{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T17:41:24Z","timestamp":1790185284531,"version":"4.1.0"},"reference-count":45,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T00:00:00Z","timestamp":1778803200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0"}],"funder":[{"name":"Beijing Hospitals Authority\u2019s Ascent Plan","award":["DFL2022060"],"award-info":[{"award-number":["DFL2022060"]}]},{"name":"Beijing Anzhen Hospital High Level Research Funding","award":["2024AZB2006"],"award-info":[{"award-number":["2024AZB2006"]}]},{"DOI":"10.13039\/501100005090","name":"Beijing Nova Program","doi-asserted-by":"crossref","award":["20250484950"],"award-info":[{"award-number":["20250484950"]}],"id":[{"id":"10.13039\/501100005090","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Tongzhou District Science & Technology Commission","award":["ws-2024-112"],"award-info":[{"award-number":["ws-2024-112"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Coronary artery stenosis is a major pathological basis of coronary artery disease (CAD), and clinical evaluation relies on coronary angiography (CAG) as the gold standard. However, manual assessment suffers from high subjectivity, low efficiency, and poor inter-observer consistency, highlighting the necessity of automated detection methods. Existing studies are mostly \u201csingle-task\u201d approaches (segmentation-only or detection-only), which fail to explicitly exploit the complementary information between the two tasks. Moreover, the scarcity of annotated samples limits generalization in complex vascular structures. To address these challenges, we propose a stenosis detection model that integrates a Multi-level Perception Fusion (MPF) framework with Pseudo-Stenosis Pretraining. The MPF framework places a vessel segmentation module before the detection module, providing vascular spatial constraints and structural cues of stenosis. By propagating detection loss to the vessel segmentation module, this module is also enhanced to highlighting the difference of stenotic regions. Meanwhile, Pseudo-Stenosis Pretraining generates morphologically realistic pseudo stenosis samples and adopts a pretraining\u2013fine-tuning scheme to mitigate overfitting in limited data scenarios. Experimental results demonstrate that the proposed model achieves superior performance compared with state-of-the-art methods on both a private dataset and the public ARCADE dataset. On the private dataset, the model achieves F1, recall, and mean Average Precision (mAP) of 68.1, 73.7, and 72.3 in the detection task, outperforming LSKNet, Oriented-RCNN, ReDet, S2ANet, and RetinaNet. On the ARCADE public dataset, the model achieves F1, sensitivity, and area under the curve (AUC) of 60.52, 62.24, and 99.17, surpassing U-Net, PoolNet, and YOLOv8m.<\/jats:p>","DOI":"10.7717\/peerj-cs.3786","type":"journal-article","created":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T08:24:24Z","timestamp":1778833464000},"page":"e3786","source":"Crossref","is-referenced-by-count":1,"title":["Multi-level perception fusion with pseudo-stenosis pretraining for robust coronary artery stenosis detection"],"prefix":"10.7717","volume":"12","author":[{"given":"Dingce","family":"Sun","sequence":"first","affiliation":[{"name":"Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiahe","family":"Zhu","sequence":"additional","affiliation":[{"name":"School of Intelligence Science and Technology, Nanjing University, Suzhou, China"},{"name":"HuaHui Jian AI Tech Co., Ltd., Tianjin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenyuan","family":"Yu","sequence":"additional","affiliation":[{"name":"Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongli","family":"Liu","sequence":"additional","affiliation":[{"name":"Center of Applied Laser, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bokang","family":"Zhai","sequence":"additional","affiliation":[{"name":"Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lekang","family":"Chen","sequence":"additional","affiliation":[{"name":"HuaHui Jian AI Tech Co., Ltd., Tianjin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"He","sequence":"additional","affiliation":[{"name":"HuaHui Jian AI Tech Co., Ltd., Tianjin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Yu","sequence":"additional","affiliation":[{"name":"Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,5,15]]},"reference":[{"issue":"13","key":"10.7717\/peerj-cs.3786\/ref-1","doi-asserted-by":"publisher","first-page":"e993","DOI":"10.7717\/peerj-cs.993","article-title":"Multi-constraints based deep learning model for automated segmentation and diagnosis of coronary artery disease in X-ray angiographic