Prediction of Patient-Reported Physical Activity Scores from Wearable Accelerometer Data: A Feasibility Study

Abstract

Many diseases are characterized by limitations in mobility, including a wide range of musculoskeletal and neurological conditions. Reduced mobility impacts a patients ability to perform activities of daily living, which in turn reduces health-related quality of life. Mobility can be assessed by collecting patient-reported outcome scores from standardized questionnaires and by directly measuring physical activity parameters from wearable accelerometer data. In this work, we explored the relationship between subjectively and objectively measured mobility by training machine learning models to predict patient responses based on features derived from realworld acceleration data. Our method achieved up to 82% accuracy using a random forest classifier and set the basis to develop novel data-driven digital biomarkers for objective, quantitative and more frequent evaluation of patients' mobility.

Continuous Digital Monitoring of Walking Speed in Frail Elderly Patients: Noninterventional Validation Study and Longitudinal Clinical Trial

“…Directly capturing context, behavior, and activities of daily living using ambient [34] and smartphone technologies [24] will help us to better relate objective measurements to events and transitions in a patient’s life. Ultimately, we aim to build on the progress presented here in terms of establishing accuracy, feasibility, and leveraging context to make meaningful comparisons to further build the link between performance, behavior, subjective perceptions of health, and clinical outcomes [35] and to predict long-term changes in health.…”

Evaluation, Acceptance, and Qualification of Digital Measures: From Proof of Concept to Endpoint

“…Once a guiding COI has been established, initial investigations typically seek to de-risk further investment by demonstrating, on a smaller scale, that the proposed approach and the expected benefits are more than just hypothetical in a proof of concept [46]. At this stage it may not be immediately obvious which data streams are most informative, requiring a head-to-head design comparing data from several sensors [47], or it may not even be clear that person-generated health data (PGHD) can even address the question at all [48, 49]. The available technologies may not be sufficiently mature to deploy into clinical development (e.g., they may lack sufficient security provisions or not meet General Data Protection Regulations (GDPR) which would exclude use in European sites, etc.)…”

“…One possible next step is to replace, prompt, or augment COA items and subscales with digital measures derived from PGHD. It has been shown that PGHD can be used to accurately predict subjective mobility [47], stress [61, 62], and fatigue [63] outcomes, demonstrating the potential of this idea.…”

The Path Forward for Digital Measures: Suppressing the Desire to Compare Apples and Pineapples

“…While objective wearable data often demonstrate lower associations with PROM scores of purported similar concepts (sleep quality, scratching) than layman expectations would posit [42][43][44][45], other areas have found PROMs to be highly aligned with sensor-derived variables [46,47]. For instance, Bahej et al [47] used objective measures to forecast PROs and achieved accuracies of around 70-80% for predicting subjectively reported mobility. Similar analyses have found equally promising results across objective and subjective outcomes in cognition [48,49] and stress [41].…”