PaSAL: A Deep Learning Pipeline for Pulmonary Artery-Vein Segmentation and Anatomical Labeling in Thoracic CT

Jasper Eppink, Hoel Kervadec, Julian van Capelleveen, Joost Verhoeff, Suresh Senan, Omar Bohoudi
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:3561-3592, 2026.

Abstract

We present PaSAL, a deep learning pipeline for pulmonary artery-vein segmentation and anatomical labeling in thoracic CT. PaSAL combines an nnU-Net-based binary vessel segmentation model with a graph-based anatomical labeling framework that assigns 19 clinically defined vascular classes. The pipeline integrates vessel enhancement, skeletonization, and topology-aware label propagation to produce anatomically coherent outputs. PaSAL is trained on the HiPaS and PTL public datasets and evaluated on an external set of 63 clinical scans from Amsterdam UMC. On HiPaS, PaSAL achieves Dice scores of $89.5%$ (arteries) and $88.1%$ (veins). On PTL, voxel-level anatomical labeling accuracy reaches $90.1%$ for arteries and $82.7%$ for veins. Expert review confirms high anatomical plausibility and clinical utility, while showing weak correlation between standard quantitative metrics and perceived quality. To our knowledge, PaSAL is the first method to jointly perform artery-vein segmentation and anatomical labeling in CT. The results demonstrate robust performance across diverse anatomical presentations, including pre- and post-radiotherapy scans, and establish PaSAL as a useful baseline tool for vascular analysis in medical imaging.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-eppink26a, title = {PaSAL: A Deep Learning Pipeline for Pulmonary Artery-Vein Segmentation and Anatomical Labeling in Thoracic CT}, author = {Eppink, Jasper and Kervadec, Hoel and van Capelleveen, Julian and Verhoeff, Joost and Senan, Suresh and Bohoudi, Omar}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {3561--3592}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/eppink26a/eppink26a.pdf}, url = {https://proceedings.mlr.press/v315/eppink26a.html}, abstract = {We present PaSAL, a deep learning pipeline for pulmonary artery-vein segmentation and anatomical labeling in thoracic CT. PaSAL combines an nnU-Net-based binary vessel segmentation model with a graph-based anatomical labeling framework that assigns 19 clinically defined vascular classes. The pipeline integrates vessel enhancement, skeletonization, and topology-aware label propagation to produce anatomically coherent outputs. PaSAL is trained on the HiPaS and PTL public datasets and evaluated on an external set of 63 clinical scans from Amsterdam UMC. On HiPaS, PaSAL achieves Dice scores of $89.5%$ (arteries) and $88.1%$ (veins). On PTL, voxel-level anatomical labeling accuracy reaches $90.1%$ for arteries and $82.7%$ for veins. Expert review confirms high anatomical plausibility and clinical utility, while showing weak correlation between standard quantitative metrics and perceived quality. To our knowledge, PaSAL is the first method to jointly perform artery-vein segmentation and anatomical labeling in CT. The results demonstrate robust performance across diverse anatomical presentations, including pre- and post-radiotherapy scans, and establish PaSAL as a useful baseline tool for vascular analysis in medical imaging.} }
Endnote
%0 Conference Paper %T PaSAL: A Deep Learning Pipeline for Pulmonary Artery-Vein Segmentation and Anatomical Labeling in Thoracic CT %A Jasper Eppink %A Hoel Kervadec %A Julian van Capelleveen %A Joost Verhoeff %A Suresh Senan %A Omar Bohoudi %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-eppink26a %I PMLR %P 3561--3592 %U https://proceedings.mlr.press/v315/eppink26a.html %V 315 %X We present PaSAL, a deep learning pipeline for pulmonary artery-vein segmentation and anatomical labeling in thoracic CT. PaSAL combines an nnU-Net-based binary vessel segmentation model with a graph-based anatomical labeling framework that assigns 19 clinically defined vascular classes. The pipeline integrates vessel enhancement, skeletonization, and topology-aware label propagation to produce anatomically coherent outputs. PaSAL is trained on the HiPaS and PTL public datasets and evaluated on an external set of 63 clinical scans from Amsterdam UMC. On HiPaS, PaSAL achieves Dice scores of $89.5%$ (arteries) and $88.1%$ (veins). On PTL, voxel-level anatomical labeling accuracy reaches $90.1%$ for arteries and $82.7%$ for veins. Expert review confirms high anatomical plausibility and clinical utility, while showing weak correlation between standard quantitative metrics and perceived quality. To our knowledge, PaSAL is the first method to jointly perform artery-vein segmentation and anatomical labeling in CT. The results demonstrate robust performance across diverse anatomical presentations, including pre- and post-radiotherapy scans, and establish PaSAL as a useful baseline tool for vascular analysis in medical imaging.
APA
Eppink, J., Kervadec, H., van Capelleveen, J., Verhoeff, J., Senan, S. & Bohoudi, O.. (2026). PaSAL: A Deep Learning Pipeline for Pulmonary Artery-Vein Segmentation and Anatomical Labeling in Thoracic CT. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:3561-3592 Available from https://proceedings.mlr.press/v315/eppink26a.html.

Related Material