DESENVOLVIMENTO E VALIDAÇÃO CONCORRENTE DE UM SISTEMA DE ANÁLISE DE MOVIMENTOS EM TRÊS DIMENSÕES BASEADOS EM CÂMERAS RGB

Abstract

The prescription of therapeutic exercises to be performed at home is a common practice adopted by physical therapists. However, adherence to this type of exercise is, in general, low, which may restrain patients’ evolution. New technologies based on virtual reality and augmented reality offer promising perspectives in an attempt to increase adherence to home rehabilitation programs through the analysis of movement and its comparison with the normal prescribed treatement. However, most of the solutions, such as Vicon 3D (Vicon Motion Systems Ltd., Oxford, UK) and Qualisys Track Manager (QTM), are expensive and not portable. Therefore, restricted to specialized centers. RGB cameras are accessible and available on various devices, such as cell phones and webcams and can become a good alternative for data collection, as long as the data processing allows the prediction of 3D variables from 2D information. Therefore, the objective of this work is to (1) develop and (2) concurrently validate a 3D motion capture (MOVA3D) and 3D motion analysis system from 2D RGB cameras. (1) For the development of this analysis, Intel RealSense cameras (Intel 435i with the depth function disabled) were used to capture the data, and a proprietary software capable of automatically inferring the joint centers in 3D and performing the angular kinematic calculations of interest was developed for such analysis. (2) For concurrent validation, 10 subjects aged, between 22 and 50 years, performed squatting, hip flexion and abduction activities, with the aim of measurement the variables hip abduction/adduction angle and hip flexion/extension. Knee flexion/extension angle was also captured simultaneously by the MOVA3De systems as well as by the QTM (gold standard). The mean error of each variable and its degree of association were calculated using Pearson's Correlation Index. Keywords: Telerehabilitation; Motion Analysis; Capture System; QualisysTrack Manager.