|
Activity Number:
|
657
- Statistical Network Models for Brain Connectivity Data Analysis
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Imaging
|
|
Abstract #329880
|
Presentation
|
|
Title:
|
Statistical Inference of Brain Connectivity Networks: a Network Topology Based Method
|
|
Author(s):
|
Yishi Xing* and Shuo Chen
|
|
Companies:
|
and University of Maryland, School of Medicine
|
|
Keywords:
|
brain connectivity ;
networks;
network topology;
imaging statistics;
graph;
multiple testing
|
|
Abstract:
|
Major neurological and psychiatric diseases often impact cerebral functional connectivity systematically at the network level. A true disease-related connectivity network can be complex and well-organized, which is further obscured by being imbedded in a myriad of functional networks supporting normal brain functions. How to discover and statistically test true disease-related networks is a major challenge to current brain imaging research. We propose a novel, integrated machine-learning and statistical methodology to uncover hidden disease-related connectivity patterns by conceptualizing disease networks as "network-objects". This approach relies on leveraging the rich graph topological structure of disease-related network. We use simulation data and actual clinical data to demonstrate that, comparing to the conventional statistical methods, the proposed approach simultaneously reduces false positive and negative discovery rates and yields remarkable reproducibility on imaging discoveries. The network-object analysis and statistics approach may provide insights into how brain connectome is systematically impaired by brain illnesses.
|
Authors who are presenting talks have a * after their name.
