Stable neural decoding improves brain machine interface performance across sessions
Stable Neural Decoding Across Sessions via Task-Conditioned Latent Alignment for Brain-Machine Interfaces
Machine Learning
Summary
Brain-machine interfaces (BMIs) use brain activity to control devices, but the signals can change a lot from day to day, making control unreliable. The authors developed a new method called Task-Conditioned Latent Alignment (TCLA) that helps keep decoding stable by aligning the brain signals according to specific tasks during each session. This method effectively maps brain signals into a shared space that remains consistent, improving the accuracy of decoding over time and across different subjects. The authors tested TCLA extensively on monkey data and found it to be more reliable than existing methods.
What this means in practice
- •For bmi device developers: Improve long-term accuracy of brain-machine interfaces by stabilizing neural decoding across multiple sessions and tasks.$Commercial implications: Enables more reliable brain-controlled devices that can be sold to medical and assistive technology providers for users needing continuous BMI use.
- •For neuroprosthetics engineers: Enhance adaptation methods to maintain stable control signals when neural recordings vary across sessions or individuals.
Authors
Canyang Zhao, Bolin Peng, J. Patrick Mayo, Ce Ju, Bing Liu
Abstract
Achieving stable long-term neural decoding in invasive brain-machine interfaces (BMIs) remains challenging due to variations in recorded neural populations across sessions. Current latent alignment approaches may overlook task-dependent structure during cross-session adaptation. We propose Task-Conditioned Latent Alignment (TCLA), a framework that stabilizes neural decoding by learning a shared latent space. TCLA learns a low-dimensional source representation using neural reconstruction and continuous behavioral supervision. During target-session adaptation, the shared representation is fixed, while target neural activity is mapped into the source latent space by aligning source and target distributions separately for each task condition. We evaluated TCLA on seven nonhuman primate datasets spanning multiple tasks. In long-term cross-session evaluation, TCLA achieved a mean $R^2$ of $0.476\pm0.014$ with a negative $R^2$ failure rate of only 6.8\%. Across 1,356 within-subject session pairs, TCLA achieved a mean $R^2$ of $0.371\pm0.009$ with a failure rate of 6.8\%. Across 2,134 cross-subject session pairs, TCLA achieved a mean $R^2$ of $0.218\pm0.004$ with a failure rate of 12.9\%, substantially better than those of the comparison methods. These results demonstrate that by preserving behaviorally relevant and task-dependent latent structure, TCLA improves the robustness of neural decoding across recording sessions and subjects. The source code is publicly available at \href{https://github.com/FAMD-CASIA/TCLA}{https://github.com/FAMD-CASIA/TCLA}.