EFRI BEGIN OI: Feasibility of 3D Biological Neurocomputers for Intelligent Biomedical Motor Control Systems

Conditions such as stroke lead to a permanent loss of brain tissue. 3D neural networks (3DNN) assembled from brain cells may offer a promising way to repair damaged tissue and restore lost brain functions after stroke. These 3DNNs can be taught to recognize intentions, perform motor functions, and self-correct for mistakes. This EFRI BEGIN OI study aims to develop 3DNNs and assess their ability to process brain signals underlying movement intentions, generate movement behavior, and use feedback to self-correct. 3D bioprinting techniques to produce robust, reliable, and reproducible 3DNNs will be developed. Core principles to program the 3DNNs with intelligent behavior will be established. Further anticipating the ethical, moral, legal, and social challenges of this technology and establishing a framework for responsible conduct of research will ensure its ethical development. The research will expand public participation in science/engineering by: providing educational opportunities in advanced bioengineering and neuroscience techniques; introducing students to the critical analysis of ethical, legal, and social dimensions of research; engaging the general public, including persons with disabilities, in relevant educational opportunities; and contributing to public understanding of ethical, legal, and social implications of developing intelligent 3DNNs. Cultured 3D neural networks offer a promising path to repair stroke-damaged tissue and restore lost neurological functions. As such, the long-term vision of this EFRI BEGIN OI project is to develop 3DNNs that interface with other brain/body areas and employ self-organizing properties to learn to interpret input and feedback signals. These 3DNNs could be trained to recognize cognitive signals, actuate motor functions, and self-correct for errors. Before this grand vision can be achieved, this study seeks to demonstrate proof-of-concept that cultured 3DNNs in vitro can indeed interpret brain signals underlying motor intentions, execute motor behavior, and utilize feedback to self-correct. While immense knowledge gaps exist across various scientific disciplines, critical areas for pursuing the above vision were identified. Specifically, this study will focus on establishing fabrication and maintenance methodologies to produce robust, reliable, reproducible 3DNNs and core principles to confer intelligent behavior upon them. Achieving these goals will provide the basis for future 3DNN iterations. Further anticipating the ethical, legal, and social challenges of this technology and establishing a framework for responsible conduct of research will ensure its ethical development and application in society. Finally, while this study is motivated by a neurorestoration application, its successful completion will advance the field of biocomputing broadly. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.