Motor Cortex Coverage Predicts Signal Strength of a Stentrode Endovascular Brain-Computer Interface

Brain-computer interfaces (BCIs) are an emerging assistive technology for individuals with motor impairments, enabling the command of digital devices using neural signals. The Stentrode BCI is an implant, positioned within the brain's neurovasculature, that can record movement-related electrocortical activity. Over 5 years, 10 participants (8 amyotrophic lateral sclerosis, 1 primary lateral sclerosis, 1 brainstem stroke) have been implanted with a Stentrode BCI and significant inter-participant variability has been observed in the recorded motor signal strength. This variability warrants a critical investigation to characterize potential predictors of signal strength to promote more successful BCI control in future participants. Therefore, we investigated the relationship between Stentrode BCI motor signal strength and a variety of user-specific factors: (1) clinical status, (2) pre-implant functional activity, (3) peri-implant neuroanatomy, (4) peri-implant neurovasculature, and (5) Stentrode device integrity. Data from 10 implanted participants, including clinical demographics, pre- and post-implant neuroimaging and longitudinal Stentrode BCI motor signal assessments were acquired over a year. Across all potential predictors, the strongest predictor of Stentrode motor signal strength was the degree to which the Stentrode BCI's deployment position overlapped with primary motor cortex (M1). These findings highlight the importance of targeting M1 during device deployment and, more generally, provides a scientific framework for investigating the role of user-specific factors on BCI device outcomes.