A Neighborhood Attention Transformer Network for Enhanced 3D Segmentation of the Left Anterior Descending Artery
2026-08-12 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors created a new computer method called NA-UNETR to better find a tiny and hard-to-see heart artery (LAD) in CT scans used for radiation therapy planning. Their method uses a special technique to look closely at small details and also understand the bigger picture in 3D images. They first trained their model on many heart CT scans, then adjusted it on a smaller set of more difficult scans, helping it learn more accurately. Their results show improvements over other methods, especially in precisely outlining the artery's edges. This work helps in more accurate heart protection during cancer treatment planning.
Left Anterior Descending artery3D segmentationtransformer modelNeighborhood AttentionDilated Neighborhood AttentionDice scoreHausdorff distanceCT imagingradiotherapy planninguncertainty-guided optimization
Authors
Rafi Ibn Sultan, Chengyin Li, Yiannos Demetriou, Ahmed I. Ghanem, Joshua P. Kim, Justine Cunningham, Hassan Bagher-Ebadian, Dongxiao Zhu, Kundan S. Thind
Abstract
Background: Accurate segmentation of the Left Anterior Descending (LAD) artery in 3D free-breathing, non-contrast CT is critical for cardiac dose sparing in thoracic radiotherapy. The LAD is extremely small, has poor soft-tissue contrast, and varies substantially across patients; even manual contours show limited inter-observer agreement, underscoring the ambiguity of the vessel boundaries. Purpose: To develop a transformer-based framework that improves LAD delineation in low-contrast, imbalanced CT through local-global context modeling and uncertainty-guided optimization. Methods: We propose NA-UNETR, a 3D transformer-based segmentation model whose Neighborhood Attention (NA) and Dilated NA (DiNA) blocks jointly capture fine structural detail and long-range context. Given the scarcity of annotated LAD data, the model is pretrained on 1,000 CTA volumes of general coronary anatomy and fine-tuned with LoRA-based parameter-efficient adaptation on 20 free-breathing institutional CT scans. A composite Dice-Focal and Hausdorff loss, dynamically balanced via homoscedastic uncertainty, improves overlap and boundary accuracy. Results: NA-UNETR reached 45.64% Dice, 38.16 mm HD95, and 10.01 mm ASD, improving Dice by 3.10 percentage points over nnU-Net and reducing HD95 by 2.96 mm relative to Swin UNETR, with the strongest boundary accuracy among all models and improved centerline stability. On ImageCAS it achieved 79.49% Dice, 8.89 mm HD95, and 1.02 mm ASD. Ablations confirmed that residual blocks, variable kernels, and uncertainty-weighted loss each contributed. Conclusions: NA-UNETR balances local precision and global context for thin, low-contrast LAD structures, offering a computationally efficient framework for substructure-level cardiac segmentation in radiotherapy planning.