Transformer network enhances 3D LAD artery segmentation in CT scans

Researchers have developed a novel transformer-based network, NA-UNETR, designed for precise 3D segmentation of the Left Anterior Descending (LAD) artery in CT scans. This model incorporates Neighborhood Attention blocks to effectively capture both fine structural details and broader contextual information, addressing the challenges posed by the LAD's small size and low contrast. The framework also utilizes an uncertainty-guided optimization approach with a composite loss function to enhance overlap and boundary accuracy, particularly in scenarios with limited annotated data. AI

IMPACT This research could lead to more accurate cardiac substructure segmentation for radiotherapy planning, improving patient outcomes.

RANK_REASON The cluster contains an academic paper detailing a new model for a specific medical imaging task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• CT
• Left Anterior Descending (LAD) artery
• NA-UNETR
• nnU-Net
• Swin UNETR
• Transformer

• paper
• model release

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