{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,27]],"date-time":"2026-10-27T16:41:11Z","timestamp":1793119271568,"version":"4.5.0"},"reference-count":25,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2026,2,23]],"date-time":"2026-02-23T00:00:00Z","timestamp":1771804800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Chonnam National University Hospital Biomedical Research Institute","award":["BCRI26035"],"award-info":[{"award-number":["BCRI26035"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JCM"],"abstract":"<jats:p>Background: Large language models (LLMs) may assist radiologists in interpreting brain tumor MRI. We compared the diagnostic accuracy of ChatGPT-4o and Claude 3.5 Sonnet with that of board-certified radiologists and trainees, and evaluated whether LLM assistance could enhance diagnostic performance. Methods: A total of 127 histologically confirmed brain tumor cases were included. Two LLMs analyzed representative MRI images together with structured radiologic reports, whereas two board-certified radiologists and three trainees reviewed representative images with basic demographic information only. All participants generated up to three differential diagnoses per case. The accuracy of the primary diagnosis and the accuracy of the top-three differential diagnoses were calculated and compared. Following the initial readings, LLM-generated differential diagnoses were provided to the readers, and their post-assistance diagnostic performance was re-evaluated. Results: Claude 3.5 Sonnet achieved a primary diagnostic accuracy of 50.4% and a top-three differential accuracy of 85.0%, comparable to ChatGPT-4o (44.9% and 82.7%, respectively). Radiologists demonstrated a higher primary diagnostic accuracy (69.3%, p &lt; 0.001) compared to LLMs, but a similar top-three differential accuracy (80.7%). In contrast, trainees showed a primary diagnostic accuracy (48.0%) comparable to LLMs, but a lower top-three differential accuracy (62.5%) than LLMs. With LLM assistance, radiologists exhibited a significant improvement in the top-three differential accuracy (from 80.7% to 90.2%, p &lt; 0.001), and trainees showed significant improvements in both the primary and top-three differential accuracy (from 48.0% to 58.8%, p &lt; 0.001, and from 62.5% to 81.1%, p &lt; 0.001, respectively). Conclusion: LLMs demonstrated the ability to expand differential diagnostic considerations when operating on structured imaging inputs. LLM assistance was associated with improved trainee performance in this constrained experimental setting. These findings should be interpreted cautiously and require validation under balanced input conditions and clinically realistic workflows.<\/jats:p>","DOI":"10.3390\/jcm15041673","type":"journal-article","created":{"date-parts":[[2026,2,23]],"date-time":"2026-02-23T16:22:55Z","timestamp":1771863775000},"page":"1673","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Impact of Large Language Model Assistance on Radiologists\u2019 Diagnostic Performance for Brain Tumors by Experience Level"],"prefix":"10.3390","volume":"15","author":[{"given":"Chae Won","family":"Song","sequence":"first","affiliation":[{"name":"Department of Radiology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8657-1052","authenticated-orcid":false,"given":"Byung Hyun","family":"Baek","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea"},{"name":"Department of Radiology, Chonnam National University Medical School, Gwangju 61469, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Seul Kee","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Medical School, Gwangju 61469, Republic of Korea"},{"name":"Department of Radiology, Chonnam National University Hwasun Hospital, Hwasun 58128, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Woong","family":"Yoon","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea"},{"name":"Department of Radiology, Chonnam National University Medical School, Gwangju 61469, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1200-9462","authenticated-orcid":false,"given":"Yun Young","family":"Lee","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea"},{"name":"Department of Radiology, Chonnam National University Medical School, Gwangju 61469, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ilwoo","family":"Park","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea"},{"name":"Department of Radiology, Chonnam National University Medical School, Gwangju 61469, Republic