ViT models adapted for cardiac MR classification using self-supervised learning

Researchers have developed a self-supervised contrastive learning method to adapt Vision Transformer (ViT) models for cardiac MR sequence classification. Pretrained ViT models showed poor transferability to medical imaging, but the new adaptation strategy significantly improved performance. The adapted model achieved an AUC exceeding 0.75 on common cardiac MR sequences and demonstrated generalization to external datasets like BraTS and ADNI. AI

IMPACT This research could improve diagnostic accuracy in cardiac MR imaging by enabling more effective use of advanced AI models.

RANK_REASON The cluster contains an academic paper detailing a new research methodology for adapting AI models to a specific domain. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• ADNI
• Vision Transformer

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →