Classification of Free-Living Body Posture with ECG Patch Accelerometers: Application to the Multicenter AIDS Cohort Study

Abstract

SummaryAs health studies increasingly monitor free-living heart performance via ECG patches with accelerometers, researchers will seek to investigate cardio-electrical responses to physical activity and sedentary behavior, increasing demand for fast, scalable methods to process accelerometer data. We provide the first published analysis of tri-axial accelerometry data from Zio XT patch and introduce an extension of posture classification algorithms for use with ECG patches worn in the free-living environment. Our novel extensions to posture classification include (1) estimation of an upright posture for each individual without the reference measurements used by existing posture classification algorithms; (2) correction for device removal and re-positioning using novel spherical change-point detection; and (3) classification of upright and recumbent periods using a clustering and voting process rather than a simple inclination threshold used in other algorithms. Methods were built using data from 14 participants from the Multicenter AIDS Cohort Study (MACS), and applied to 1, 250 MACS participants. As no posture labels exist in the free-living environment, we evaluate the algorithm against labelled data from the Towson Accelerometer Study and against data labelled by hand from the MACS study.