DualMiT-Net enhances breast mass segmentation with dual-branch AI

Researchers have developed DualMiT-Net, a novel deep learning model for segmenting breast masses in mammograms. This dual-branch network combines a focused view of the mass with a broader context of the surrounding breast tissue. The local branch utilizes a Mix Transformer (MiT-B5) for detailed shape and texture analysis, while the global branch employs an EfficientNet-B5 to capture contextual information. Experimental results on the CBIS-DDSM dataset show DualMiT-Net achieving a mean Dice coefficient of 0.9375 and a mean Intersection over Union of 0.8834, outperforming several baseline models. AI

IMPACT This model could improve the accuracy and consistency of breast mass segmentation, aiding in earlier and more precise diagnosis.

RANK_REASON The cluster contains a research paper detailing a new AI model for medical image segmentation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• CBIS-DDSM
• Curated Breast Imaging Subset
• Digital Database for Screening Mammography
• DualMiT-Net
• EfficientNet-B5
• Gábor
• MiT-B5
• Mix Transformer

• paper
• other

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