{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T14:19:06Z","timestamp":1787062746294,"version":"build-2736575974"},"reference-count":23,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,6,11]],"date-time":"2025-06-11T00:00:00Z","timestamp":1749600000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,6,11]],"date-time":"2025-06-11T00:00:00Z","timestamp":1749600000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100004721","name":"The University of Tokyo","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100004721","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Abdom Radiol"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>Appropriate categorization based on magnetic resonance imaging (MRI) findings is important for managing intraductal papillary mucinous neoplasms (IPMNs). In this study, a large language model (LLM) that classifies IPMNs based on MRI findings was developed, and its performance was compared with that of less experienced human readers.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The medical image management and processing systems of our hospital were searched to identify MRI reports of branch-duct IPMNs (BD-IPMNs). They were assigned to the training, validation, and testing datasets in chronological order. The model was trained on the training dataset, and the best-performing model on the validation dataset was evaluated on the test dataset. Furthermore, two radiology residents (Readers 1 and 2) and an intern (Reader 3) manually sorted the reports in the test dataset. The accuracy, sensitivity, and time required for categorizing were compared between the model and readers.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The accuracy of the fine-tuned LLM for the test dataset was 0.966, which was comparable to that of Readers 1 and 2 (0.931\u20130.972) and significantly better than that of Reader 3 (0.907). The fine-tuned LLM had an area under the receiver operating characteristic curve of 0.982 for the classification of cyst diameter\u2009\u2265\u200910\u00a0mm, which was significantly superior to that of Reader 3 (0.944). Furthermore, the fine-tuned LLM (25\u00a0s) completed the test dataset faster than the readers (1,887\u20132,646\u00a0s).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The fine-tuned LLM classified BD-IPMNs based on MRI findings with comparable performance to that of radiology residents and significantly reduced the time required.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00261-025-05062-z","type":"journal-article","created":{"date-parts":[[2025,6,10]],"date-time":"2025-06-10T23:04:10Z","timestamp":1749596650000},"page":"417-423","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Efficacy of a large language model in classifying branch-duct intraductal papillary mucinous neoplasms"],"prefix":"10.1007","volume":"51","author":[{"given":"Mai","family":"Sato","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Koichiro","family":"Yasaka","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shimon","family":"Abe","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joji","family":"Kurashima","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yusuke","family":"Asari","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shigeru","family":"Kiryu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Osamu","family":"Abe","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,6,11]]},"reference":[{"issue":"12","key":"5062_CR1","doi-asserted-by":"publisher","first-page":"1880","DOI":"10.4251\/wjgo.v13.i12.1880","volume":"13","author":"B Jablonska","year":"2021","unstructured":"Jablonska B, Szmigiel P, Mrowiec S (2021) Pancreatic intraductal papillary mucinous neoplasms: Current diagnosis and management. World J Gastrointest Oncol 13(12):1880\u20131895. