Data from: High-performance brain-to-text communication via handwriting
Willett, Francis et al. (2021), Data from: High-performance brain-to-text communication via handwriting, Dryad, Dataset, https://doi.org/10.5061/dryad.wh70rxwmv
Brain-computer interfaces (BCIs) can restore communication to people who have lost the ability to move or speak. In this study, we demonstrated an intracortical BCI that decodes attempted handwriting movements from neural activity in motor cortex and translates it to text in real-time, using a recurrent neural network decoding approach. With this BCI, our study participant, whose hand was paralyzed from spinal cord injury, achieved typing speeds that exceed those of any other BCI yet reported: 90 characters per minute at 94.1% raw accuracy online, and >99% accuracy offline with a general-purpose autocorrect.
This dataset contains all of the neural activity recorded during these experiments, consisting of 1,000 sentences (43,501 characters) over 10.7 hours. The neural activity was recorded with two microleectrode arrays implanted in hand area of motor cortex (96 electrodes each). The dataset also contains all of the real-time outputs of the handwriting BCI.
See the readme.pdf for a description of the data format (and this code for an example of how to train RNNs on this data).
U.S. Department of Veterans Affairs, Award: A2295R, N2864C
NIH: National Institute of Neurological Disorders and Stroke and BRAIN Initiative, Award: UH2NS095548, U01-NS098968
National Institute on Deafness and Other Communication Disorders, Award: R01DC009899, U01DC017844, R01-DC014034
Howard Hughes Medical Institute
Simons Foundation Collaboration on the Global Brain, Award: 543045
Larry and Pamela Garlick
Samuel and Betsy Reeves
Wu Tsai Neurosciences Institute at Stanford