Meeting brain–computer interface user performance expectations using a deep neural network decoding framework

Brain–computer interface (BCI) neurotechnology has the potential to reduce disability associated with paralysis by translating neural activity into control of assistive devices1–9. Surveys of potential end-users have identified key BCI system features10–14, including high accuracy, minimal daily setup, rapid response times, and multifunctionality. These performance characteristics are primarily influenced by the BCI’s neural decoding algorithm1,15, which is trained to associate neural activation patterns with intended user actions. Here, we introduce a new deep neural network16 decoding framework for BCI systems enabling discrete movements that addresses these four key performance characteristics. Using intracortical data from a participant with tetraplegia, we provide offline results demonstrating that our decoder is highly accurate, sustains this performance beyond a year without explicit daily retraining by combining it with an unsupervised updating procedure3,17–20, responds faster than competing methods8, and can increase functionality with minimal retraining by using a technique known as transfer learning21. We then show that our participant can use the decoder in real-time to reanimate his paralyzed forearm with functional electrical stimulation (FES), enabling accurate manipulation of three objects from the grasp and release test (GRT)22. These results demonstrate that deep neural network decoders can advance the clinical translation of BCI technology.Intracortical activity data recorded over 2 years in a tetraplegic patient is used to develop an artificial intelligence algorithm that achieves fast, accurate, and stable movement decoding to reenable real-time control of the paralyzed forearm.