{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T02:29:35Z","timestamp":1787020175082,"version":"build-2736575974"},"reference-count":21,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T00:00:00Z","timestamp":1775865600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T00:00:00Z","timestamp":1779321600000},"content-version":"vor","delay-in-days":40,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Imaging"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Accurate preoperative staging is vital in bladder cancer management, particularly for assessing muscle invasion. Multiparametric MRI (mpMRI) combined with the Vesical Imaging\u2013Reporting and Data System (VI-RADS) offers a non-invasive and standardized approach for tumor stratification. However, inter-observer variability among radiologists remains a significant limitation. This study investigates the performance of three large language models (LLMs)\u2014ChatGPT (OpenAI; GPT-5.2), Gemini (Google; Gemini 2.0), and Copilot (Microsoft; Copilot based on GPT-5 architecture)\u2014in classifying bladder lesions according to VI-RADS using zero-shot and few-shot prompting strategies.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and Methods<\/jats:title>\n                    <jats:p>A synthetic dataset of 100 simulated bladder cancer cases was developed from expert-crafted synthetic radiology reports based on structured text descriptors and expert consensus, comprising 20 cases for each VI-RADS category (1\u20135). Each case included structured imaging descriptors aligned with VI-RADS scoring criteria. The LLMs were evaluated under zero-shot (no examples provided) and few-shot (with illustrative examples) prompting conditions. Performance metrics included accuracy, macro F1 scores, and Cohen\u2019s kappa, with statistical significance assessed using McNemar\u2019s test.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Few-shot prompting significantly improved the classification performance of ChatGPT (OpenAI; GPT-5.2) (accuracy: 94%, F1: 0.939) and Gemini (Google; Gemini 2.0) (accuracy: 83%, F1: 0.823) compared to zero-shot. In contrast, Copilot (Microsoft; Copilot based on GPT-5 architecture)\u2018s accuracy (53%) and F1 score (0.503) declined under few-shot conditions. ChatGPT (OpenAI; GPT-5.2) demonstrated the highest consistency in identifying high-risk lesions (VI-RADS 4\u20135), followed by Gemini (Google; Gemini 2.0).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Few-shot prompting enhances LLM performance in VI-RADS classification, particularly for ChatGPT (OpenAI; GPT-5.2) and Gemini (Google; Gemini 2.0). These findings highlight the potential of AI tools to support radiological decision-making in bladder cancer staging. Further studies using real imaging data and explainable AI are warranted.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12880-026-02334-y","type":"journal-article","created":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T11:55:15Z","timestamp":1775908515000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Evaluation of large language models for VI-RADS reports: a comparative analysis of zero-shot and few-shot prompting"],"prefix":"10.1186","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3460-9124","authenticated-orcid":false,"given":"Ahmet","family":"Halis","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9152-5600","authenticated-orcid":false,"given":"Deniz","family":"Celiker","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,4,11]]},"reference":[{"key":"2334_CR1","unstructured":"MacVicar D, Husband JE. Radiology in the staging of bladder cancer. Br J Hosp Med. 1994;51(9):454\u201358."},{"issue":"1","key":"2334_CR2","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1016\/j.ajur.2024.06.001","volume":"12","author":"H Gungor","year":"2025","unstructured":"Gungor H, Camtosun A, Topcu I, Karaca L. Evaluating the concordance between vesical imaging reporting and data system scores and bladder tumor histopathology. Asian J Urology. 2025;12(1):87\u201392. https:\/\/doi.org\/10.1016\/j.ajur.2024.06.001.","journal-title":"Asian J Urology"},{"issue":"1","key":"2334_CR3","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.eururo.2009.08.024","volume":"57","author":"GS Kulkarni","year":"2010","unstructured":"Kulkarni GS, Hakenberg OW, Gschwend JE, Thalmann G, Kassouf W, Kamat A, et al. An updated critical analysis of the treatment strategy for newly diagnosed high-grade T1 (previously T1G3) bladder cancer. Eur Urology. 2010;57(1):60\u201370.","journal-title":"Eur Urology"},{"issue":"3","key":"2334_CR4","doi-asserted-by":"publisher","first-page":"649","DOI":"10.1002\/jmri.27090","volume":"52","author":"I Caglic","year":"2020","unstructured":"Caglic I, Panebianco V, Vargas HA, Bura V, Woo S, Pecoraro M, et al. MRI of bladder cancer: local and nodal staging. J Magnetic Reson Imag. 2020;52(3):649\u201367.","journal-title":"J Magnetic Reson Imag"},{"issue":"22","key":"2334_CR5","doi-asserted-by":"publisher","first-page":"5468","DOI":"10.3390\/cancers15225468","volume":"15","author":"O Akin","year":"2023","unstructured":"Akin O, Lema-Dopico A, Paudyal R, Konar AS, Chenevert TL, Malyarenko D, et al. Multiparametric MRI in era of artificial intelligence for bladder cancer therapies. Cancers. 2023;15(22):5468. https:\/\/doi.org\/10.3390\/cancers15225468.","journal-title":"Cancers"},{"key":"2334_CR6","doi-asserted-by":"crossref","unstructured":"Panebianco V, Narumi Y, Altun E, Bochner BH, Efstathiou JA, Hafeez S, et al. Multiparametric magnetic resonance imaging for bladder cancer: development of VI-RADS (Vesical imaging-reporting and data system). Eur Urology. 