Continual Learning approach for adaptive neural decoders

The proposed project aims to develop neural decoder models that can interpret brain signals in real-time to control stimulation devices, specifically focusing on Epidural Electrical Stimulation (EES) for upper limb rehabilitation. By implementing a continuous learning paradigm in decoders, the project seeks to enable the models to compensate for the effects of stimulation and adapt to different stimulating settings with minimal retraining. The expected benefit of this research is to advance the development of cutting-edge neuroprosthetic devices for rehabilitation, benefiting both the Université de Montreal and EPFL in Lausanne. Additionally, the project has the potential to contribute to the development of multi-subject models and guide the next generation of rehabilitation devices, aligning with the expertise and objectives of the participating institutions.

Faculty Supervisor:

Matthew Perich

Student:

Partner:

École polytechnique fédérale de Lausanne

Discipline:

Life Sciences

Sector:

Education

University:

Université de Montréal

Program: