Ninapro is a a publicly available multimodal database to foster research on robotic & prosthetic hands controlled with artificial intelligence. Ninapro includes electromyography, kinematic, inertial, eye tracking, visual, clinical and neurocognitive data. Ninapro data are used worldwide by scientific researchers in machine learning, robotics, medical and neurocognitive sciences. ProHand is an innovative robotic prosthetic hand controlled with machine learning and produced with additive manufacturing. PaWFE is a signal feature extraction code that can extract widely used signal features in parallel, strongly reducing computational times. Dataset 1 includes data from 27 intact subjects. Dataset 2 includes data from 40 intact subjects. Dataset 3 includes data from 11 transradial amputees (with amputation levels as represented in the figure at the end of the page). Dataset 4 includes 10 intact subjects recorded with "Cometa" electrodes. Dataset 5 includes 10 intact subjects recorded with two Thalmic Myo armbands, putting them on the same forearm simultaneously. Dataset 6 contains repeatability data from 10 intact subjects repeating data acquisitions 2 times per day for 5 days. Dataset 7 contains data from 20 intact subjects and 2 transradial amputees. Dataset 8 is intended to be used for estimation/reconstruction of finger movement rather than motion/grip classification. Data are from 10 intact subjects and 2 amptuees. Dataset 9 includes kinematic data from 77 intact subjects acquired with a Cyberglove-II. Dataset 10, MeganePro, is the first multimodal database from intact and hand amputated subjects including sEMG, intertial, gaze tracking, visual, behavioral and clinical data for prosthetics and the analysis of phantom limb sensation.
In our research, we worked on DB2 including 40 intact subjects and 0 amputees. The NINAPRO DB2 could be downloaded from the official website http://ninaweb.hevs.ch/ with a file size of around 50GB inclusive of all exercises for all subjects.