Estimation of Movement in Physical Exercise Programs Using Depth Cameras: Validation Against a Gold Standard

Physical exercise has been identified as an effective intervention for preventing and treating frailty. Exercise programs such as ViviFil® have been designed to improve the functional capacities of frail older adults through strength, balance, flexibility, and endurance exercises. Proper supervision is crucial to ensure the safety and effectiveness of these programs, and with current technological advances, healthcare professionals have effective options for more detailed supervision. In this work, we propose a system comprised of depth cameras capable of estimating movements during the execution of physical exercises. The objective of this study is to verify the accuracy of the system in estimating the parameters characterizing the exercise, by comparing the proposed system with a high-resolution commercial optical motion capture system as gold standard. The study includes several phases: first, the system's ability to obtain the human skeleton is evaluated, then the kinematic characteristics of each exercise are estimated. For this purpose, five exercises from the ViviFil program have been selected for evaluation: squat, knee extension, shoulder elevation, ankle extension, and hip extension. Preliminary results show the capability of the proposed system to estimate the movement parameters of the exercises performed, providing a non-invasive and easy-to-use tool suitable for assessing the quality of exercise program execution.