Brain-computer interfaces (BCIs) can make your life easier if you have movement or speech disorders. BCIs could als0 allow you to manipulate prosthetic limbs and use computers, and also play games.
And maybe one day you could control robots to do your bidding.
But unless you want to have an invasive Neuralink-type device drill into your brain (or via other methods), you're currently stuck with wearing an EEG (electroencephalogram) device, like the guy above. Yes, it could analyze your brain waves, but for limited uses.
AI to the rescue
So Bin He in the Department of Biomedical Engineering at Carnegie Mellon University and colleagues turned to deep-learning decoders to hopefully improve a BCI’s performance.
The idea was to enable continuous and sustained movement of a virtual object. Twenty-eight adult participants were instructed to imagine moving their right hand to move the cursor right and their left hand to move the cursor left, moving both hands simultaneously to move upwards, and moving neither hand to go down.
The authors evaluated two different deep-learning architectures and a traditional decoder over seven BCI sessions.
A step towards neuro-assistive robotics
Both deep-learning decoders improved throughout the study and outperformed the traditional decoder by the final session.
The participants were able to control a fast, continuously moving computer cursor—tracking randomly moving objects with a high level of performance without moving a muscle.
Citation: Forenzo, D., Zhu, H., Shanahan, J., Lim, J., & He, B. (2024). Continuous tracking using deep learning-based decoding for noninvasive brain–computer interface. PNAS Nexus, 3(4). https://doi.org/10.1093/pnasnexus/pgae145 (open-access)