Novel neural signal features permit robust machine-learning of natural tactile- and proprioception-dominated dorsal column nuclei signals

Neural prostheses enable users to effect movement through a variety of actuators by translating brain signals into movement control signals. However, to achieve more natural limb movements from these devices, restoration of somatosensory feedback and advances in neural decoding of motor control-related brain signals are required. We used a machine-learning approach to assess signal features for their capacity to enhance decoding performance of neural signals evoked by natural tactile and proprioceptive somatosensory stimuli, recorded from the surface of the dorsal column nuclei in urethane-anaesthetised rats. We determined signal features that are highly informative for decoding somatosensory stimuli, yet these appear underutilised in neuroprosthetic applications. We found that proprioception-dominated stimuli generalise across animals better than tactile-dominated stimuli, and we demonstrate how information that signal features contribute to neural decoding changes over a time-course of dynamic somatosensory events. These findings may improve neural decoding for various applications including novel neuroprosthetic design.