NeuroNetwork 16

Project titel: Diagnostic Glove: Disease diagnosis in daily life from wearable kinematics Project leader: Esther Kühn, Elena Azañón, Stefanie Schreiber, Christoph Reichert

Clinicians suffer from the limited meaningfulness of patient data acquired in clinical settings. Not only is the number of available tests limited, the acquired data are also influenced by motivational, emotional, and subjective factors, such as doctor-patient relationships or a physician’s experience. In addition, quantitative data on the patients’ real life behavior in a familiar environment are usually not available. This is a particular problem for sensorimotor disorders, where accurate diagnosis depends on knowledge on real life motor functions. Here, we propose to use recent advances in basic research on the tracking and classification of hand movements to develop a novel medical product (the Diagnostic Glove) that helps clinicians to diagnose, track, and classify sensorimotor disorders of the upper limbs. We will start with a medical problem that is common yet difficult to solve: the distinction of certain upper limb muscular patterns between lower motor neuron dominant (LMND) amyotrophic lateral sclerosis (ALS), inclusion body myositis (IBM), and brachial monomelic amyotrophy Hirayama (MMA). All three diseases are characterized by upper limb motor deficits, which are, however, difficult to differentiate early in disease progress applying the “clinical view” only and with the motor tests currently available. This translational proof-of-concept project will (i) show that the Diagnostic Glove can be used to reliably categorize motor movements relevant for clinical diagnosis, (ii) develop algorithms to reliably distinguish between ALS, IBM, and MMA, and (iii) receive a patent for this software as starting point to commercialize the product. This project is part of a trend in the Medical Sciences, where novel tools allow tracking patient behavior at home. This “Medicine to take away” is expected to found novel treatment strategies that are based on Big Data, Machine Learning, and analysis algorithms for a quantitative, evidencebased medicine.