Exercise Recognition for Kinect-based Telerehabilitation

An aging population and increasing survival of diseases and traumas that leave physical consequences are difficult aspects in the context of an efficient health management. This is why telerehabilitation systems are being developed, to allow monitoring and support of physiotherapy sessions at home, which could reduce healthcare costs while also improving the quality of life of the users. The objective was to develop a Kinect-based algorithm that provides highly accurate real-time monitoring of physical rehabilitation exercises and that also provides an approachable interface oriented both to users and physiotherapists. The two main constituents of the algorithm are the posture classification method and the exercises recognition method. The exercises consist of series of movements. Each movement is composed of an initial posture, a final posture and the angular trajectories of the limbs involved in the movement. The algorithm was designed and tested with datasets of real movements performed by volunteers. The paper also explains how the optimal values for the trade-off values for posture and trajectory recognition were obtained. Two relevant aspects of the algorithm were evaluated in the tests, classification accuracy and real-time data processing. 91.9% accuracy in posture classification and 93.75% accuracy in trajectory recognition was achieved. In addition, the algorithm was tested for and able to process the data in real-time. The study found that the algorithm could process more than 20,000 postures per second and all the required trajectory data-series in real-time, which in practice guarantees no perceptible delays. Later on, two clinical trials with real patients that suffered shoulder disorders were conducted. An exercise monitoring accuracy of 95.16% was achieved.

This study presents an exercise recognition algorithm that handles the data provided by Kinect efficiently. The algorithm has been validated in a real scenario where its suitability was verified. Moreover, positive feedback was received from both users and the physiotherapists who took part in the tests.