Activity Number:
|
265
- Stochastic Processes in Medicine and Medical Engineering: Theoretical Foundations and Applications
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Medical Devices and Diagnostics
|
Abstract #322321
|
|
Title:
|
Adaptive Frequency Band Analysis for Multivariate Biomedical Time Series
|
Author(s):
|
Scott Alan Bruce* and Raanju Sundararajan
|
Companies:
|
Texas A&M University and Southern Methodist University
|
Keywords:
|
Frequency band estimation;
Local periodogram;
Locally stationary;
Multivariate time series;
Scan statistics;
Spectrum analysis
|
Abstract:
|
The frequency-domain properties of nonstationary multivariate biomedical time series often contain valuable information and are characterized through its time-varying spectral matrix. Practitioners seeking low-dimensional summary measures of spectral matrices often partition frequencies into bands and create collapsed measures of power within bands. However, standard frequency bands have largely been developed through manual inspection of time series data and may not adequately summarize spectral matrices. In this work, we propose a framework for data-driven frequency band estimation of multivariate time series that optimally summarizes the time-varying dynamics of the series. We develop a scan statistic and search algorithm to detect changes in the frequency domain. We establish theoretical properties of this framework and develop a computationally-efficient implementation. The validity of our method is also justified through numerous simulation studies. The methodology is then used to analyze the characteristics of functional magnetic resonance imaging time series data for autism spectrum disorder patients.
|
Authors who are presenting talks have a * after their name.