{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,19]],"date-time":"2026-04-19T19:41:52Z","timestamp":1776627712789,"version":"3.51.2"},"reference-count":20,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T00:00:00Z","timestamp":1774310400000},"content-version":"vor","delay-in-days":23,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/100020595","name":"National Science and Technology Council","doi-asserted-by":"publisher","award":["NSTC-113-2123-M-002-017-"],"award-info":[{"award-number":["NSTC-113-2123-M-002-017-"]}],"id":[{"id":"10.13039\/100020595","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100020595","name":"National Science and Technology Council","doi-asserted-by":"publisher","award":["NSTC-114-2123-M-002-010-"],"award-info":[{"award-number":["NSTC-114-2123-M-002-010-"]}],"id":[{"id":"10.13039\/100020595","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,3,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Pancreatic cancer (PC) is frequently missed on unenhanced CT examinations performed for unrelated clinical indications, where the pancreas is included incidentally and clinical suspicion is low.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To develop and validate a deep learning-based tool for PC diagnosis and risk stratification on unenhanced CT.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and Methods<\/jats:title>\n                    <jats:p>This retrospective study included 3080 unenhanced CT studies of Taiwanese patients with PC, other pancreatic diseases and normal pancreas between 2004 and 2019 from a tertiary hospital, randomly divided into training, validation, and internal test sets. Unenhanced CT studies from United States institutions were used for external testing. A hybrid convolutional neural network\u2013transformer model was trained for PC diagnosis and risk stratification. Performance was evaluated using sensitivity, specificity, and area under the curve (AUC), with comparisons to 2 radiologists by McNemar\u2019s test and exploratory decision curve analysis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The internal dataset included 713 PCs (mean age, 64.6\u2009\u00b1\u200912.0\u2009years; 384 men), 1661 normal pancreas and 706 other pancreatic diseases. In an exploratory comparison restricted to unenhanced CT (29 PCs, 31 controls), the sensitivity of computer-aided diagnosis (CAD) tool (89.7%, 72.6-97.8) seemed comparable with that of 1 radiologist (86.2%, 68.3-96.1) and higher than another (41.4%, 23.5-61.1); but wide confidence intervals and inter-radiologist variability warrant cautious interpretation. In the internal test set (142 PCs, 474 controls), sensitivity was 90.8% (84.9-95.0) and specificity 93.0% (90.4-95.2) (AUC: 0.98), with sensitivity comparable to radiologist reports based on enhanced and unenhanced CT (95.4%, 90.2-98.3; P = .21). In the external set (42 PCs, 22 controls), sensitivity was 76.2% (60.5-87.9) and specificity 86.4% (65.1-97.1) (AUC: 0.89). The tool stratified cases into 7 risk levels with likelihood ratios ranging from &amp;lt;0.01 to 173.46. Exploratory decision curve analysis suggested potential net benefit across threshold probabilities.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This tool may assist in the opportunistic detection and risk stratification of PC on unenhanced CT.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/radadv\/umag017","type":"journal-article","created":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T12:46:23Z","timestamp":1774010783000},"source":"Crossref","is-referenced-by-count":0,"title":["Pancreatic cancer diagnosis on unenhanced CT with deep learning for opportunistic diagnosis"],"prefix":"10.1093","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8675-5863","authenticated-orcid":false,"given":"Po-Ting","family":"Chen","sequence":"first","affiliation":[{"name":"Department of Medical Imaging, National Taiwan University Hospital, National Taiwan University College of Medicine , Taipei, 100233,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5313-0453","authenticated-orcid":false,"given":"Dawei","family":"Chang","sequence":"additional","affiliation":[{"name":"Data Science Degree Program, National Taiwan University and Academia Sinica , Taipei, 106319,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-2741-0222","authenticated-orcid":false,"given":"Yenjia","family":"Chen","sequence":"additional","affiliation":[{"name":"Data Science Degree