Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 80 - Contributed Poster Presentations: Mental Health Statistics Section
Type: Contributed
Date/Time: Monday, August 3, 2020 : 10:00 AM to 2:00 PM
Sponsor: Mental Health Statistics Section
Abstract #313643
Title: Hourly, Daily, Weekly or Monthly? Choosing the Right Data Time Granularity for Analysis of Digital Biomarker Trajectories
Author(s): Nicole Wakim* and Thomas Braun and Hiroko Dodge
Companies: University of Michigan and University of Michigan and University of Michigan
Keywords: digital biomarkers; longitudinal data analysis; time granularity; exploratory analysis
Abstract:

The use of digital biomarker data in dementia research provides the opportunity for frequent cognitive assessments that were not previously available using neuropsychological tests. However, the frequent measurements in digital biomarker data may be hard to store or computationally analyze. It is important to find a balance between computational efficiency and integrity of the data, which is reflected in a chosen time granularity, which is defined as the frequency at which measurements are observed or summarized. When faced with a time granularity decision, we focus on factors of the data that maintain the biomarker trajectory and improve analysis. These factors are explained in depth, and then applied to a real example involving walking speed in dementia patients. The example sheds light on typical problems data present and how we can use the above factors in exploratory analysis to choose an appropriate time granularity.


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

Back to the full JSM 2020 program