probe devices that connect brain organoids and computers are typically microelectrode arrays (MEAs) which are devices that contain multiple microscopic electrodes and can detect and record electrical activity from neurons. When these probes are inserted into a brain organoid, they pick up the electrical signals generated by the organoid’s neurons. These signals are then amplified, digitized, and transmitted to a computer for analysis [65]. The use of MEAs in brain organoid research has several significant implications. Firstly, it allows for the real-time monitoring of organoid activity, providing insights into how neurons in the organoid communicate with each other. This can help researchers understand the fundamental processes of brain development and function. Secondly, MEAs can be used to stimulate the neurons in the organoid, allowing researchers to investigate how different types of stimulation affect neuronal activity. This could lead to the development of new treatments for neurological disorders which are characterized by abnormal neuronal activity. Thirdly, the integration of brain organoids and computers via MEAs opens up the possibility of creating biological computers or biocomputers. Biocomputers use biological materials, such as DNA or neurons, to perform computational operations. Brain organoids connected to computers could potentially be trained to perform specific tasks, such as pattern recognition or data processing, offering a novel approach to computing that leverages the unique capabilities of biological systems [55].

7. Training of brain organoid with AI

AI, particularly machine learning and deep learning, has the potential to transform the way one analyzes and interprets complex biological data. In the context of brain organoids, AI can be used to analyze large-scale data sets, identify patterns, and make predictions that would be impossible for humans to do manually. This can include everything from predicting the development of specific cell types to identifying the effects of different drugs or genetic modifications [23]. The first step in training brain organoids using AI is the creation of the organoids themselves. This is typically done using human PSCs which are cultured, and they differentiate into brain cells through the application of various growth factors. Over time, these cells self-organize into 3D that resemble aspects of the human brain. Once the organoids have been created, they are used to generate data for AI training. This typically involves using advanced imaging techniques to capture detailed images of the organoids at various stages of development. These images are then analyzed using machine learning algorithms that identify patterns and make predictions about the organoids’ behavior. Training brain organoids with AI involves using machine learning algorithms to analyze the data generated by the organoids and make predictions about their behavior. This is done in a supervised or unsupervised manner. In supervised learning, the algorithm is trained on a set of labeled data, where the correct outcome is known. The algorithm then uses this training to make predictions on new, unlabeled data. In the context of brain organoids, this could involve training the algorithm to recognize specific cell types or developmental stages based on a set of labeled images [66]. On the other hand, unsupervised learning involves training the algorithm to identify patterns in the data without any pre-existing labels. This can be particularly useful in the context of brain organoids, where the complexity of the data can make it difficult to identify relevant patterns manually.

After the initial training, the model’s performance is validated. This typically involves using the model to make predictions on the testing set and comparing these predictions to the actual outcomes. The model’s parameters are adjusted and the model retrained multiple times to achieve optimal performance. Once the AI model has been trained and validated, it is used to guide the development of the brain organoids. This involve the use of the model to predict the effects of different growth factors or genetic modifications on the organoids’ development, and then adjust these variables based on the model’s predictions [51]. The development of the organoids should be continuously monitored, with the AI model’s predictions being used to make ongoing adjustments. This iterative process of prediction, adjustment, and monitoring is key to guiding the organoids towards the development of organoid intelligence. Finally, the level of OI achieved is evaluated. This involve assessing the organoids’ ability to perform basic information processing tasks, their short- and long-term memory capabilities, and other cognitive functions [67].

8. Challenges and future directions

Despite the exciting developments, there are also significant challenges in the field of biocomputing. One of the main challenges is the development of reliable and efficient methods for programming biological systems. Unlike traditional computers, which use a standard binary language, there is no universal language for programming biological systems. Additionally, biological systems are inherently noisy and variable, which can lead to errors in computation [68]. Another challenge is the integration of biocomputers with traditional electronic devices. While there has been progress in developing interfaces between biological and electronic systems, there is still much work to be done to achieve seamless integration. Besides culturing neurons and maintaining them in a healthy state is a complex task that requires precise control over the cellular environment. Furthermore, interfacing neurons with electronic devices is a significant technical challenge, due to the differences in scale and material properties between biological and electronic systems. However, ongoing research and technological advancements are helping to overcome these challenges. As our understanding of biology and our computing capabilities continue to grow, the potential of biocomputing in medical research and treatment will only increase.

