Heterogeneous Graph Neural Networks for Evolving Multimodal Data

Integrate relational attributes + topological structure over time

Oct 1, 2024 1 min read Research

Heterogeneous Graph Neural Networks

Overview

Heterogeneous Graph Neural Networks integrate relational attributes and topological structure over time to analyze complex, evolving multimodal data.

Key Features

Heterogeneous Graphs: Support for multiple node and edge types
Temporal Dynamics: Tracking structure evolution over time
Multimodal Integration: Combining diverse data sources
Explainable Predictions: Understanding graph-based decisions

Applications

Brain Disease Detection: Neurological disorder diagnosis
Metabolic Mechanisms: Understanding biological pathways
Molecular Analysis: Drug-target interaction prediction
Drug Discovery: Identifying potential therapeutic compounds

Deliverables

Heterogeneous Graph Neural Network Explainability Framework
Interactive Graph expLainability with Uncertainty (IGLU)
HiTCO (High-Fidelity Temporal Representation with Combined-Objective Training)

Impact

These tools enable researchers and clinicians to better understand complex biological systems and make data-driven discoveries.

Graph Neural Networks Multimodal Data Healthcare

Lu Wang

Assistant Professor of Biomedical Engineering

My research interests are developing and applying Machine Learning, Data Mining and Statistical methods (e.g., Multi-task Learning, Survival Analysis, Clustering, Risk Factor Analysis and Causal Discovery) on various data including gene expression, electronic health/medical records, and DNA sequencing reads for both cognitive disorders (e.g., delirium, Alzheimer’s disease, dementia, major depressive disorder) and chronic diseases (e.g., cancer, obesity, hypertension). Inspired by the human factors approach, she also designs and develops Human-Centered Artificial Intelligence tools for users to integrate, visualize, analyze, and interpret health data in order to improve the interoperability and accessibility of AI-assisted healthcare decision support.