Intracortical brain-computer interfaces (iBCIs) have been successful at decoding motor intention in individuals with sever neuromuscular impairments, enabling such tasks as typing, word recognition, and control of a computer cursor. However, although many different models have been proposed for decoding the high-dimensional neural data, collected from hundreds of electrodes on millisecond timescales, many of these models are forgotten or are not properly compared against other models. This project implemented a variety of decoding algorithms for computer cursor control and compared the performance of each. Due to the interdisciplinary nature of BCI research, these models have been developed using methods from electrical engineering, applied mathematics, machine learning, and neuroscience, and thus differ in their frameworks. Specifically, this project examined models of neural state dynamics in a Kalman filter framework, the ultimate output of each model being cursor position at each timestep. The data were from simulations using eyetracking (eye image) and noisy cursor data instead of real neural data. We found that an extended Kalman filter outperformed other algorithms, and work is ongoing to implement that algorithm into the neural signal processing pipeline. Selecting the best decoding methods allows the field of iBCI research to continuously improve performance in the face of such variables as fluctuations of brain state over time, multiple neurological impairments, and neurodegeneration, and can also translate into quality-of-life improvements for users of iBCIs.
Advani, Anand,
"Comparison of Decoding Algorithms for Cursor Control for an Intracortical Brain-Computer Interface"
(2024).
Summer Research Symposium.
Brown Digital Repository. Brown University Library.
https://doi.org/10.26300/r0pk-g211
Each year, Brown University showcases the research of its undergraduates at the Summer Research Symposium. More than half of the student-researchers are UTRA recipients, while others receive funding from a variety of Brown-administered and national programs and fellowships and go …
