An intelligent sport shoe to prevent ankle inversion sprain injury

Materials and methods (1) Sensing: Five subjects performed various sporting motions with data collected from a plantar pressure system to reconstruct the ankle supination torque determined from a motion capture system with a force plate. Validation test on another five subjects was conducted. (2) Identification: Six subjects performed simulated subinjury and non-injury trials with the dorsal foot kinematics measured by 8 wearable motion sensors. Data was used to train a support vector machine to establish a mathematics algorithm for identification, which was validated on another 6 subjects, with an expected accuracy of 90%. An uni-axial gyrometer was placed at the position with the best accuracy for identifying ankle sprain hazard, with a threshold suggested from a database of ankle inversion velocity from real injury incidents, sub-injury trials and non-injury motions. (3) Correction: Myoelectric stimulations with different delay time (0, 5, 10 and 15ms) were delivered to the peroneal muscles of 10 subjects performing unanticipated subinjury trials in a laboratory. The effect was quantified by the heel tilting angle and its velocity as determined by a motion analysis system.