A Finger Motion Monitoring Glove for Hand Rehabilitation Training and Assessment Based on Gesture Recognition

Hand rehabilitation training and assessment require the frequent participation of professional doctors, which is time-consuming, laborious, and sometime non-quantitative. To liberate both doctors and patients, a finger motion monitoring glove is expected to give a helping hand by monitoring the speed and state of hand movement in real time. However, the existing finger motion monitoring gloves suffer from low sensitivity, narrow detection range, and a lack of fabrication consistency. Herein, a highly sensitive finger motion monitoring glove with an intrinsic surface microstructure is fabricated based on the stable printing method, and hand rehabilitation training and assessment are realized with the help of the machine learning method. The printed strain sensors achieve a high strain sensitivity (gauge factor (GF) reaches 100.69 at 30%–50% strain), wide response range (0.1%–50%), fast response/recovery speed (71.43/178.49 ms), satisfied durability (function properly after 5000 cycles), and low batch-to-batch variation (within 0.10). These advantages enable the printed sensors to monitor finger movements quickly and comprehensively, thus making it practicable for hand rehabilitation training and assessment. This work provides a simple and stable method to obtain a highly sensitive finger motion monitoring glove, which is expected to make hand rehabilitation training and assessment more convenient and reliable.