images","volume":"8","author":"Algarni","year":"2022","journal-title":"PeerJ Computer Science"},{"issue":"1","key":"10.7717\/peerj-cs.3786\/ref-2","doi-asserted-by":"publisher","first-page":"142","DOI":"10.1038\/s41746-023-00880-1","article-title":"CathAI: fully automated coronary angiography interpretation and stenosis estimation","volume":"6","author":"Avram","year":"2023","journal-title":"NPJ Digital Medicine"},{"key":"10.7717\/peerj-cs.3786\/ref-3","doi-asserted-by":"publisher","first-page":"205","DOI":"10.1007\/978-3-031-25066-8_9","article-title":"Swin-Unet: Unet-like pure transformer for medical image segmentation","volume-title":"Lecture Notes in Computer Science","author":"Cao","year":"2023"},{"key":"10.7717\/peerj-cs.3786\/ref-4","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2211.09788","article-title":"DiffusionDet: diffusion model for object detection","author":"Chen","year":"2022"},{"issue":"12","key":"10.7717\/peerj-cs.3786\/ref-5","doi-asserted-by":"publisher","first-page":"102513","DOI":"10.1016\/j.compmedimag.2025.102513","article-title":"Intraoperative stenosis detection in X-ray coronary angiography via temporal fusion and attention-based CNN","volume":"122","author":"Chen","year":"2025","journal-title":"Computerized Medical Imaging and Graphics: the Official Journal of the Computerized Medical Imaging Society"},{"key":"10.7717\/peerj-cs.3786\/ref-6","doi-asserted-by":"publisher","first-page":"1299","DOI":"10.1109\/isbi.2014.6868115","article-title":"Automatic detection of coronary stenosis in X-ray angiography through spatio-temporal tracking","author":"Compas","year":"2014"},{"issue":"1","key":"10.7717\/peerj-cs.3786\/ref-7","doi-asserted-by":"publisher","first-page":"7582","DOI":"10.1038\/s41598-021-87174-2","article-title":"Real-time coronary artery stenosis detection based on modern neural networks","volume":"11","author":"Danilov","year":"2021","journal-title":"Scientific Reports"},{"issue":"Pt A","key":"10.7717\/peerj-cs.3786\/ref-8","doi-asserted-by":"publisher","first-page":"111172","DOI":"10.1016\/j.compbiomed.2025.111172","article-title":"FedStenoNet: tackling domain shift in X-ray coronary angiography through a personalized federated detection framework","volume":"198","author":"Di Cosmo","year":"2025","journal-title":"Computers in Biology and Medicine"},{"issue":"24","key":"10.7717\/peerj-cs.3786\/ref-9","doi-asserted-by":"publisher","first-page":"8134","DOI":"10.3390\/s24248134","article-title":"DCA-YOLOv8: a novel framework combined with AICI loss function for coronary artery stenosis detection","volume":"24","author":"Duan","year":"2024","journal-title":"Sensors"},{"issue":"2","key":"10.7717\/peerj-cs.3786\/ref-10","doi-asserted-by":"publisher","first-page":"442","DOI":"10.1109\/tcsvt.2019.2892986","article-title":"Topology optimization using multiple-possibility fusion for vasculature extraction","volume":"30","author":"Fang","year":"2020","journal-title":"IEEE Transactions on Circuits and Systems for Video Technology"},{"key":"10.7717\/peerj-cs.3786\/ref-11","doi-asserted-by":"publisher","first-page":"106546","DOI":"10.1016\/j.compbiomed.2023.106546","article-title":"Coronary artery stenosis detection via proposal-shifted spatial-temporal transformer in X-ray angiography","volume":"153","author":"Han","year":"2023","journal-title":"Computers in Biology and Medicine"},{"key":"10.7717\/peerj-cs.3786\/ref-12","doi-asserted-by":"publisher","first-page":"106787","DOI":"10.1016\/j.cmpb.2022.106787","article-title":"Recursive centerline- and direction-aware joint learning network with ensemble strategy for vessel segmentation in X-ray angiography images","volume":"220","author":"Han","year":"2022a","journal-title":"Computer Methods and Programs in Biomedicine"},{"key":"10.7717\/peerj-cs.3786\/ref-13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2021.3062048","article-title":"Align deep features for oriented object detection","volume":"60","author":"Han","year":"2022b","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3786\/ref-14","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2103.07733","article-title":"ReDet: a rotation-equivariant detector for aerial object detection (Version 1)","author":"Han","year":"2021"},{"key":"10.7717\/peerj-cs.3786\/ref-15","article-title":"Denoising diffusion probabilistic models","volume-title":"34th Conference on Neural Information Processing Systems (NeurIPS 2020), Vancouver, Canada. Advances in Neural Information Processing Systems 33","author":"Ho","year":"2020"},{"key":"10.7717\/peerj-cs.3786\/ref-16","doi-asserted-by":"publisher","first-page":"1055","DOI":"10.1109\/icassp40776.2020.9053405","article-title":"UNet 3+: a full-scale connected UNet for medical image