Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 325 - Machine Learning Methods for Better-Informed Decision-Making in Heath Care
Type: Invited
Date/Time: Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
Sponsor: Health Policy Statistics Section
Abstract #314460
Title: High-Dimensional Directional Brain Network Analysis for Focal Epileptic Seizures
Author(s): Tingting Zhang*
Companies: University of Pittsburgh
Keywords: Epilepsy treatment; directional network; Bayesian model

The brain is a high-dimensional directional network system consisting of many regions that exert influences onto each other. The directional influence from one region to another is referred to as a directional connection. Epilepsy is a directional network disorder, as epileptic activity spreads from a seizure onset zone (SOZ) to many other regions after seizure onset. However, studying epileptic directional brain networks has been limited to low-dimensional directional networks due to the lack of efficient methods for identifying high-dimensional directional brain networks. To address this knowledge gap, we studied high-dimensional directional networks in epileptic brains by using a clustering-enabled multivariate autoregressive state-space model (MARSS) to analyze multi-channel intracranial EEG recordings of focal seizures. With the new MARSS, we revealed, for the first time, that many regions outside the SOZ cluster had no changes in directional connections, although these regions showed ictal activity. We used these high-dimensional network results to localize the SOZ for patients with focal seizures and achieved 100% true positive rates and < 3% false positive rates.

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

Back to the full JSM 2021 program