Related Experiment Videos
Individualized dynamic latent factor model for multi-resolutional data with application to mobile health.
1Department of Statistics, University of California, Irvine, Donald Bren Hall 2219, Irvine, California, 92697, U.S.A.
Biometrika
|June 29, 2026
Summary
This study introduces a new model to accurately fill in missing health data from smartphones and wearables. The individualized dynamic latent factor model improves tracking of individual health status using irregular time series data.
Area of Science:
- Biomedical Informatics
- Data Science
- Time Series Analysis
Background:
- Mobile health (mHealth) utilizes smartphones and wearables for health tracking.
- mHealth generates heterogeneous, multi-resolution, and irregular time series data.
- Existing methods struggle with integrating and interpolating such complex data.
Purpose of the Study:
- To propose an individualized dynamic latent factor model.
- To address challenges in handling irregular multi-resolution time series data in mHealth.
- To enable accurate interpolation of unsampled measurements.
Main Methods:
- Developed an individualized dynamic latent factor model.
- Mapped multi-resolution data to a latent space for integration.
- Utilized B-spline approximation methods for interpolation.
Main Results:
- The proposed model effectively integrates multiple irregular time series and subjects.
- Demonstrated superior performance in interpolating unsampled measurements compared to existing methods.
- Validated through simulation studies and real-world smartwatch data.
Conclusions:
- The individualized dynamic latent factor model offers a robust solution for mHealth data.
- Enables accurate analysis of heterogeneous longitudinal health information.
- Advances the field of mHealth data processing and analysis.
Related Concept Videos
01:23Mechanistic Models: Compartment Models in Individual and Population Analysis
Mechanistic models are utilized in individual analysis using single-source data, but imperfections arise due to data collection errors, preventing perfect prediction of observed data. The mathematical equation involves known values (Xi), observed concentrations (Ci), measurement errors (εi), model parameters (ϕj), and the related function (ƒi) for i number of values. Different least-squares metrics quantify differences between predicted and observed values. The ordinary least squares (OLS)...
01:21Dimensions of Health and Illness
The factors influencing the health-illness continuum can be internal or external and may or may not be under conscious control. They are related to the following eight human dimensions, and each dimension is interrelated to one other.