Characterizing Data-driven Rehabilitation After Stroke: A Scoping Review

This scoping review aims to develop a theoretical framework for data-driven rehabilitation interventions for people with stroke by systematically mapping the available literature. Current rehabilitation efforts are often hindered by repetitiousness, performance plateau or lack of progressive overload, limited engagement or adherence, and suboptimal exercise prescription. While advances in data science offer the potential to address these shortcomings, evidence-based and clinically relevant methodologies for implementing data-driven rehabilitation interventions remain nebulous and undefined. By synthesizing the existing evidence on data-driven approaches to stroke rehabilitation, this review will: (1) define core components (e.g., sensor-derived data, analytics, personalized prescriptions) of data-driven rehabilitation; (2) propose a standardized, evidence-based framework; and (3) outline potential care pathways for transform rehabilitation from experiential to objective, adaptive, and patient/context-specific in order to enhance clinical outcomes.