Towards Zero-Shot Task-Generalizable Learning on fMRI

2025-02-15

AI Summary

Task-based fMRI models suffer from poor cross-task generalization due to task diversity. To address this, we propose the Task-Aware Graph Attention Network (TA-GAT), a plug-and-play architecture that explicitly decouples a universal brain functional encoder from a task-context representation module. TA-GAT employs graph attention mechanisms to model dynamic functional connectivity among brain regions and incorporates task-conditioned embeddings to enable multi-task joint supervised learning. Crucially, it supports zero-shot transfer to unseen tasks. Evaluated across multiple neurocognitive fMRI datasets, TA-GAT achieves significant improvements in cross-task decoding performance. This work constitutes the first effort to explicitly separate task-agnostic brain representations from task-specific priors—establishing a novel paradigm for universal brain functional modeling.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Functional MRI measuring BOLD signal is an increasingly important imaging modality in studying brain functions and neurological disorders. It can be acquired in either a resting-state or a task-based paradigm. Compared to resting-state fMRI, task-based fMRI is acquired while the subject is performing a specific task designed to enhance study-related brain activities. Consequently, it generally has more informative task-dependent signals. However, due to the variety of task designs, it is much more difficult than in resting state to aggregate task-based fMRI acquired in different tasks to train a generalizable model. To resolve this complication, we propose a supervised task-aware network TA-GAT that jointly learns a general-purpose encoder and task-specific contextual information. The encoder-generated embedding and the learned contextual information are then combined as input to multiple modules for performing downstream tasks. We believe that the proposed task-aware architecture can plug-and-play in any neural network architecture to incorporate the prior knowledge of fMRI tasks into capturing functional brain patterns.

Problem

• Generalizing task-based fMRI across diverse designs
• Learning task-specific contextual information effectively
• Enhancing fMRI analysis with task-aware neural networks

Innovation

• Supervised task-aware network
• General-purpose encoder learning
• Task-specific contextual integration