Wrist-Forearm Differential Kinematics Using Monocular Vision Capture

Each year the number of patients requiring motor rehabilitation for the upper limb, particularly for the wrist and forearm, increases, due to a loss in their fine motor skills. Although physiotherapists assist them, the progress evaluation remains subjective. Modern technology can be used to create solutions to provide a quantitative description allowing objective metrics evaluations for motor recovery of the patients, which typically yield complex and expensive solutions. This paper presents a methodology to obtain the differential kinematic trajectories of the wrist and forearm performed by a patient, through video capture of its movement, using the open-source MediaPipe library. The angles of the wrist and forearm are obtained through inverse kinematics analysis using hand-specific landmarks. A second-order filter estimates the angular trajectories for position, velocity, and acceleration. The study was performed with basic movements used in rehabilitation: adduction-abduction, flexion-extension, pronation-supination, and complex movements such as circumduction with the participation of six healthy patients. The results show that it is possible to obtain estimates of differential wrist trajectories. The applications of this methodology in the fields of biomechanics and rehabilitation are discussed.