AI for brain-machine interfaces

The field of brain machine interfaces is advancing rapidly with many academic labs and neurotechnology companies entering the field. AI can vastly improve this new technology. In collaboration with the Emami lab at Caltech, we have used neural networks to extract features of electrically recorded activity, increasing the amount of information available for interpretation by neural decoders. Additionally, a new class of brain machine interfaces that use ultrasound technology has the bottleneck of requiring rapid analysis of large amounts of imaging data. Collaborating with the Anandkumar lab, we have used machine learning techniques to improve the amount of imaging data that can be analyzed in real time for brain control applications.

Watch Video

Bio

Richard Andersen is the James G. Boswell Professor of Neuroscience and the T&C Chen Brain-Machine Interface Center Leadership Chair at Caltech. Andersen obtained his Ph.D. from the University of California, San Francisco and completed a postdoctoral fellowship at the Johns Hopkins Medical School. He was a faculty member of the Salk Institute and MIT before coming to Caltech. Andersen discovered gain-fields, the method the brain uses to transform signals between spatial representations. He also discovered neural signals of intention, proving that they are not sensory in nature but rather reflect the planning of the subject. He has applied this discovery of intention to advance research in brain-machine interfaces, showing that paralyzed patients' intentions can be decoded from brain activity to control computers and robotic devices. Andersen is a member of the National Academy of Sciences, the National Academy of Medicine, and the American Academy of Arts and Sciences.