Rhamba framework integrates attention and Mamba for fMRI self-supervised learning

Researchers have developed Rhamba, a novel framework for self-supervised learning on resting-state fMRI data. This framework combines region-aware masking with hybrid Attention-Mamba architectures to improve the analysis of neuroimaging data. Experiments on the ABIDE dataset and fine-tuning on COBRE and ADHD-200 datasets demonstrated that Rhamba, particularly the Mamba-Attention configuration, achieved superior performance in discriminating between conditions like schizophrenia and ADHD compared to existing methods. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Introduces a new framework for neuroimaging analysis that could improve diagnostic capabilities for neurological disorders.

RANK_REASON This is a research paper detailing a new framework for analyzing fMRI data using self-supervised learning. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Rhamba
• ABIDE
• COBRE
• ADHD-200
• Mamba
• Attention
• Integrated Gradients

• paper
• other