This summer, I worked as a research assistant at the Ganguly Lab (UCSF Weill Institute of Neuroscience). The lab develops brain-computer interfaces (BCI) that enable individuals with paralysis to control neuroprosthetic devices using brain signals (see this Cell publication from the lab). The slides below an overview of the lab's work on BCI.
A major challenge in BCI systems is signal drift: the neural signals recorded from brain implants degrade and change over time. This instability means that gesture decoding algorithms, limiting the long-term reliability of neuroprosthetic control.
Currently, I am working on anomaly detection for long-term gesture decoding stability for the control of a neuroprosthesis. The following slides provide some details on my work presented at the Lab in August 2025.