JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 308
Type: Contributed
Date/Time: Tuesday, August 11, 2015 : 8:30 AM to 10:20 AM
Sponsor: Mental Health Statistics Section
Abstract #317406
Title: Understanding Activity Patterns via Functional Data Approach and Quantifying Similarities Across Species
Author(s): Haochang Shou* and Vadim Zipunnikov and Lihong Cui and Kathleen Merikangas and Sonja Greven and Ciprian Crainiceanu
Companies: University of Pennsylvania and Johns Hopkins Bloomberg School of Public Health and National Institute of Mental Health and National Institute of Mental Health and LMU and The Johns Hopkins University
Keywords: accelerometer ; activity patterns ; principal component analysis ; multilevel model ; functional data ; mood disorder
Abstract:

Changes in activity have long been known as a core feature of mental disorders. Several studies using objective measures of activity with accelerometers have shown differential daytime activity levels for people with bipolar and major depressive disorders. However, most analyses are based on simple statistics and ignore the time-dependent effect. We develop a functional data analysis approach and use principal components (PCs) to define the intrinsic daily activity patterns. Our methods account for multilevel correlations in the design, adapt to high-dimensionality and adjust for potential confounders. We apply the methods to: 1) NIMH family study of spectrum disorders with both the minute-by-minute activity measures and mood/energy rating scores; 2) macaque models of Parkinson's disease with three phases of study design; 3) transgenic mice models of Autism. The methods enable us to identify difference in activity behaviors across disease subgroups within each study. Moreover, by using a bootstrap-based method to quantify similarities of the space expanded by individual sets of PCs, we are able to evaluate the translational potentials of animal models for certain human diseases.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home