Abstract: Disclosed examples relate to a system and method for creating and integrating self-supervised learning (SSL) digital movement biomarkers into electronic health records (EHRs) for enhanced diagnostic support and treatment monitoring, particularly for rare diseases and movement disorders. Example systems employ SSL techniques to analyze video data, tracking body key points and extracting kinematic data without the need for human-labeled annotations. Disclosed examples generate digital movement biomarker profiles that quantify symptoms and track disease progression, facilitating the objective measurement of drug impacts on symptoms. Example systems include components for video processing, compliance checking, pose estimation, context analysis, kinematic measurement, and classification.

Abstract: In one example, a system for measuring body movement in a movement disorder disease is provided. The system may comprise at least one processor and a memory storing processor executable codes, which, when implemented by the at least one processor, cause the system to perform operations comprising, at least receiving a video including a sequence of images and detecting at least one object of interest in one or more of the images. Feature reference points of the at least one object of interest are located, and a virtual movement-detection framework is generated in one or more of the images. The operations may include detecting, over the sequence of images, at least one singular or reciprocating movement of the feature reference point relative to the virtual movement-detection framework and generating a virtual path tracking a path of the at least one detected singular or reciprocating movement of the feature reference point.