Activity Number:
|
285
- Advances in Dimension Reduction and Model Selection for Statistically Challenging Data
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, July 31, 2018 : 8:30 AM to 10:20 AM
|
Sponsor:
|
IMS
|
Abstract #328304
|
|
Title:
|
Improved Selection of High-Dimensional Neuroimaging Biomarkers Associated with Neurodegenerative Disease Progression
|
Author(s):
|
Tanya Garcia* and Jeffrey S Morris
|
Companies:
|
Texas A&M University and The University of Texas M.D. Anderson Cancer Center
|
Keywords:
|
High-dimensional data;
Bayesian framework;
Functional data analysis
|
Abstract:
|
Neuroimaging data now plays an imperative role in understanding the progression of neurodegenerative diseases such as Huntington's, Parkinson's and Alzheimer's disease. A primary focus with such data is discovering and evaluating those neural regions most predictive of clinical outcomes such as age of disease onset. Identifying these neural regions is of high public health relevance as it aids to determine when and how therapeutic treatments should intervene. But handling the complex structure of high-dimensional neuroimaging data, extracting relevant regions and overcoming large censoring rates in event responses complicates image analysis. Overcoming these challenges requires developing advanced statistical tools for neuroimaging analysis which is the focus of this talk. We propose a new Bayesian framework for identifying brain regions of interest associated with disease onset. Our methods are shown to be consistent, and leads to new scientific discoveries in a study of Huntington's disease.
|
Authors who are presenting talks have a * after their name.