The clinical environment all over the world has been changing rapidly, with a higher demand for healthcare availability without an increase in available resources to account for this demand. Thus, computational tools are required to help alleviate this burden and allow for patients to not only be seen faster but provide more objective data in the decision-making process. Orthopaedics is a particularly challenging field to implement novel computational tools given the subjective nature of many of the currently used diagnostic workflows. Additionally, much of the research in this field has been focused on biomedical image analysis which requires the patient to have radiographic imaging, which is an expensive and time-consuming task. In this chapter we propose the use of marker-less motion capture techniques to allow the collection of clinically relevant motion data from patients. This motion data can be used to initially quantify the severity of a problem and thus lift the load on the more expensive services and those with long waiting lists. In this work we showed that marker-less motion capture is sensitive enough to distinguish between a patient with knee pain and a patient without knee pain, this was done using a diagnostic block to remove the effects of pain and allow us to see the knee’s physiological capabilities. This motion data has also been further utilised in a biomarker identification pipeline to outline actions and kinematic metrics that are clinically relevant to both diagnosis and rehabilitative progress tracking.