Proprietary: This proposal includes trade secrets and other proprietary or confidential information of Highland Instruments and is being provided for use by the National Institutes of Health (NIH) for the sole purpose of evaluating this SBIR proposal. No other rights are conferred. This proposal and the trade secrets and other proprietary or confidential information contained herein shal further not be disclosed in whole or in parts, outside of NIH without Highland Instrument's permission. This restriction does not limit the NIH's right to use information contained in the data if it is obtained from another source without restriction. This legend applies to the Abstract, Specific Aims, Research Plan (al components), Commercialization Plan, and Human Subject's Sections of this proposal. Abstract The AHA Stroke Rehabilitation guideline states that in the current fiscal climate, ?the provision of comprehensive rehabilitation programs with adequate resources, dose, and duration is an essential aspect of stroke care and should be a priority? [1]. While near and long-term rehabilitation goals should be guided by patient baseline motor status and potential for motor recovery [2-12], acute motor assessment and prognostication remain a clinically difficult task [13, 14]. Conventional clinical assessments (e.g. NIH Stroke Scale, Fugl-Meyer (FM) Scale) [15, 16] that power prognosis are highly dependent on the initial severity and care provider/point of care, and often reduced further to even coarser prognostic scales (e.g. Orpington Prognostic Scale (OPS)) [17]. Overall they lack the level of sophistication required to predict motor recovery and tailor rehabilitation due to ceiling effects, omission of fractionated and complex distal movements, and/or unequal weighting of the two extremities in assessments [18]. Furthermore, many survivors do not even receive comprehensive assessments prior to discharge, and telestroke approaches which are being implemented to address such issues are limited in their scope of care and still not fully developed for functional assessments [19-23]. This is critical because ?all patients benefit from a formal assessment of the patient?s rehabilitation needs prior to discharge? [24]. To address these and other limitations the AHA guidelines specifically call for the development of ?computer-adapted assessments for personalized and tailored interventions?, ?newer technologies such as?body-worn sensors?, and ?better predictor models to identify responders and nonresponders? [1]. To address these limitations, we propose to develop a system of multimodal, integrated sensors (i.e., motion capture cameras, force sensors, accelerometers, and gyroscopes) that will interface with a patient for recording their movement; a software suite with signal processing and data fusion algorithms to reduce data dimensionality and provide kinematic/kinetic based evaluations; and system software to track, classify, and predict patient disease severity. We will test this computational Integrated sensor-based Motion Analysis Suite (IMAS) in 30 stroke patients undergoing IMAS directed motor evaluations (repeated 4 times in a 2 day period) within 5 days post stroke and 6-8 weeks post stroke (after having undergone rehabilitation) focused on assessments identified as important to patient independence, activities of daily living, and ability to return to work. Quantitative kinematic/kinetic metrics descriptive of motor behavior will be derived from IMAS sensor recordings to characterize subjects? motor performance (e.g., joint displacement, velocity, acceleration, jerk, and movement quality measures). Then, we will investigate the statistical relationships between the IMAS collected data (and with additional clinical data gathered during the patient assessments including demographics; imaging derived information; and FM, OPS, and Barthel index scores) to build statistical models to: 1. extract a low dimensional representation of disease state, 2. predict the FM scale, and 3. predict likelihood of motor recovery following treatment.
Stroke is a major cause of death and the leading cause of disability. In the clinic, the lack of quantitative, objective methods for outcome assessment and outcome prediction considerably hamper management of the disease, ultimately leading to sub-optimal strategies for managing patient care. This study will develop a motion analysis suite for quantitative, objective outcome assessment and prediction in stroke patients.
