Deep learning improves brain disorder diagnosis from MRI data
HyperAMS-Net: Adaptive Multi-Scale Spatial Hypergraph Network for Brain Disorder Classification
Computer Vision and Pattern RecognitionMachine Learning
Summary
Classifying brain disorders like autism, depression, and Alzheimer’s from brain scans is difficult because people’s brains differ a lot and the brain patterns happen at many scales. The authors developed HyperAMS-Net, a new deep learning system that looks at these complex, multi-scale brain patterns more effectively. It uses different specialized methods to capture detailed brain information and how parts connect in complex ways. Tests on multiple brain disorder datasets showed it outperforms existing methods, especially thanks to its way of modeling connections with hypergraphs.
What this means in practice
- •For medical imaging teams: Improve brain disorder diagnosis accuracy by integrating multi-scale neuroimaging features with hypergraph-based neural networks.
- •For clinical ai developers: Develop advanced AI tools for brain disorder detection that use adaptive feature fusion and attention mechanisms from MRI data.$Commercial implications: Enables creation of more accurate diagnostic software for healthcare providers by exploiting novel deep learning architectures.
Authors
Proloy Kumar Mondal, Md Kamran Hussin Chowdhury, Hoi Leong Lee
Abstract
Accurate classification of brain disorders from neuroimaging data remains challenging because of substantial inter-subject heterogeneity and the complex multi-scale patterns present in functional connectivity and morphological representations. To address these challenges, we propose HyperAMS-Net, a deep learning framework for brain disorder classification using neuroimaging representations derived from resting-state functional MRI or structural MRI. HyperAMS-Net integrates adaptive multi-scale convolution, hypergraph attention, spatial-channel attention, and adaptive feature fusion. Specifically, adaptive multi-scale convolution learns data-driven weights over multiple receptive fields to capture complementary patterns at different scales. Hypergraph attention models higher-order dependencies among learned feature representations through node--hyperedge--node message passing, while spatial-channel attention enhances discriminative feature learning. Adaptive feature fusion further aggregates complementary information across parallel network branches. HyperAMS-Net is evaluated on three benchmark datasets spanning distinct brain disorders: ABIDE for autism spectrum disorder, REST-meta-MDD for major depressive disorder, and ADNI for Alzheimer's disease, using 5-fold stratified cross-validation. HyperAMS-Net achieves state-of-the-art performance across all evaluated datasets, attaining the highest accuracy and AUC among the compared methods. Ablation studies further demonstrate the contribution of each proposed component, with the largest performance degradation observed when hypergraph attention is removed.