Program, National Taiwan University and Academia Sinica , Taipei, 106319,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3856-048X","authenticated-orcid":false,"given":"Pochuan","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Information Engineering, National Taiwan University , Taipei, 106319,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andre Yanchen","family":"Yeh","sequence":"additional","affiliation":[{"name":"Department of Medicine, National Taiwan University College of Medicine , Taipei, 100233,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4100-8909","authenticated-orcid":false,"given":"Kao-Lang","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Medical Imaging, National Taiwan University Hospital, National Taiwan University College of Medicine , Taipei, 100233,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ming-Shiang","family":"Wu","sequence":"additional","affiliation":[{"name":"Division of Gastroenterology and Hepatology, Department of Internal Medicine, National Taiwan University Hospital, National Taiwan University College of Medicine , Taipei, 100233,","place":["Taiwan"]},{"name":"Internal Medicine, College of Medicine, National Taiwan University , Taipei, 100233,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5362-6953","authenticated-orcid":false,"given":"Wei-Chih","family":"Liao","sequence":"additional","affiliation":[{"name":"Division of Gastroenterology and Hepatology, Department of Internal Medicine, National Taiwan University Hospital, National Taiwan University College of Medicine , Taipei, 100233,","place":["Taiwan"]},{"name":"Internal Medicine, College of Medicine, National Taiwan University , Taipei, 100233,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6154-7750","authenticated-orcid":false,"given":"Weichung","family":"Wang","sequence":"additional","affiliation":[{"name":"Institute of Applied Mathematical Sciences, National Taiwan University , Taipei, 106319,","place":["Taiwan"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2026,3,24]]},"reference":[{"issue":"1","key":"2026041915070051400_umag017-B1","first-page":"17","article-title":"Cancer statistics, 2023","volume":"73","author":"Siegel","year":"2023","journal-title":"CA Cancer J Clin"},{"issue":"4","key":"2026041915070051400_umag017-B2","doi-asserted-by":"crossref","first-page":"2422","DOI":"10.1007\/s00330-020-07307-5","article-title":"Factors associated with missed and misinterpreted cases of pancreatic ductal adenocarcinoma","volume":"31","author":"Kang","year":"2020","journal-title":"Eur Radiol"},{"issue":"10","key":"2026041915070051400_umag017-B3","first-page":"634","article-title":"Rationale use of unenhanced multi-detector CT (CT KUB) in evaluation of suspected renal colic","volume":"10","author":"Nadeem","year":"2012","journal-title":"Int J Surg Ovid Technol (Wolters Kluwer Health)"},{"issue":"7","key":"2026041915070051400_umag017-B4","first-page":"e231112","article-title":"Diagnostic accuracy of unenhanced computed tomography for evaluation of acute abdominal pain in the emergency department","volume":"158","author":"Shaish","year":"2023","journal-title":"JAMA Surg Am Med Assoc"},{"issue":"3","key":"2026041915070051400_umag017-B5","first-page":"168","article-title":"Imaging of pancreatic cancer: An overview","volume":"2","author":"Tummala","year":"2011","journal-title":"J Gastrointest Oncol"},{"issue":"12","key":"2026041915070051400_umag017-B6","doi-asserted-by":"crossref","first-page":"3033","DOI":"10.1038\/s41591-023-02640-w","article-title":"Large-scale pancreatic cancer detection via non-contrast CT and deep learning","volume":"29","author":"Cao","year":"2023","journal-title":"Nat Med"},{"issue":"11","key":"2026041915070051400_umag017-B7","doi-asserted-by":"crossref","first-page":"e230217","DOI":"10.1148\/rg.230217","article-title":"Invited commentary: Failures in imaging diagnosis of pancreatic cancer: can we avoid them and how?","volume":"43","author":"Zins","year":"2023","journal-title":"Radiographics"},{"issue":"5","key":"2026041915070051400_umag017-B8","doi-asserted-by":"crossref","first-page":"umaf028","DOI":"10.1093\/radadv\/umaf028","article-title":"Early detection of pancreatic cancer on computed tomography: advancements with deep learning","volume":"2","author":"Lopez-Ramirez","year":"2025","journal-title":"Radiol