Enhancing Brain Machine Interface Decoding Accuracy through Domain Knowledge Integration.

This paper introduces a novel decoding approach for Brain Machine Interface (BMI) that enhances the estimation accuracy and stability of muscle activity by incorporating domain knowledge of motor control. Our approach uniquely integrates domain knowledge, focusing on the relationship between torque direction and muscle activity in isometric wrist tasks. We demonstrate the effectiveness of our approach through decoding analysis with non-human primates performing a wrist torque tracking task. By implementing a Kalman filter augmented with models of muscle activity and torque for specific movement directions, we show significant improvements compared to vanilla Kalman filter in the accuracy of muscle activity estimation. The proposed approach presents a promising direction for enhancing the performance of BMI by leveraging domain-specific insights into motor control.