Designing Wearable Sensor-Based Analytics for Quantitative Mobility Assessment

Wearable sensors are gaining traction in various healthcare domains, including patient mobility assessment performed in rehabilitation environments. Typically, clinical observations by therapists are used to characterize patient movement abilities and progress. More precise quantitative measurements of patient performance can be collected with wearable inertial sensors. Highly useful quantitative information and visual presentations of wearable sensor data are critical in gaining therapist acceptance of the technology and improving the therapy experience for patients. To bridge the gap between design of mobility monitoring technology and actual use of the technology, we report responses from interviews conducted with physical therapy providers at an inpatient rehabilitation facility. The information presented during the interviews includes results from our wearable sensor-based mobility assessment algorithms. Our smart computing algorithms utilize wearable sensor data to extract patient movement metrics, train clinical assessment prediction models, and visualize the data. The interview results indicate therapy providers are interested in using wearable sensors and wearable sensor- based metrics, prediction tools, and visualizations while they provide therapy services for their patients. Based on therapist feedback, we suggest future research directions that may increase the clinical utility and adoption of wearable sensor systems and data visualization for mobility assessment.