“…However, in both cases [21,22], no machine learning model was developed to discriminate between the two populations of interest. Wu et al [49] built a neural network based on the features extrapolated from the 3D phase space reconstruction of the knee mechanics during the internal-external rotation, and flexion-extension, antero-poster, and proximal-distal translations while walking on a treadmill. On the other hand, Richter et al [50] considered a wide range of biomechanical features, including ground reaction forces and impulses, center of mass velocity and power in pelvis, hip, knee, and ankle, as well as joint angles of the ankle, knee, hip, pelvis, thorax, and thorax on pelvis in sagittal, frontal, and transversal planes, joint angular velocities of the ankle, knee, hip, pelvis, thorax, and thorax on pelvis in sagittal, frontal, and transversal plane, joint powers, moments, work, and impulse of ankle, knee, hip, and pelvis in sagittal, frontal, and transversal plane, time, and the rotation foot angle to pelvis.…”