Generalizable Representation Learning for fMRI-based Neurological Disorder Identification

2024-12-16

arXiv.org

AI Summary

To address poor generalization in neurologic disorder identification caused by fMRI data scarcity and high heterogeneity of functional patterns, this paper proposes a novel representation learning framework integrating self-supervised pretraining with model-agnostic meta-learning (MAML). It is the first to jointly leverage time-series masked reconstruction, instance-discriminative contrastive learning, and dynamic graph convolutional networks (dGCNs) to model time-varying functional connectivity—enabling transferable brain functional representation learning solely from healthy control data. The framework achieves cross-domain generalization from healthy controls to few-shot clinical tasks across four heterogeneous datasets involving multiple rare disorders (ASD, ADHD, schizophrenia, and Parkinson’s disease), improving average classification accuracy by 5.2–11.7%. Under few-shot settings (≤20 samples per disorder), it attains an AUC of 0.89. The implementation code is publicly available.

Technology Category

• Machine Learning: Representation Learning
• Computer Vision: Representation Learning for Vision
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Despite the impressive advances achieved using deep learning for functional brain activity analysis, the heterogeneity of functional patterns and the scarcity of imaging data still pose challenges in tasks such as identifying neurological disorders. For functional Magnetic Resonance Imaging (fMRI), while data may be abundantly available from healthy controls, clinical data is often scarce, especially for rare diseases, limiting the ability of models to identify clinically-relevant features. We overcome this limitation by introducing a novel representation learning strategy integrating meta-learning with self-supervised learning to improve the generalization from normal to clinical features. This approach enables generalization to challenging clinical tasks featuring scarce training data. We achieve this by leveraging self-supervised learning on the control dataset to focus on inherent features that are not limited to a particular supervised task and incorporating meta-learning to improve the generalization across domains. To explore the generalizability of the learned representations to unseen clinical applications, we apply the model to four distinct clinical datasets featuring scarce and heterogeneous data for neurological disorder classification. Results demonstrate the superiority of our representation learning strategy on diverse clinically-relevant tasks. Code is publicly available at https://github.com/wenhui0206/MeTSK/tree/main

Problem

• Overcoming fMRI data scarcity for neurological disorder identification
• Improving generalization from normal to clinical fMRI features
• Enhancing representation learning for heterogeneous neurological data

Innovation

• Meta-learning with self-supervised representation learning
• Generalizing from normal to clinical fMRI features
• Leveraging control data for clinical task generalization