Fully Automated Segmentation of Chronic Total Occlusion Lesions on Coronary CT Angiography Using Deep Learning
- 16 November 2026
- journal article
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
Abstract
Reliable quantitative assessment for chronic total occlusion (CTO) lesions depends on accurate segmentation of coronary computed tomography angiography (CCTA) images. However, incomplete contrast opacification and complex lesion morphology make precise delineation particularly challenging. We developed an Atrous Attention U-Net (AAU-Net) for CTO segmentation and integrated it into a fully automated workflow comprising heart localization, CTO localization, and lesion segmentation. This retrospective study included 565 CTO lesions from 551 patients who underwent prospectively ECG-gated CCTA at three centers. Standalone segmentation used reference-mask-derived lesion-containing slices and regions of interest, whereas the automated workflow used predicted localization without reference-mask information or manual correction. Under reference-defined localization, AAU-Net achieved a Dice score of 0.8421 and intersection over union of 0.7312 in the internal test set, although performance varied by anatomical location and center. Performance decreased in the fully automated workflow, with Dice scores of 0.6922 in the internal test set, 0.5822 in external test set 1, and 0.6480 in external test set 2, potentially reflecting localization errors and differences between the separately trained segmentation models. In an exploratory locked-model radiomics analysis, features derived from automated segmentations showed numerically lower predictive performance than those derived from manual segmentations. Overall, AAU-Net showed favorable segmentation performance under reference-defined localization, but the fully automated workflow demonstrated lower performance that varied across centers. The radiomics findings demonstrate technical feasibility rather than established clinical utility. Further optimization and larger prospective multicenter validation are required before clinical deployment.Keywords
Funding Information
- National Natural Science Foundation of China (82272068)
- National Natural Science Foundation of China (32571731)
- National Natural Science Foundation of China (82572311)
- National Natural Science Foundation of China (U25A20148)
- Natural Science Foundation of Beijing Municipality (L256013)
- Natural Science Foundation of Beijing Municipality (L244060)
- Fundamental Research Funds for the Central Universities (GW2025-76)
- Key Technologies Research and Development Program (2021YFA1301603)
- Capital Health Research and Development of Special Fund (2022-1-2061)
- Beijing Natural Science Foundation-Xiaomi Innovation Joint Fund (L233009)
- High-level public health technical talent construction project of Beijing Municipal Health Commission (Leading Talent-02-01)
- Beijing Hospitals Authority’s Ascent Plan (DFL20220603)
- Beijing Anzhen Hospital High Level Research Funding (2025AZC2019)
- Beijing Anzhen Hospital High Level Research Funding (2025AZC2018)
This publication has 50 references indexed in Scilit:
- CT Angiography Image Postprocessing and Data Analysis2024
- Machine learning and deep learning approach for medical image analysis: diagnosis to detectionMultimedia Tools and Applications, 2022
- Automated CT pancreas segmentation for acute pancreatitis patients by combining a novel object detection approach and U-NetBiomedical Signal Processing and Control, 2022
- A novel end‐to‐end deep learning solution for coronary artery segmentation from CCTAMedical Physics, 2022
- Automatic coronary artery segmentation and diagnosis of stenosis by deep learning based on computed tomographic coronary angiographyEuropean Radiology, 2022
- nnDetection: A Self-configuring Method for Medical Object Detection2021
- DGFAU-Net: Global feature attention upsampling network for medical image segmentationNeural Computing and Applications, 2021
- UNet++: A Nested U-Net Architecture for Medical Image Segmentation2018
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Prevalence and management of coronary chronic total occlusions in a tertiary veterans affairs hospitalCatheterization and Cardiovascular Interventions, 2013