B0761
Title: A Riemann manifold model framework for longitudinal changes in physical activity patterns
Authors: Jingjing Zou - University of California, San Diego (United States) [presenting]
Chongzhi Di - Fred Hutchinson Cancer Center (United States)
Loki Natarajan - University of California San Diego (United States)
Abstract: The wide usage of wearable accelerometer-based activity trackers in recent years has provided a unique opportunity for in-depth research on physical activity (PA) and its relationship with health outcomes and interventions. Past analysis of activity tracker data relies heavily on aggregating minute-level PA records into day-level summary statistics, in which important information on diurnal PA patterns is lost. We propose a novel functional data analysis approach based on the theory of Riemann manifolds for modeling PA records and longitudinal changes in PA temporal patterns. We model smoothed minute-level PA of a day as one-dimensional Riemann manifolds and longitudinal changes in PA in different visits as deformations between manifolds. With the proposed approach, we conduct comprehensive analyses of data from two clinical trials, focusing on the effect of interventions on longitudinal changes in PA patterns and how different patterns of changes in PA influence weight loss, respectively. For both studies, important modes of variation in PA were identified to be significantly associated with lifestyle interventions/health outcomes.