of Korea"},{"name":"Department of Data Science, Chonnam National University, Gwangju 61180, Republic of Korea"},{"name":"Department of Artificial Intelligence Convergence, Chonnam National University, Gwangju 61180, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jae Hyun","family":"Park","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1497-1604","authenticated-orcid":false,"given":"Seol Bin","family":"Park","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4607-9909","authenticated-orcid":false,"given":"In Woo","family":"Choi","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chonnam National University Hwasun Hospital, Hwasun 58128, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,2,23]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Ali, H., Qadir, J., Alam, T., Househ, M., and Shah, Z. (2023, January 16\u201317). ChatGPT and Large Language Models in Healthcare: Opportunities and Risks. Proceedings of the 2023 IEEE International Conference on Artificial Intelligence, Blockchain, and Internet of Things (AIBThings), Mount Pleasant, MI, USA.","DOI":"10.1109\/AIBThings58340.2023.10291020"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Wada, A., Akashi, T., Shih, G., Hagiwara, A., Nishizawa, M., Hayakawa, Y., Kikuta, J., Shimoji, K., Sano, K., and Kamagata, K. (2024). Optimizing GPT-4 Turbo Diagnostic Accuracy in Neuroradiology through Prompt Engineering and Confidence Thresholds. Diagnostics, 14.","DOI":"10.3390\/diagnostics14141541"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"269","DOI":"10.1055\/s-0043-1777289","article-title":"Radiological Differential Diagnoses Based on Cardiovascular and Thoracic Imaging Patterns: Perspectives of Four Large Language Models","volume":"34","author":"Sarangi","year":"2024","journal-title":"Indian J. Radiol. Imaging"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"e000451","DOI":"10.1136\/bmjno-2023-000451","article-title":"Evaluating the limits of AI in Medical Specialisation: ChatGPT\u2019s Performance on the UK Neurology Specialty Certificate Examination","volume":"5","author":"Giannos","year":"2023","journal-title":"BMJ Neurol. Open"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1938","DOI":"10.1007\/s00330-024-11032-8","article-title":"Comparative Analysis of GPT-4-Based ChatGPT\u2019s Diagnostic Performance with Radiologists Using Real-World Radiology Reports of Brain Tumors","volume":"35","author":"Mitsuyama","year":"2025","journal-title":"Eur. Radiol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1007\/s00234-023-03252-4","article-title":"Accuracy of ChatGPT-generated diagnosis from patient history and imaging findings in neuroradiology cases","volume":"66","author":"Horiuchi","year":"2024","journal-title":"Neuroradiology"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Wada, A., Akashi, T., Shih, G., Hagiwara, A., Nishizawa, M., Hayakawa, Y., Kikuta, J., Shimoji, K., Sano, K., and Kamagata, K. (2024). Prompt Engineering Strategies Improve the Diagnostic Accuracy of GPT-4 Turbo in Neuroradiology Cases. medRxiv, medRxiv:2024.04.29.24306583.","DOI":"10.1101\/2024.04.29.24306583"},{"key":"ref_8","first-page":"e67766","article-title":"Comparative Evaluation of AI Models Such as ChatGPT 3.5, ChatGPT 4.0, and Google Gemini in Neuroradiology Diagnostics","volume":"16","author":"Gupta","year":"2024","journal-title":"Cureus"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e240273","DOI":"10.1148\/radiol.240273","article-title":"Comparing Diagnostic Accuracy of Radiologists versus GPT-4V and Gemini Pro Vision Using Image Inputs from Diagnosis Please Cases","volume":"312","author":"Suh","year":"2024","journal-title":"Radiology"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"e231040","DOI":"10.1148\/radiol.231040","article-title":"ChatGPT\u2019s Diagnostic Performance from Patient History and Imaging Findings on the Diagnosis Please Quizzes","volume":"308","author":"Ueda","year":"2023","journal-title":"Radiology"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"779","DOI":"10.1007\/s00062-024-01426-y","article-title":"Comparing the Diagnostic Performance of GPT-4-based ChatGPT, GPT-4V-based ChatGPT and Radiologists in Challenging Neuroradiology Cases","volume":"34","author":"Horiuchi","year":"2024","journal-title":"Clin. Neuroradiol."