A method comprising: receiving, with respect to each of a plurality of subjects, data comprising at least one of: (i) a data representing center of pressure (COP) of a subject during at least on gait phase; (ii) a data representing center of gravity (COG) of the subject during at least on gait phase; and; (iii) a data representing posture of the subject during at least on gait phase; processing the data to extract a plurality of features representing the data, at a training stage, training a machine learning model on a training set comprising all of the plurality of features; and at an inference stage, applying the trained machine learning model to a target set of the features obtained for ma target subject, to classify posture, proprioception and/or kinesthesis of the target subject.