ABSTRACT
Neurorehabilitation is a highly individualized process, in which a multidisciplinary clinical team attempts to address each patient’s unique impairments, priorities, and recovery profile over time. The heterogeneous nature of this process makes it challenging to carry out randomized controlled trials and therefore to extract strong evidence for clinical best practices. It has been suggested that a practice-based evidence approach can move the field forward by helping to characterize the complex interplay between patient characteristics, interventions, environments, and outcome metrics. While collecting the rich data sets necessary to answer these questions is a daunting task, a number of evolving technologies will provide new opportunities to collect large-scale data about the neurorehabilitation of the upper limb. Examples include robotic rehabilitation devices, wearable sensors, and markerless motion capture systems. As large and complex data sets become increasingly prevalent in this setting, big data science will be vital to extracting the insights needed to optimize the outcomes of the neurorehabilitation process.
