Related Experiment Video
Updated: Jul 10, 2026
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
AdapHBNA: Adaptive hierarchical spatio-temporal brain network analysis for brain disease detection
Junze Wang1, Guangyu Wang2, Dequan Meng1
1School of Computer and Artificial Intelligence, Shandong Jianzhu University, Jinan Shandong, 250101, China.
Summary
AdapHBNA improves brain disorder diagnosis by analyzing brain networks at multiple scales. This adaptive hierarchical framework captures complementary insights for better prediction of Autism Spectrum Disorder, Cognitive Impairment, and Major Depressive Disorder.
Area of Science:
- Neuroscience
- Machine Learning
- Medical Imaging
Background:
- Resting-state functional MRIs (rs-fMRIs) are used for brain network analysis (BNA) to understand brain disorders.
- Existing BNA methods often use static networks at a single spatial scale, ignoring the brain's hierarchical structure.
Purpose of the Study:
- To propose AdapHBNA, an Adaptive Hierarchical spatio-temporal Brain Network Analysis framework.
- To improve brain disorder diagnosis by capturing multi-scale, hierarchical brain representations.
Main Methods:
- Incorporated feature-channel-guided temporal and Modular Brain Clustering (MBC)-driven spatial hierarchical learning.
- Utilized Mamba and Graph Neural Networks for spatio-temporal encoding.
- Developed an end-to-end framework for adaptive multi-scale learning and fusion.
Main Results:
- AdapHBNA demonstrated superior performance compared to state-of-the-art methods.
- The framework effectively leveraged complementary diagnostic insights across multiple scales.
- Validated on Autism Spectrum Disorder (ABIDE), Early Mild Cognitive Impairment (ADNI), and Major Depressive Disorder (REST_MDD) datasets.
Conclusions:
- AdapHBNA offers a novel approach to brain disorder diagnosis by embracing hierarchical and multi-scale network properties.
- The adaptive spatio-temporal framework enhances the identification of biomarkers for neurological and psychiatric conditions.