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/35070031.","journal-title":"World J Gastrointest Oncol"},{"key":"5062_CR2","doi-asserted-by":"crossref","unstructured":"Nista EC, Schepis T, Candelli M, Giuli L, Pignataro G, Franceschi F, Gasbarrini A, Ojetti V (2021) Humoral Predictors of Malignancy in IPMN: A Review of the Literature. Int J Mol Sci 22(23) https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/34884643.","DOI":"10.3390\/ijms222312839"},{"issue":"3","key":"5062_CR3","doi-asserted-by":"publisher","first-page":"236","DOI":"10.1016\/j.hpb.2015.10.010","volume":"18","author":"N Rezaee","year":"2016","unstructured":"Rezaee N, Barbon C, Zaki A, He J, Salman B, Hruban RH, Cameron JL, Herman JM, Ahuja N, Lennon AM, Weiss MJ, Wood LD, Wolfgang CL (2016) Intraductal papillary mucinous neoplasm (IPMN) with high-grade dysplasia is a risk factor for the subsequent development of pancreatic ductal adenocarcinoma. HPB (Oxford) 18(3):236\u2013246. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/27017163.","journal-title":"HPB (Oxford)"},{"issue":"9","key":"5062_CR4","doi-asserted-by":"publisher","first-page":"1652","DOI":"10.1136\/gutjnl-2017-315062","volume":"67","author":"M Felsenstein","year":"2018","unstructured":"Felsenstein M, Noe M, Masica DL, Hosoda W, Chianchiano P, Fischer CG, Lionheart G, Brosens LAA, Pea A, Yu J, Gemenetzis G, Groot VP, Makary MA, He J, Weiss MJ, Cameron JL, Wolfgang CL, Hruban RH, Roberts NJ, Karchin R, Goggins MG, Wood LD (2018) IPMNs with co-occurring invasive cancers: neighbours but not always relatives. Gut 67(9):1652\u20131662. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/29500184.","journal-title":"Gut"},{"issue":"5","key":"5062_CR5","doi-asserted-by":"publisher","first-page":"738","DOI":"10.1016\/j.pan.2017.07.007","volume":"17","author":"M Tanaka","year":"2017","unstructured":"Tanaka M, Fernandez-Del Castillo C, Kamisawa T, Jang JY, Levy P, Ohtsuka T, Salvia R, Shimizu Y, Tada M, Wolfgang CL (2017) Revisions of international consensus Fukuoka guidelines for the management of IPMN of the pancreas. Pancreatology 17(5):738\u2013753. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/28735806.","journal-title":"Pancreatology"},{"issue":"6","key":"5062_CR6","doi-asserted-by":"publisher","first-page":"1021","DOI":"10.1097\/XCS.0000000000000176","volume":"234","author":"P Zelga","year":"2022","unstructured":"Zelga P, Hernandez-Barco YG, Qadan M, Ferrone CR, Kambadakone A, Horick N, Jah A, Warshaw AL, Lillemoe KD, Balakrishnan A, Fernandez-Del Castillo C (2022) Number of Worrisome Features and Risk of Malignancy in Intraductal Papillary Mucinous Neoplasm. J Am Coll Surg 234(6):1021\u20131030. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/35703792.","journal-title":"J Am Coll Surg"},{"issue":"4","key":"5062_CR7","doi-asserted-by":"publisher","first-page":"257","DOI":"10.1007\/s11604-018-0726-3","volume":"36","author":"K Yasaka","year":"2018","unstructured":"Yasaka K, Akai H, Kunimatsu A, Kiryu S, Abe O (2018) Deep learning with convolutional neural network in radiology. Jpn J Radiol 36(4):257\u2013272. http:\/\/www.ncbi.nlm.nih.gov\/pubmed\/29498017.","journal-title":"Jpn J Radiol"},{"issue":"6","key":"5062_CR8","doi-asserted-by":"publisher","first-page":"e220133","DOI":"10.1148\/rg.220133","volume":"43","author":"S Kiryu","year":"2023","unstructured":"Kiryu S, Akai H, Yasaka K, Tajima T, Kunimatsu A, Yoshioka N, Akahane M, Abe O, Ohtomo K (2023) Clinical Impact of Deep Learning Reconstruction in MRI. Radiographics 43(6):e220133. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/37200221.","journal-title":"Radiographics"},{"issue":"5","key":"5062_CR9","doi-asserted-by":"publisher","first-page":"476","DOI":"10.1007\/s11604-021-01225-2","volume":"40","author":"K Yasaka","year":"2022","unstructured":"Yasaka K, Akai H, Sugawara H, Tajima T, Akahane M, Yoshioka N, Kabasawa H, Miyo R, Ohtomo K, Abe O, Kiryu S (2022) Impact of deep learning reconstruction on intracranial 1.5 T magnetic resonance angiography. Jpn J Radiol 40(5):476\u2013483. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/34851499.","journal-title":"Jpn J Radiol"},{"issue":"4","key":"5062_CR10","doi-asserted-by":"publisher","first-page":"1280","DOI":"10.1007\/s00261-023-03834-z","volume":"48","author":"N Okimoto","year":"2023","unstructured":"Okimoto N, Yasaka K, Kaiume M, Kanemaru N, Suzuki Y, Abe O (2023) Improving detection performance of hepatocellular carcinoma and interobserver agreement for liver imaging reporting and data system on CT using deep learning reconstruction. Abdom Radiol (NY) 48(4):1280\u20131289. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/36757454.","journal-title":"Abdom Radiol (NY)"},{"issue":"3","key":"5062_CR11","doi-asserted-by":"publisher","first-page":"887","DOI":"10.1148\/radiol.2017170706","volume":"286","author":"K Yasaka","year":"2018","unstructured":"Yasaka K, Akai H, Abe O, Kiryu S (2018) Deep Learning with Convolutional Neural Network for Differentiation of Liver Masses at Dynamic Contrast-enhanced CT: A Preliminary Study. Radiology 286(3):887\u2013896. http:\/\/www.ncbi.nlm.nih.gov\/pubmed\/29059036.","journal-title":"Radiology"},{"issue":"1","key":"5062_CR12","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1148\/radiol.2018180901","volume":"290","author":"D Ueda","year":"2019","unstructured":"Ueda D, Yamamoto A, Nishimori M, Shimono T, Doishita S, Shimazaki A, Katayama Y, Fukumoto S, Choppin A, Shimahara Y, Miki Y (2019) Deep Learning for MR Angiography: Automated Detection of Cerebral Aneurysms. Radiology 290(1):187\u2013194. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/30351253.","journal-title":"Radiology"},{"issue":"1","key":"5062_CR13","doi-asserted-by":"publisher","first-page":"146","DOI":"10.1148\/radiol.2017171928","volume":"287","author":"K Yasaka","year":"2018","unstructured":"Yasaka K, Akai H, Kunimatsu A, Abe O, Kiryu S (2018) Liver Fibrosis: Deep Convolutional Neural Network for Staging by Using Gadoxetic Acid-enhanced Hepatobiliary Phase MR Images. Radiology 287(1):146\u2013155. http:\/\/www.ncbi.nlm.nih.gov\/pubmed\/29239710.","journal-title":"Radiology"},{"key":"5062_CR14","doi-asserted-by":"crossref","unstructured":"Kanemaru N, Yasaka K, Okimoto N, Sato M, Nomura T, Morita Y, Katayama A, Kiryu S, Abe O (2025) Efficacy of Fine-Tuned Large Language Model in CT Protocol Assignment as Clinical Decision-Supporting System. J Imaging Inform Med https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/39909993.","DOI":"10.1007\/s10278-025-01433-6"},{"key":"5062_CR15","doi-asserted-by":"crossref","unstructured":"Kanzawa J, Yasaka K, Fujita N, Fujiwara S, Abe O (2024) Automated classification of brain MRI reports using fine-tuned large language models. Neuroradiology https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/38995393.","DOI":"10.1007\/s00234-024-03427-7"},{"key":"5062_CR16","doi-asserted-by":"crossref","unstructured":"Yasaka K, Nomura T, Kamohara J, Hirakawa H, Kubo T, Kiryu S, Abe O (2024) Classification of Interventional Radiology Reports into Technique Categories with a Fine-Tuned Large Language Model. J Imaging Inform Med https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/39673010.","DOI":"10.1007\/s10278-024-01370-w"},{"key":"5062_CR17","doi-asserted-by":"crossref","unstructured":"Kanemaru N, Yasaka K, Fujita N, Kanzawa J, Abe O (2024) The Fine-Tuned Large Language Model for Extracting the Progressive Bone Metastasis from Unstructured Radiology Reports. J Imaging Inform Med https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/39187702.","DOI":"10.1007\/s10278-024-01242-3"},{"key":"5062_CR18","doi-asserted-by":"crossref","unstructured":"Yasaka K, Kanzawa J, Kanemaru N, Koshino S, Abe O (2024) Fine-Tuned Large