2018;74(3):294\u2013306.","DOI":"10.1016\/j.eururo.2018.04.029"},{"key":"2334_CR7","unstructured":"Rajpurkar P, et al. CheXNet: radiologist-level pneumonia detection on chest X-Rays with deep learning. 2017. arXiv: 1711.05225."},{"key":"2334_CR8","unstructured":"Heller MT, et al. AI in radiology: impact and applications. Radiographics. 2020;40(5):1294\u2013306."},{"key":"2334_CR9","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1038\/s41591-018-0300-7","volume":"25","author":"EJ Topol","year":"2019","unstructured":"Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019;25:44\u201356.","journal-title":"Nat Med"},{"key":"2334_CR10","first-page":"8099","volume":"33","author":"K Jeblick","year":"2023","unstructured":"Jeblick K, Schachtner B, Dexl J, et al. ChatGPT makes medicine easy to swallow: an exploratory case study on simplified radiology reports. Eur Radiol. 2023;33:8099\u2013104.","journal-title":"Eur Radiol"},{"issue":"3","key":"2334_CR11","doi-asserted-by":"publisher","first-page":"649","DOI":"10.1002\/jmri.27090","volume":"52","author":"I Caglic","year":"2020","unstructured":"Caglic I, Panebianco V, Vargas HA, et al. MRI of bladder cancer: local and nodal staging. J Magn Reson Imag. 2020;52(3):649\u201367.","journal-title":"J Magn Reson Imag"},{"issue":"3","key":"2334_CR12","doi-asserted-by":"publisher","first-page":"294","DOI":"10.1016\/j.eururo.2018.04.029","volume":"74","author":"V Panebianco","year":"2018","unstructured":"Panebianco V, Narumi Y, Altun E, et al. Multiparametric magnetic resonance imaging for bladder cancer: development of VI-RADS. Eur Urol. 2018;74(3):294\u2013306.","journal-title":"Eur Urol"},{"key":"2334_CR13","doi-asserted-by":"publisher","first-page":"368","DOI":"10.1093\/bjr\/tqae236","volume":"98","author":"KL Lee","year":"2025","unstructured":"Lee KL, Kessler DA, Caglic I, Kuo YH, Shaida N, Barrett T. Assessing the performance of ChatGPT and Bard\/Gemini against radiologists for prostate imaging-Reporting and data System classification based on prostate multiparametric MRI text reports. The Br J Radiol. 2025;98, (1167): 368\u201374. https:\/\/doi.org\/10.1093\/bjr\/tqae236.","journal-title":"The Br J Radiol"},{"issue":"8","key":"2334_CR14","doi-asserted-by":"publisher","first-page":"767","DOI":"10.3348\/kjr.2024.0438","volume":"25","author":"E \u00c7amur","year":"2024","unstructured":"\u00c7amur E, Cesur T, G\u00fcne\u015f YC. Comparative evaluation of the accuracies of large language models in answering VI-RADS-related questions. Korean J Radiol. 2024;25(8):767\u201368. https:\/\/doi.org\/10.3348\/kjr.2024.0438.","journal-title":"Korean J Radiol"},{"issue":"1","key":"2334_CR15","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.eururo.2009.08.024","volume":"57","author":"GS Kulkarni","year":"2010","unstructured":"Kulkarni GS, Hakenberg OW, Gschwend JE, et al. An updated critical analysis of the treatment strategy for newly diagnosed high-grade T1 bladder cancer. Eur Urol. 2010;57(1):60\u201370.","journal-title":"Eur Urol"},{"issue":"3","key":"2334_CR16","doi-asserted-by":"publisher","first-page":"512","DOI":"10.1016\/j.jacr.2017.12.028","volume":"15","author":"ML Giger","year":"2018","unstructured":"Giger ML. Machine learning in medical imaging. J Am Coll Radiol. 2018;15(3):512\u201320.","journal-title":"J Am Coll Radiol"},{"key":"2334_CR17","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1038\/s41746-019-0148-3","volume":"2","author":"P Shah","year":"2019","unstructured":"Shah P, Kendall F, Khozin S, et al. Artificial intelligence and machine learning in clinical development: a translational perspective. NPJ Digit Med. 2019;2:69.","journal-title":"NPJ Digit Med"},{"key":"2334_CR18","unstructured":"Rajpurkar P, Irvin J, Zhu K, et al. CheXNet: radiologist-level pneumonia detection on chest X-rays with deep learning. arXiv preprint. 2017, arXiv: 1711.05225."},{"key":"2334_CR19","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1038\/s41746-020-00376-2","volume":"4","author":"A Esteva","year":"2021","unstructured":"Esteva A, Chou K, Yeung S, et al. Deep learning-enabled medical computer vision. NPJ Digit Med. 2021;4:5.","journal-title":"NPJ Digit Med"},{"issue":"1","key":"2334_CR20","first-page":"39","volume":"1","author":"W Samek","year":"2017","unstructured":"Samek W, Wiegand T, M\u00fcller KR. Explainable artificial intelligence: understanding, visualizing and interpreting deep learning models. ITU J ICT Discov. 2017;1(1):39\u201348.","journal-title":"ITU J ICT Discov"},{"key":"2334_CR21","unstructured":"Doshi-Velez F, Kim B. Towards a rigorous science of interpretable machine learning. arXiv preprint. 2017, arXiv: 1702.08608."}],"container-title":["BMC Medical Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12880-026-02334-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-026-02334-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-026-02334-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T21:50:38Z","timestamp":1781646638000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s12880-026-02334-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,11]]},"references-count":21,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["2334"],"URL":"https:\/\/doi.org\/10.1186\/s12880-026-02334-y","relation":{},"ISSN":["1471-2342"],"issn-type":[{"value":"1471-2342","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,11]]},"assertion":[{"value":"9 January 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 March 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 April 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study was based entirely on synthetically generated clinical scenarios and did not involve any real patient data. Therefore, ethics committee approval and informed consent to participate were not required.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"265"}}