Explore a cutting-edge approach to integrating multimodal brain networks in this 49-minute conference talk by Moo K. Chung. Delve into a novel topological learning framework that utilizes persistent homology to integrate networks with different topologies obtained from diffusion and functional MRI. Learn about the Wasserstein distance-based topological loss on graph filtrations, which enables efficient topological computations and optimizations. Discover the application of this framework in a twin brain imaging study to determine the genetic heritability of brain networks. Gain insights into the methodology, including network data analysis, graph mutation, decomposition, and matching techniques. Access the related paper and Matlab codes for hands-on understanding of the concepts presented.
Topological Learning and Its Application to Multimodal Brain Network Integration
Applied Algebraic Topology Network via YouTube
Overview
Syllabus
Introduction
Data
Network Data
Approach
Position Diagram
Graphitization
Graph Mutation
Graph Structure
Decomposition
Graph Matching
Average Operation
Subject Level
Loss Function
Group Level
Next Project
Questions
Taught by
Applied Algebraic Topology Network