Language Model for Extracting Patients on Pretreatment for Lung Cancer from a Picture Archiving and Communication System Based on Radiological Reports. J Imaging Inform Med https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/38955964.","DOI":"10.1007\/s10278-024-01186-8"},{"issue":"1","key":"5062_CR19","doi-asserted-by":"publisher","first-page":"e241073","DOI":"10.1148\/radiol.241073","volume":"314","author":"CH Savage","year":"2025","unstructured":"Savage CH, Kanhere A, Parekh V, Langlotz CP, Joshi A, Huang H, Doo FX (2025) Open-Source Large Language Models in Radiology: A Review and Tutorial for Practical Research and Clinical Deployment. Radiology 314(1):e241073. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/39873598.","journal-title":"Radiology"},{"issue":"8","key":"5062_CR20","doi-asserted-by":"publisher","first-page":"803","DOI":"10.1080\/00365521.2017.1318948","volume":"52","author":"L Aronsson","year":"2017","unstructured":"Aronsson L, Andersson R, Ansari D (2017) Intraductal papillary mucinous neoplasm of the pancreas - epidemiology, risk factors, diagnosis, and management. Scand J Gastroenterol 52(8):803\u2013815. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/28446039.","journal-title":"Scand J Gastroenterol"},{"issue":"10","key":"5062_CR21","doi-asserted-by":"publisher","first-page":"1100","DOI":"10.1007\/s11604-024-01608-1","volume":"42","author":"SL Walston","year":"2024","unstructured":"Walston SL, Seki H, Takita H, Mitsuyama Y, Sato S, Hagiwara A, Ito R, Hanaoka S, Miki Y, Ueda D (2024) Data set terminology of deep learning in medicine: a historical review and recommendation. Jpn J Radiol 42(10):1100\u20131109. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/38856878.","journal-title":"Jpn J Radiol"},{"issue":"4","key":"5062_CR22","doi-asserted-by":"publisher","first-page":"1256","DOI":"10.4251\/wjgo.v16.i4.1256","volume":"16","author":"S Ren","year":"2024","unstructured":"Ren S, Qian LC, Cao YY, Daniels MJ, Song LN, Tian Y, Wang ZQ (2024) Computed tomography-based radiomics diagnostic approach for differential diagnosis between early- and late-stage pancreatic ductal adenocarcinoma. World J Gastrointest Oncol 16(4):1256\u20131267. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/38660647.","journal-title":"World J Gastrointest Oncol"},{"issue":"1","key":"5062_CR23","doi-asserted-by":"publisher","first-page":"217","DOI":"10.1007\/s11596-022-2535-2","volume":"42","author":"S Ren","year":"2022","unstructured":"Ren S, Tang HJ, Zhao R, Duan SF, Chen R, Wang ZQ (2022) Application of Unenhanced Computed Tomography Texture Analysis to Differentiate Pancreatic Adenosquamous Carcinoma from Pancreatic Ductal Adenocarcinoma. Curr Med Sci 42(1):217\u2013225. https:\/\/www.ncbi.nlm.nih.gov\/pubmed\/35089491.","journal-title":"Curr Med Sci"}],"container-title":["Abdominal Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00261-025-05062-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00261-025-05062-z","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00261-025-05062-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,23]],"date-time":"2026-01-23T06:03:49Z","timestamp":1769148229000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00261-025-05062-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,11]]},"references-count":23,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["5062"],"URL":"https:\/\/doi.org\/10.1007\/s00261-025-05062-z","relation":{},"ISSN":["2366-0058"],"issn-type":[{"value":"2366-0058","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,11]]},"assertion":[{"value":"25 March 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"28 May 2025","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 June 2025","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 June 2025","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}