Tiny Machine Learning for Controlling Brain-Computer Interfaces

Summary

This study aims to evaluate if we can use a single ultra small, low-power and cheap computer chip processor to run a brain-computer interface (BCI) application. In this work, the computer typically responsible for processing real-time brain data used in BCIs that is recorded by sensors on the scalp will be replaced by these ultra small computers, called ‘edge-devices’. This is an initial proof-of-concept study to demonstrate the feasibility of this approach. This work will enable efficient, affordable, BCI applications to be built which could be used on anywhere. Success in this will support ongoing research and implementation of BCIs for children living with complex needs happening in the BCI4Kids Clinical and Engineering programs.