All Times EDT
Keywords: Acceleromery, physical activity, objective, digital biomarkers
Wearable devices are increasingly used in observational and clinical trials because they can provide objective, continuous, high quality data that has the potential to produce digital biomarkers that are sensitive to changes in disease status status and to treatment modifications. However, transforming the large, heterogenous data produced by wearable devices into digital biomarkers requires open source, reproducible, easy to understand analytic pre-processing pipelines. In this talk I will discuss the necessary steps for transforming accelerometer data into digital biomarkers and I will describe the main bottlenecks and provide potential solutions for FDA approval of accelerometer-based digital biomarkers.