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Nakaura, T., Uetani, H., Yoshida, N., Kobayashi, N., Nagayama, Y., Kidoh, M., Kuroda, J.I., Mukasa, A., and Hirai, T. (2025). Intra-axial primary brain tumor differentiation: Comparing large language models on structured MRI reports vs. radiologists on images. Eur. Radiol., advance online publication.","DOI":"10.1007\/s00330-025-11924-3"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"383","DOI":"10.1515\/revneuro-2021-0101","article-title":"Current Uses, Emerging Applications, and Clinical Integration of Artificial Intelligence in Neuroradiology","volume":"33","author":"Fiani","year":"2022","journal-title":"Rev. Neurosci."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"6099","DOI":"10.1007\/s00405-024-08828-1","article-title":"Assessing the Use of the Novel Tool Claude 3 in Comparison to ChatGPT 4.0 as an Artificial Intelligence Tool in the Diagnosis and Therapy of Primary Head and Neck Cancer Cases","volume":"281","author":"Schmidl","year":"2024","journal-title":"Eur. Arch. Otorhinolaryngol."},{"key":"ref_15","first-page":"e41435","article-title":"ChatGPT in Radiology: The Advantages and Limitations of AI for Medical Imaging Diagnosis","volume":"15","author":"Srivastav","year":"2023","journal-title":"Cureus"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Khosravi, P., and Schweitzer, M. (2023). Artificial intelligence in neuroradiology: A scoping review of ethical challenges. Front. Radiol., 3.","DOI":"10.3389\/fradi.2023.1149461"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"101566","DOI":"10.1016\/j.dcn.2025.101566","article-title":"Neural, Cognitive and Psychopathological Signatures of a Prosocial or Delinquent Peer Environment during Early Adolescence","volume":"73","author":"Liu","year":"2025","journal-title":"Dev. Cogn. Neurosci."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"e231593","DOI":"10.1148\/radiol.231593","article-title":"Quantitative Evaluation of Large Language Models to Streamline Radiology Report Impressions: A Multimodal Retrospective Analysis","volume":"310","author":"Doshi","year":"2024","journal-title":"Radiology"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"451","DOI":"10.1038\/s41586-025-08869-4","article-title":"Towards Accurate Differential Diagnosis with Large Language Models","volume":"642","author":"McDuff","year":"2025","journal-title":"Nature"},{"key":"ref_20","first-page":"4678554","article-title":"ChatGPT-powered deep learning: Elevating brain tumor detection in MRI scans","volume":"2024","author":"Rawas","year":"2024","journal-title":"Appl. Comput. Inform."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"685","DOI":"10.1007\/s11604-024-01552-0","article-title":"The Impact of Large Language Models on Radiology: A Guide for Radiologists on the Latest Innovations in AI","volume":"42","author":"Nakaura","year":"2024","journal-title":"Jpn J. Radiol."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"80","DOI":"10.4274\/dir.2023.232417","article-title":"Large language models in radiology: Fundamentals, applications, ethical considerations, risks, and future directions","volume":"30","author":"Stanzione","year":"2024","journal-title":"Diagn. Interv. Radiol."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"97","DOI":"10.5853\/jos.2025.01725","article-title":"Dynamic Interactions Between Hemispheres Reveal a Compensatory Pathway for Motor Recovery in Moderate-to-Severe Subcortical Stroke","volume":"28","author":"Fan","year":"2026","journal-title":"J. Stroke"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"857622","DOI":"10.3389\/fonc.2022.857622","article-title":"Review of current principles of the diagnosis and management of brain metastases","volume":"12","author":"Brenner","year":"2022","journal-title":"Front Oncol."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"20230203","DOI":"10.1259\/bjr.20230203","article-title":"ChatGPT from radiologists\u2019 perspective","volume":"96","author":"Sendur","year":"2023","journal-title":"Br. J. Radiol."}],"container-title":["Journal of Clinical Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/4\/1673\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,23]],"date-time":"2026-02-23T16:27:27Z","timestamp":1771864047000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/4\/1673"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,23]]},"references-count":25,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,2]]}},"alternative-id":["jcm15041673"],"URL":"https:\/\/doi.org\/10.3390\/jcm15041673","relation":{},"ISSN":["2077-0383"],"issn-type":[{"value":"2077-0383","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,23]]}}}