Events Calendar

Real-Time Brain-Machine Interface Architectures: Neural Decoding from Plan to Movement

Speaker: Maryam M. Shanechi, Harvard Medical School, MIT EECS, Massachusetts General Hospital

Talk Title: Real-Time Brain-Machine Interface Architectures: Neural Decoding from Plan to Movement

Abstract: Developing brain-machine interfaces (BMI) that aim to enable motor function in patients with movement disabilities is an active area of research in computational neuroscience and neuroengineering. BMIs work by recording the neural activity, mapping or decoding it into a motor command, and then controlling a device such as a robotic arm. Research in this area has largely focused on the problem of restoring the original motor function. However, performance of such BMIs needs to be significantly improved before they become clinically viable. Moreover, while developing high-performance BMIs with the goal of matching the original motor function is indeed valuable, a compelling goal is that of designing BMIs that can surpass original motor function. In this work, I first develop a novel real-time BMI for restoration of natural motor function and then introduce a BMI architecture aimed at enhancing original motor function. I demonstrate the successful implementation of both these designs in rhesus monkeys.

To facilitate the restoration of lost motor function, I develop a two-stage decoder to decode jointly the target and trajectory of a reaching movement.
First, the decoder predicts the intended target from the spiking activity prior to movement. Second, it combines the decoded target with the spiking activity during movement to estimate the trajectory. The second stage uses an optimal feedback-control design that emulates the sensorimotor processing underlying actual motor control and directly processes the spiking activity using point process modeling in real time. I show that the two stages of the BMI result in a significantly more robust and accurate estimation of movement than is possible by using either stage alone or by using common regression approaches.

To enable enhancement of the original motor function, I introduce a real-time concurrent BMI architecture for performing complex tasks that involve a sequence of planned movements. In contrast to a traditional BMI, in this architecture, the BMI decodes all the elements of the sequential motor plan concurrently from working memory prior to movement. This in turn allows the BMI to analyze the complete sequence before action and find potential ways to perform the task more effectively, such as more quickly, than is possible by natural movement. I demonstrate the feasibility of such a concurrent architecture and that indeed sequential motor plans can be decoded simultaneously, accurately, robustly, and in advance of movement.

Biography: Maryam M. Shanechi received the B.A.Sc. degree with honors in
Engineering Science from the University of Toronto in 2004 and the S.M. and Ph.D. degrees in Electrical Engineering and Computer Science (EECS) from the Massachusetts Institute of Technology (MIT) in 2006 and 2011, respectively. She is currently a postdoctoral fellow with joint appointments at Harvard Medical School, MIT EECS, and Massachusetts General Hospital. Her research interests are at the interface of computational neuroscience, statistical signal processing, and
information and control theories. She has received various awards for academic achievement including the Professional Engineers of Ontario (PEO) gold medal and the Wilson Medal. She is the recipient of the Natural Sciences and Engineering Research Council of Canada (NSERC) doctoral fellowship and the Hewlett-Packard (HP) doctoral scholarship.

Host: Alice C. Parker

Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248

Audiences: Everyone Is Invited

Contact: Annie Yu