Decoding phantom limb neuro-mechanical function for a new paradigm of mind-controlled bionic limbs

Mind controlled bionic limbs promise to replace mechanical function of lost biological extremities and restore amputees’ motor capacity. State of the art approaches use machine learning for establishing a mapping function between electromyography (EMG) and joint kinematics. However, current approaches require frequent recalibration with lack of robustness, thus providing control paradigms that are sensitive to external conditions. This paper presents an alternative method based on the authors’ recent findings. That is, a biomimetic decoder comprising a computational model that explicitly synthesizes the dynamics of the musculoskeletal system as controlled by EMG-derived neural activation signals