Automated quantification of motor dysfunction in multiple sclerosis using depth-sensing computer vision (P3.213)

Objectives: To test classification performance of an automated machine learning algorithm in predicting Expanded Disability Status Scale (EDSS) sub-scores and understand the clinical relevance of the dominant features identified in the movements. Background: Several measurement tools were developed to capture disease progression and disease activity in MS. The EDSS is the most widely used scale in MS but has been criticized for its low reliability. The ASSESS MS system is being developed to provide a more consistent and finer grained measurement of motor dysfunction using advanced machine learning techniques. It classifies movements recorded non-invasively using a 3D depth sensing camera. Methods: Pre-defined movements from standardized neurological assessment, covering upper extremities and trunk, as well as movements typical of activities of daily living (ADL), were recorded in 200 patients and 200 healthy volunteers. For all patients a standardized EDSS-assessment (the Neurostatus) was performed. Movement information extracted from the depth sensor recordings were pre-processed and analysed by an automated image analysis algorithm to correctly classify patients according to the Neurostatus sub scores. Results: We report quantification of the motor dysfunction of 200 MS patients in pre-defined recorded movements. Each movement’s EDSS sub-score was correctly predicted with average sensitivity/specificity scores in the range of 80[percnt]. The automated image analysis algorithm identified movement properties with highest discriminative power, which are medically plausible, confirming the validity of our approach. Conclusions: Automated quantification of motor dysfunction using depth-sensing computer vision enables an accurate and sensitive quantitative assessment of motor dysfunction in MS patients. With a finer classification to capture different degrees of movement disability in future, this approach may improve the evaluation of disability and disease progression in MS. Disclosure: Dr. D9Souza has received personal compensation for activities with Bayer AG, Teva, and Genzyme. Dr. D9Souza has received research support from the University of Basel. Dr. Burggraaff has received personal compensation for activities with Novartis. Dr. Kontschieder has received personal compensation for activities with Microsoft Research as an employee. Dr. Dorn has received personal compensation for activities with Novartis Pharma AG as an employee. Dr. Kamm has received personal compensation for activities with Biogen Idec, Novartis, Bayer Pharmaceuticals, Merck Serono, Genzyme, and Pfizer Inc. Dr. Tewarie has received personal compensation for activities with Novartis. Dr. Morrison has received personal compensation for activities with Microsoft Research as an employee. Dr. Vogel has received personal compensation for activities with Novartis as an employee. Dr. Sellen has received personal compensation for activities with Microsoft Research as an employee. Dr. Criminisi has received personal compensation for activities with Microsoft as an employee. Dr. Dahlke has received personal compensation for activities with Novartis as an employee. Dr. Uitdehaag has received personal compensation for activities with Biogen Idec, Novartis, EMD Serono, Teva Neuroscience, Genzyme Corporation, and Roche Diagnostics Corporation. Dr. Kappos has received personal compensation for activities with Actelion Pharmaceuticals.