Adv"},{"issue":"6","key":"2026041915070051400_umag017-B9","doi-asserted-by":"crossref","first-page":"e303","DOI":"10.1016\/S2589-7500(20)30078-9","article-title":"Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation","volume":"2","author":"Liu","year":"2020","journal-title":"Lancet Digit. Health"},{"issue":"4","key":"2026041915070051400_umag017-B10","doi-asserted-by":"crossref","first-page":"e210010","DOI":"10.1148\/rycan.2021210010","article-title":"Radiomic features at CT can distinguish pancreatic cancer from noncancerous pancreas","volume":"3","author":"Chen","year":"2021","journal-title":"Radiol Imaging Cancer"},{"issue":"5","key":"2026041915070051400_umag017-B11","doi-asserted-by":"crossref","first-page":"1435","DOI":"10.1053\/j.gastro.2022.06.066","article-title":"Radiomics-based machine-learning models can detect pancreatic cancer on prediagnostic computed tomography scans at a substantial lead time before clinical diagnosis","volume":"163","author":"Mukherjee","year":"2022","journal-title":"Gastroenterology"},{"issue":"1","key":"2026041915070051400_umag017-B12","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1148\/radiol.220152","article-title":"Pancreatic cancer detection on CT scans with deep learning: a nationwide population-based study","volume":"306","author":"Chen","year":"2023","journal-title":"Radiology"},{"issue":"1","key":"2026041915070051400_umag017-B13","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1186\/s12885-023-10536-8","article-title":"Detection of pancreatic cancer with two- and three-dimensional radiomic analysis in a nationwide population-based real-world dataset","volume":"23","author":"Chang","year":"2023","journal-title":"BMC Cancer"},{"key":"2026041915070051400_umag017-B14","doi-asserted-by":"publisher","author":"The Clinical Proteomic Tumor Analysis Consortium Pancreatic Ductal Adenocarcinoma Collection (CPTAC-PDA)","DOI":"10.7937\/k9\/tcia.2018.sc20fo18"},{"issue":"6","key":"2026041915070051400_umag017-B15","doi-asserted-by":"crossref","first-page":"333","DOI":"10.1097\/PPO.0000000000000290","article-title":"Diagnosis and detection of pancreatic cancer","volume":"23","author":"Chu","year":"2017","journal-title":"Cancer J"},{"issue":"7458","key":"2026041915070051400_umag017-B16","doi-asserted-by":"crossref","first-page":"168","DOI":"10.1136\/bmj.329.7458.168","article-title":"Diagnostic tests 4: likelihood ratios","volume":"329","author":"Deeks","year":"2004","journal-title":"BMJ"},{"key":"2026041915070051400_umag017-B17","volume-title":"Statistical Methods in Diagnostic Medicine","author":"Zhou","year":"2009"},{"issue":"5","key":"2026041915070051400_umag017-B18","doi-asserted-by":"crossref","first-page":"1524","DOI":"10.1007\/s00261-020-02506-6","article-title":"Diagnostic accuracy of unenhanced CT texture analysis to differentiate mass-forming pancreatitis from pancreatic ductal adenocarcinoma","volume":"45","author":"Ren","year":"2020","journal-title":"Abdom Radiol (NY)"},{"issue":"1","key":"2026041915070051400_umag017-B19","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1007\/s11596-022-2535-2","article-title":"Application of unenhanced computed tomography texture analysis to differentiate pancreatic adenosquamous carcinoma from pancreatic ductal adenocarcinoma","volume":"42","author":"Ren","year":"2022","journal-title":"Curr Med Sci"},{"issue":"11","key":"2026041915070051400_umag017-B20","doi-asserted-by":"crossref","first-page":"e230054","DOI":"10.1148\/rg.230054","article-title":"Pancreatic cancer and its mimics","volume":"43","author":"Miller","year":"2023","journal-title":"Radiographics"}],"container-title":["Radiology Advances"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/radadv\/advance-article-pdf\/doi\/10.1093\/radadv\/umag017\/67502276\/umag017.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/radadv\/article-pdf\/3\/2\/umag017\/67502276\/umag017.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/radadv\/article-pdf\/3\/2\/umag017\/67502276\/umag017.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,19]],"date-time":"2026-04-19T19:07:08Z","timestamp":1776625628000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/radadv\/article\/doi\/10.1093\/radadv\/umag017\/8539946"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3]]},"references-count":20,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,3,3]]}},"URL":"https:\/\/doi.org\/10.1093\/radadv\/umag017","relation":{},"ISSN":["2976-9337"],"issn-type":[{"value":"2976-9337","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,3]]},"published":{"date-parts":[[2026,3]]},"article-number":"umag017"}}