In contrast, the concept of OI is not without its challenges and ethical considerations. As brain organoids become increasingly complex and start to exhibit higher-order functions, questions arise about their moral and legal status. Some researchers have raised concerns about the potential for organoids to develop consciousness, although this is currently purely speculative and far from the capabilities of current organoid technology. Moreover, the use of AI in organoid research brings its own set of challenges. AI algorithms require large amounts of data for training, and the data generated by organoid research is both complex and high-dimensional. Ensuring the accuracy and interpretability of AI models is crucial for their effective application in organoid research [62]. Along with creating these intelligent, human-like brains, some consider the possibility of organoids autonomously developing sentience, awareness, or pain perception. The ethical issue of OI includes the possibility of being fed biased or harmful information as well as concerns regarding cost, privacy and data mining, equity among users, intellectual property protections, and disruption of human-occupied jobs. To overcome these relevant issues surrounding OI, researchers suggest performing a comprehensive ethical analysis that includes a review from diverse groups to foster responsibility, accountability, and trust with the stakeholders. To ensure the successful integration of this technology, unmet limitations will need to be addressed first.

Currently, the main limitation of OI and brain organoids is the discrepancy in complexity compared to actual human brains. Today, brain organoids are below 500 micrometers in diameter and have less than 100,000 cells. These relatively non-complex models fail to show the developmental asymmetry nor the predictable anatomy that is needed to supplement preclinical trials of neurological conditions. As a result, decades of work will likely be necessary to develop organoids advanced enough to replace modern AI. Moreover, additional time is needed to develop the technology required for fully transmitting the computations of the organoid to a digital interface before it can practically convey clinically relevant information. Despite these hurdles, the proposition of OI “intelligence-in-a-dish” is an exciting new field that will surely attract more research. Complex brain organoids pose beneficial applications for understanding the development and treatment of neurological diseases. A long road of rigorous testing and embedded ethical mediation will be required before OI and brain organoids reach their potential.

9. Conclusion

Despite various challenges, the potential benefits of OI are immense. By providing a platform for studying the human brain in detail, brain organoids could revolutionize the understanding of neurodevelopmental and neurodegenerative disorders. The integration of AI could accelerate the research, enabling rapid analysis of complex organoid data and potentially leading to the development of novel therapeutic strategies. In conclusion, the intersection of brain organoid technology and AI in the form of OI represents a promising new frontier in neuroscience and biocomputing. As this field continues to evolve, it holds the potential to significantly advance the understanding of human brain and pave the way for novel treatments for a range of neurological disorders. OI represent a groundbreaking technology that has the potential to revolutionize our understanding of human brain development and disorders. As this field progresses, it is essential to strike a balance between scientific advancement and ethical considerations. With continued research and innovation, OI hold the promise of shaping the future of neuroscience, providing novel insights into brain complexity and unlocking new therapeutic opportunities for brain disorders.

Acknowledgments

The authors are thankful to Dr. D. Y. Patil Biotechnology and Bioinformatics Institute, Dr. D. Y. Patil Vidyapeeth, Pune for the physical infrastructure. Sangeeta Ballav is thankful to Dr. D. Y. Patil Vidyapeeth, Pune for Junior Research Fellowship (DPU/291/2021). This work is supported by the Ramalingaswami Re-entry fellowship, Department of Biotechnology, Govt. of India to S. Sur [BT/RLF/Re-entry/47/2021] and Intramural Grants, Dr. D. Y. Patil Vidyapeeth (DPU), Pimpri, Pune, India to S. Basu [DPU/644-43/2021].

Conflict of interest

There are no conflicts of interest.