{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T19:07:28Z","timestamp":1771960048042,"version":"3.50.1"},"reference-count":29,"publisher":"Public Library of Science (PLoS)","issue":"2","license":[{"start":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T00:00:00Z","timestamp":1771891200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Rapid and accurate detection of coronary plaques on CCTA is critical for timely CAD diagnosis but is limited by reader workload and interobserver variability. Our objective was to evaluate the effectiveness of machine learning (ML) based on automated segmentation of coronary plaques in coronary computed tomography angiography (CCTA). We retrospectively analyzed CCTA scans from 1,642 patients (4,711 training vessels, 1,112 test vessels, plus 1,613 negative vessels) using an nnU-Net 3D full resolution architecture. Hyperparameters (batch size, learning-rate scheduler, epochs) were optimized on a 10% subset of the training dataset, and the final model was trained with 5-fold cross-validation on positive cases. Test performance was assessed at the plaque, vessel, and exam levels. Of 2,090 ground-truth plaques, the model achieved 1,772 TP for 84.8% sensitivity (95% CI 83.2\u201386.3%), 382 FP for 82.3% precision (95% CI 80.6\u201383.8), and a median Dice score of 0.88 (95% CI 0.87\u20130.89). At the vessel level, sensitivity was 94.7% (95% CI 93.2\u201395.9%) and specificity was 84.9% (95% CI 83.0\u201386.5%). At the examination level, sensitivity was 97.4% (95% CI 94.5\u201398.8%) and specificity was 79.7% (95% CI 76.5\u201382.6%). Detection of small, non-calcified plaques remained challenging, and false positives were often due to motion or step artifacts. Our model showed promising results in detecting coronary plaques in CCTA examinations, particularly for medium to large plaques and high negative predictive value for ruling out significant CAD, offering potential to streamline radiology workflows. Future work will focus on small-plaque detection, artifact robustness, and multi-center validation to enhance clinical utility.<\/jats:p>","DOI":"10.1371\/journal.pone.0343887","type":"journal-article","created":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T18:26:12Z","timestamp":1771957572000},"page":"e0343887","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Automatic segmentation of coronary plaques in coronary CT angiography using neural networks"],"prefix":"10.1371","volume":"21","author":[{"given":"Mahdi","family":"Moosavi","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Keno","family":"Bressem","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rafael","family":"Adolf","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anastasiya","family":"Valentik","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Albrecht","family":"Will","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Eva","family":"Hendrich","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6267-1692","authenticated-orcid":true,"given":"Martin","family":"Hadamitzky","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2026,2,24]]},"reference":[{"issue":"9967","key":"pone.0343887.ref001","doi-asserted-by":"crossref","first-page":"540","DOI":"10.1016\/S0140-6736(14)60569-9","article-title":"Causes of international increases in older age life expectancy","volume":"385","author":"CD Mathers","year":"2015","journal-title":"Lancet"},{"issue":"10159","key":"pone.0343887.ref002","doi-asserted-by":"crossref","first-page":"1736","DOI":"10.1016\/S0140-6736(18)32203-7","article-title":"Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980-2017: a systematic analysis for the Global Burden of Disease Study 2017","volume":"392","author":"GBD 2017 Causes of Death Collaborators","year":"2018","journal-title":"Lancet"},{"issue":"10","key":"pone.0343887.ref003","doi-asserted-by":"crossref","first-page":"7217","DOI":"10.1007\/s00330-022-08801-8","article-title":"Deep learning-based atherosclerotic coronary plaque segmentation on coronary CT angiography","volume":"32","author":"N J\u00e1vorszky","year":"2022","journal-title":"Eur Radiol"},{"issue":"3","key":"pone.0343887.ref004","doi-asserted-by":"crossref","first-page":"407","DOI":"10.1093\/eurheartj\/ehz425","article-title":"2019 