segmentation","author":"Huang","year":"2020"},{"key":"10.7717\/peerj-cs.3786\/ref-17","doi-asserted-by":"publisher","first-page":"692","DOI":"10.1007\/978-3-031-43898-1_66","article-title":"ACC-UNet: a completely convolutional UNet model for the 2020s","volume-title":"Lecture Notes in Computer Science","author":"Ibtehaz","year":"2023"},{"issue":"3","key":"10.7717\/peerj-cs.3786\/ref-18","doi-asserted-by":"publisher","first-page":"259","DOI":"10.1007\/s10554-009-9526-2","article-title":"New approaches for the assessment of vessel sizes in quantitative (cardio-)vascular X-ray analysis","volume":"26","author":"Janssen","year":"2010","journal-title":"The International Journal of Cardiovascular Imaging"},{"key":"10.7717\/peerj-cs.3786\/ref-19","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2410.17725","article-title":"YOLOv11: an overview of the key architectural enhancements (Version 1)","author":"Khanam","year":"2024"},{"issue":"1","key":"10.7717\/peerj-cs.3786\/ref-20","doi-asserted-by":"publisher","first-page":"138","DOI":"10.1038\/s41746-024-01134-4","article-title":"Evaluation of stenoses using AI video models applied to coronary angiography","volume":"7","author":"Labrecque Langlais","year":"2024","journal-title":"NPJ Digital Medicine"},{"issue":"2","key":"10.7717\/peerj-cs.3786\/ref-21","doi-asserted-by":"publisher","first-page":"108617","DOI":"10.1016\/j.bspc.2025.108617","article-title":"Stenosis detection of coronary arteries in X-ray angiography images using Swin UNETR with self-supervised pre-training approach and PEFT strategy","volume":"112","author":"Lalinia","year":"2026","journal-title":"Biomedical Signal Processing and Control"},{"key":"10.7717\/peerj-cs.3786\/ref-22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2311.10281","article-title":"SSASS: semi-supervised approach for stenosis segmentation (Version 1)","author":"Lee","year":"2023"},{"issue":"3","key":"10.7717\/peerj-cs.3786\/ref-23","doi-asserted-by":"publisher","first-page":"1410","DOI":"10.1007\/s11263-024-02247-9","article-title":"LSKNet: a foundation lightweight backbone for remote sensing","volume":"133","author":"Li","year":"2025","journal-title":"International Journal of Computer Vision"},{"key":"10.7717\/peerj-cs.3786\/ref-24","doi-asserted-by":"publisher","first-page":"2999","DOI":"10.1109\/iccv.2017.324","article-title":"Focal loss for dense object detection","author":"Lin","year":"2017"},{"key":"10.7717\/peerj-cs.3786\/ref-25","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2310.14961","article-title":"StenUNet: automatic stenosis detection from X-ray coronary angiography (Version 1)","author":"Lin","year":"2023"},{"key":"10.7717\/peerj-cs.3786\/ref-26","doi-asserted-by":"publisher","first-page":"3912","DOI":"10.1109\/cvpr.2019.00404","article-title":"A simple pooling-based design for real-time salient object detection","author":"Liu","year":"2019"},{"issue":"9","key":"10.7717\/peerj-cs.3786\/ref-27","doi-asserted-by":"publisher","first-page":"4623","DOI":"10.1109\/JBHI.2022.3188710","article-title":"Full-resolution network and dual-threshold iteration for retinal vessel and coronary angiograph segmentation","volume":"26","author":"Liu","year":"2022","journal-title":"IEEE Journal of Biomedical and Health Informatics"},{"key":"10.7717\/peerj-cs.3786\/ref-28","doi-asserted-by":"publisher","first-page":"7516","DOI":"10.1109\/iccv48922.2021.00744","article-title":"Self-supervised vessel segmentation via adversarial learning","author":"Ma","year":"2021"},{"issue":"2","key":"10.7717\/peerj-cs.3786\/ref-29","doi-asserted-by":"publisher","first-page":"108023","DOI":"10.1016\/j.bspc.2025.108023","article-title":"Enhanced coronary artery segmentation and stenosis detection: leveraging novel deep learning techniques","volume":"109","author":"Morshedul Abedin","year":"2025","journal-title":"Biomedical Signal Processing and Control"},{"issue":"1","key":"10.7717\/peerj-cs.3786\/ref-30","doi-asserted-by":"publisher","first-page":"116112","DOI":"10.1016\/j.eswa.2021.116112","article-title":"Hybrid classical-quantum Convolutional Neural Network for stenosis detection in X-ray coronary angiography","volume":"189","author":"Ovalle-Magallanes","year":"2022","journal-title":"Expert Systems with Applications"},{"key":"10.7717\/peerj-cs.3786\/ref-31","doi-asserted-by":"publisher","first-page":"101900","DOI":"10.1016\/j.compmedimag.2021.101900","article-title":"Stenosis-DetNet: sequence consistency-based stenosis detection for X-ray coronary angiography","volume":"89","author":"Pang","year":"2021","journal-title":"Computerized Medical Imaging and Graphics: the Official Journal of the Computerized Medical Imaging Society"},{"issue":"Pt