Walking is paramount in the activities of the daily life of the human being and can be altered in neuro-musculoskeletal disorders. It is therefore essential to identify and better understand abnormal gait patterns linked to these disorders and to treat them. Clinical gait analysis (CGA) is a medical examination that objectively identifies gait deviations. CGA includes video recordings, spatiotemporal parameters, kinematics, kinetics and electromyography of a patient's gait. The objective of this article is to critically review conventional methods to identify gait deviations and methods used to improve understanding of gait deviations. After addressing conventional CGA, this article provides new avenues to better identify gait deviations based on existing literature. Different approaches to improve the understanding of gait deviations are presented. They are based on pathological models, experimental procedures, numerical simulations, anthropomorphic robots and animal models.