ESC Guidelines for the diagnosis and management of chronic coronary syndromes","volume":"41","author":"J Knuuti","year":"2020","journal-title":"Eur Heart J"},{"issue":"5","key":"pone.0343887.ref005","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1007\/s12410-017-9412-6","article-title":"The Updated NICE Guidelines: Cardiac CT as the First-Line Test for Coronary Artery Disease","volume":"10","author":"AJ Moss","year":"2017","journal-title":"Curr Cardiovasc Imaging Rep"},{"issue":"5","key":"pone.0343887.ref006","doi-asserted-by":"crossref","first-page":"342","DOI":"10.1016\/j.jcct.2014.07.003","article-title":"SCCT guidelines for the interpretation and reporting of coronary CT angiography: a report of the Society of Cardiovascular Computed Tomography Guidelines Committee","volume":"8","author":"J Leipsic","year":"2014","journal-title":"J Cardiovasc Comput Tomogr"},{"issue":"5","key":"pone.0343887.ref007","doi-asserted-by":"crossref","first-page":"489","DOI":"10.21037\/cdt.2016.11.06","article-title":"Plaque imaging with CT-a comprehensive review on coronary CT angiography based risk assessment","volume":"7","author":"M Kolossv\u00e1ry","year":"2017","journal-title":"Cardiovasc Diagn Ther"},{"key":"pone.0343887.ref008","article-title":"Diagnosis of obstructive coronary artery disease using computed tomography angiography in patients with stable chest pain depending on clinical probability and in clinically important subgroups: meta-analysis of individual patient data","volume":"365","author":"R Haase","year":"2019","journal-title":"BMJ"},{"issue":"2","key":"pone.0343887.ref009","doi-asserted-by":"crossref","first-page":"124","DOI":"10.1016\/j.jcct.2021.07.001","article-title":"Quantitative assessment of atherosclerotic plaque, recent progress and current limitations","volume":"16","author":"MC Williams","year":"2022","journal-title":"J Cardiovasc Comput Tomogr"},{"issue":"6","key":"pone.0343887.ref010","doi-asserted-by":"crossref","first-page":"1209","DOI":"10.1007\/s10554-023-02824-y","article-title":"Convolutional neural networks on risk stratification of patients with suspected coronary artery disease undergoing coronary computed tomography angiography","volume":"39","author":"R Adolf","year":"2023","journal-title":"Int J Cardiovasc Imaging"},{"key":"pone.0343887.ref011","doi-asserted-by":"crossref","DOI":"10.1093\/eurheartjsupp\/suad038","article-title":"Artificial intelligence for advanced analysis of coronary plaque","volume":"25","author":"M van Assen","year":"2023","journal-title":"Eur Heart J Suppl"},{"issue":"24","key":"pone.0343887.ref012","doi-asserted-by":"crossref","first-page":"1975","DOI":"10.1093\/eurheartj\/ehy404","article-title":"Clinical applications of machine learning in cardiovascular disease and its relevance to cardiac imaging","volume":"40","author":"SJ Al\u2019Aref","year":"2019","journal-title":"Eur Heart J"},{"issue":"12","key":"pone.0343887.ref013","doi-asserted-by":"crossref","DOI":"10.1161\/CIRCIMAGING.117.006843","article-title":"Radiomic features are superior to conventional quantitative computed tomographic metrics to identify coronary plaques with napkin\u2011ring sign","volume":"10","author":"M Kolossv\u00e1ry","year":"2017","journal-title":"Circ Cardiovasc Imaging"},{"issue":"6","key":"pone.0343887.ref014","doi-asserted-by":"crossref","first-page":"2655","DOI":"10.1007\/s00330-017-5223-z","article-title":"Integrated prediction of lesion-specific ischaemia from quantitative coronary CT angiography using machine learning: a multicentre study","volume":"28","author":"D Dey","year":"2018","journal-title":"Eur Radiol"},{"issue":"1","key":"pone.0343887.ref015","doi-asserted-by":"crossref","first-page":"4938","DOI":"10.1038\/s41467-023-40564-8","article-title":"Deep-learning based detection of vessel occlusions on CT-angiography in patients with suspected acute ischemic stroke","volume":"14","author":"G Brugnara","year":"2023","journal-title":"Nat Commun"},{"key":"pone.0343887.ref016","author":"S Zhang","year":"2022"},{"key":"pone.0343887.ref017","doi-asserted-by":"crossref","first-page":"1057800","DOI":"10.3389\/fphys.2022.1057800","article-title":"The application of the nnU-Net-based automatic segmentation model in assisting carotid artery