Everything we do starts as a flicker of intention in the brain. In this project, we leverage this fundamental insight to predict human motor movements by processing brainwave signals captured through Electroencephalography (EEG). In this project, utilizing a public dataset, we employ Long Short-Term Memory networks (LSTM) and Convolutional Neural Networks (CNN) to accurately classify six distinct phases of hand motor events:
- Hand Start: The initiation of hand movement.
- First Digit Touch: The first contact of the fingers with the object.
- Both Start Load Phase: Secure hold of the object with both fingers.
- Lift Off: The lifting of the object.
- Replace: The replacement of the object on the original surface.
- Both Released: Release of the object by both fingers.
The Road Ahead:
As we look to the future, our next step is to explore the potential of Spiking Neural Networks for this classification task. These networks offer the promise of more closely mimicking the natural processes of the human brain, opening up new avenues for research and application.