EEG motor imagery classification

Compact deep learning models advance EEG motor imagery on consumer headsets

Finishing a master's thesis is no small feat, and benchmarking three deep learning architectures on a consumer EEG headset like OpenBCI's Galea shows real initiative.

3 min readMachine Learning
From Machine Learning

I may be being silly but I have just finished my masters thesis which benchmarked three deep learning architectures (one of which is novel), across varying preprocessing pipelines for the purpose of EEG motor imagery task classification SPECIFICALLY on a consumer grade EEG cap (OpenBCI’s Galea).

The novel architecture was built to be compact in param count as well as prevent overfitting, etc.

Read the original at Machine Learning