stenosis and carotid atherosclerotic plaque evaluation","volume":"13","author":"Y Zhu","year":"2022","journal-title":"Front Physiol"},{"issue":"1","key":"pone.0343887.ref018","doi-asserted-by":"crossref","first-page":"208","DOI":"10.1186\/s13244-023-01561-z","article-title":"Pricing and cost-saving potential for deep-learning computer-aided lung nodule detection software in CT lung cancer screening","volume":"14","author":"Y Du","year":"2023","journal-title":"Insights Imaging"},{"issue":"2","key":"pone.0343887.ref019","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"F Isensee","year":"2021","journal-title":"Nat Methods"},{"key":"pone.0343887.ref020","doi-asserted-by":"crossref","first-page":"115","DOI":"10.4028\/www.scientific.net\/AMM.611.115","article-title":"Methodology and Application of the Kruskal-Wallis Test","volume":"611","author":"E Ostertagov\u00e1","year":"2014","journal-title":"AMM"},{"key":"pone.0343887.ref021","first-page":"131","article-title":"A Vessel-Focused 3D Convolutional Network for Automatic Segmentation and Classification of Coronary Artery Plaques in Cardiac CTA","volume-title":"Lecture Notes in Computer Science","author":"J Liu","year":"2019"},{"issue":"3","key":"pone.0343887.ref022","doi-asserted-by":"crossref","first-page":"661","DOI":"10.1016\/j.radi.2022.05.002","article-title":"Corrigendum to \u201cDeep learning with convolutional neural network for estimation of the characterisation of coronary plaques: Validation using IB-IVUS\u201d [Radiography 28 (2022) 61-67]","volume":"28","author":"T Masuda","year":"2022","journal-title":"Radiography (Lond)"},{"issue":"7","key":"pone.0343887.ref023","doi-asserted-by":"crossref","first-page":"1588","DOI":"10.1109\/TMI.2018.2883807","article-title":"A Recurrent CNN for Automatic Detection and Classification of Coronary Artery Plaque and Stenosis in Coronary CT Angiography","volume":"38","author":"M Zreik","year":"2019","journal-title":"IEEE Trans Med Imaging"},{"key":"pone.0343887.ref024","article-title":"Deep learning-based stenosis quantification from coronary CT Angiography","author":"Y Hong","year":"2019","journal-title":"Proc SPIE Int Soc Opt Eng"},{"issue":"1","key":"pone.0343887.ref025","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1148\/radiol.2020191621","article-title":"Deep Learning for Automatic Calcium Scoring in CT: Validation Using Multiple Cardiac CT and Chest CT Protocols","volume":"295","author":"SGM van Velzen","year":"2020","journal-title":"Radiology"},{"key":"pone.0343887.ref026","doi-asserted-by":"crossref","unstructured":"Lin T-Y, Goyal P, Girshick R, He K, Dollar P. Focal Loss for Dense Object Detection. In: 2017 IEEE International Conference on Computer Vision (ICCV), 2017. 2999\u20133007. doi: 10.1109\/iccv.2017.324","DOI":"10.1109\/ICCV.2017.324"},{"key":"pone.0343887.ref027","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-01234-2_1","article-title":"CBAM: Convolutional Block Attention Module","volume-title":"Computer Vision \u2013 ECCV 2018. ECCV 2018","author":"S Woo","year":"2018"},{"issue":"1","key":"pone.0343887.ref028","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1038\/s41598-018-37168-4","article-title":"Automatic Coronary Wall and Atherosclerotic Plaque Segmentation from 3D Coronary CT Angiography","volume":"9","author":"AM Ghanem","year":"2019","journal-title":"Sci Rep"},{"issue":"8","key":"pone.0343887.ref029","doi-asserted-by":"crossref","first-page":"1987","DOI":"10.3390\/diagnostics12081987","article-title":"Automated Identification of Coronary Arteries in Assisting Inexperienced Readers: Comparison between Two Commercial Vendors","volume":"12","author":"D De Santis","year":"2022","journal-title":"Diagnostics (Basel)"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0343887","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T18:26:18Z","timestamp":1771957578000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0343887"}},"subtitle":[],"editor":[{"given":"Redoy","family":"Ranjan","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,2,24]]},"references-count":29,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,2,24]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0343887","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,24]]}}}