B","key":"10.7717\/peerj-cs.3786\/ref-32","doi-asserted-by":"publisher","first-page":"110697","DOI":"10.1016\/j.compbiomed.2025.110697","article-title":"Hyperparameter optimization of YOLO models for invasive coronary angiography lesion detection and assessment","volume":"196","author":"Pascual-Gonz\u00e1lez","year":"2025","journal-title":"Computers in Biology and Medicine"},{"key":"10.7717\/peerj-cs.3786\/ref-33","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2407.19894","article-title":"CardioSyntax: end-to-end SYNTAX score prediction\u2013dataset, benchmark and method","author":"Ponomarchuk","year":"2024"},{"issue":"1","key":"10.7717\/peerj-cs.3786\/ref-34","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1038\/s41597-023-02871-z","article-title":"Dataset for automatic region-based coronary artery disease diagnostics using X-ray angiography images","volume":"11","author":"Popov","year":"2024","journal-title":"Scientific Data"},{"issue":"13","key":"10.7717\/peerj-cs.3786\/ref-35","doi-asserted-by":"publisher","first-page":"e2771","DOI":"10.7717\/peerj-cs.2771","article-title":"Coronary artery disease classification using ConvMixer based classifier from CT angiography images","volume":"11","author":"Rajeev","year":"2025","journal-title":"PeerJ Computer Science"},{"key":"10.7717\/peerj-cs.3786\/ref-36","doi-asserted-by":"publisher","first-page":"234","DOI":"10.1007\/978-3-319-24574-4_28","article-title":"U-Net: convolutional networks for biomedical image segmentation","volume-title":"Medical Image Computing and Computer-Assisted Intervention\u2013MICCAI 2015. MICCAI 2015. Lecture Notes in Computer Science","volume":"9351","author":"Ronneberger","year":"2015"},{"issue":"Suppl 6","key":"10.7717\/peerj-cs.3786\/ref-37","doi-asserted-by":"publisher","first-page":"270","DOI":"10.1186\/s12911-019-0966-x","article-title":"Inter\/intra-frame constrained vascular segmentation in X-ray angiographic image sequence","volume":"19","author":"Song","year":"2019","journal-title":"BMC Medical Informatics and Decision Making"},{"key":"10.7717\/peerj-cs.3786\/ref-38","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2010.02502","article-title":"Denoising diffusion implicit models","author":"Song","year":"2020"},{"key":"10.7717\/peerj-cs.3786\/ref-39","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2502.12524","article-title":"YOLOv12: attention-centric real-time object detectors (Version 1)","author":"Tian","year":"2025"},{"key":"10.7717\/peerj-cs.3786\/ref-40","article-title":"YOLOv8: the next generation of YOLO models","author":"Ultralytics","year":"2023"},{"issue":"10","key":"10.7717\/peerj-cs.3786\/ref-41","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1016\/j.cmpb.2018.10.013","article-title":"Automated identification and grading of coronary artery stenoses with X-ray angiography","volume":"167","author":"Wan","year":"2018","journal-title":"Computer Methods and Programs in Biomedicine"},{"key":"10.7717\/peerj-cs.3786\/ref-42","doi-asserted-by":"publisher","first-page":"103657","DOI":"10.1016\/j.compbiomed.2020.103657","article-title":"Automatic detection of coronary artery stenosis by convolutional neural network with temporal constraint","volume":"118","author":"Wu","year":"2020","journal-title":"Computers in Biology and Medicine"},{"key":"10.7717\/peerj-cs.3786\/ref-43","doi-asserted-by":"publisher","first-page":"3500","DOI":"10.1109\/iccv48922.2021.00350","article-title":"Oriented R-CNN for object detection","author":"Xie","year":"2021"},{"issue":"3","key":"10.7717\/peerj-cs.3786\/ref-44","doi-asserted-by":"publisher","first-page":"864","DOI":"10.1109\/TMI.2022.3219126","article-title":"Progressive perception learning for main coronary segmentation in X-ray angiography","volume":"42","author":"Zhang","year":"2023","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"10.7717\/peerj-cs.3786\/ref-45","doi-asserted-by":"publisher","first-page":"2242","DOI":"10.1109\/iccv.2017.244","article-title":"Unpaired image-to-image translation using cycle-consistent adversarial networks","author":"Zhu","year":"2017"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3786.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3786.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3786.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3786.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T08:24:29Z","timestamp":1778833469000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3786"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,15]]},"references-count":45,"alternative-id":["10.7717\/peerj-cs.3786"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3786","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,15]]},"article-number":"e3786"}}