# A Dual-Stream Transformer with Self-Supervised Contrastive Training for fMRI-Based Autism Spectrum Disorder Classification

## TL;DR

This paper introduces a dual-stream Transformer model that improves autism classification by combining dynamic and global brain connectivity features using self-supervised learning.

## Contribution

The novel TwoTST model integrates ROI time series and PCC matrices with self-supervised pre-training and contrastive learning for better ASD classification.

## Key findings

- Contrastive learning, pre-training, and dual-stream structure improved mean AUC by 3–6%, 3–7%, and 3–4% respectively.
- Attention Pooling was identified as the optimal fusion strategy for combining dual-stream features.
- Relative parameter changes were higher in contrastive projection heads compared to TST modules during fine-tuning.

## Abstract

Background/Objectives: Autism Spectrum Disorder (ASD) diagnosis is difficult due to heterogeneity. Current Time-series Transformer (TST) methods cannot capture both dynamic and global brain connectivity simultaneously, which limits ASD classification performance. Methods: We propose TwoTST, a dual-stream Transformer that combines raw Region of Interest(ROI) time series and Pearson correlation matrices(PCC).We pre-train the two TST branches via self-supervised learning by randomly masking ROIs and PCC, use contrastive learning and fine-tuning for feature alignment, evaluate five fusion strategies, and analyze relative parameter changes during fine-tuning. Results: Experiments were conducted on the ABIDE I dataset using the CC200 atlas. Contrastive learning, pre-training, and the dual-stream structure improve mean AUC by 3–6%, 3–7%, and 3–4% respectively. Attention Pooling is the optimal fusion strategy. Relative parameter changes are 0.32–0.44 for TST modules and 0.31–1.45 for contrastive projection heads. Conclusions: TwoTST effectively integrates dynamic and global connectivity for ASD identification. The proposed design outperforms single-stream models and provides a reliable approach for neuroimaging-based disorder classification.