Neuromuscular OpenCap Dataset

The Neuromuscular OpenCap Dataset (Neuromuscular_OpenCap_Dataset.zip) contains time series of body segment kinematics, marker positions, scaled musculoskeltal models, and metadata dictionaries from 415 individuals (675 data collection visits) performing movement tasks. All data were recorded and processed using OpenCap, a video-based markerless motion capture smartphone application. Our cohort includes individuals with 12 different rare, genetically-defined neuromuscular disorders and age-matched healthy controls. We also release an acompanying demographics CSV (nmd_opencap_participant_info.csv) that contains de-identified demographic information, time function test (TFT) times, and patient-reported outcome measures (ACTIVLIM survey results) for all participants in our dataset. See our GitHub to download the disease detection and impairment severity quantification code and models that were trained and tested on this dataset. Follow the instructions on our README.md to reproduce all figures, tables, and statistics from the below manuscript (currently in submission), which also contains more details about our dataset: Deep learning models detect neuromuscular disease and quantify functional impairment from video-derived biomechanics data Sydney Covitz, Parker S. Ruth, Shelby Vogt-Domke, Carmichael Ong, Tian Tan, Audrey Chun, Sarah Ismail, Lin Karman, Julie Muccini, Shannon Li, Melina Rogers, Scott Uhlrich, John W. Day, Jennifer L. Hicks, Constance de Monts, Tina Duong, Scott L. Delp