{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T22:38:55Z","timestamp":1789166335943,"version":"build-2803163510"},"reference-count":25,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T00:00:00Z","timestamp":1781136000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To evaluate the performance of DeepSeek (R1 version), an open-source large language model, in three core clinical scenarios: answering patients\u2019 common questions, interpreting PET\/CT reports with follow-up inquiries, and diagnosing complex cases, and comparison with GPT-5.3, to verify the clinical applicability of DeepSeek-R1 as an alternative AI assistant.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      A total of 39 standardized tasks were assigned to both models, including responding to 15 frequently asked questions about [\n                      <jats:sup>18<\/jats:sup>\n                      F]FDG PET\/CT, interpreting 12 anonymized reports of lung cancer and lymphoma (with follow-up inquiries regarding tumor staging or treatment), and providing primary and differential diagnoses for 10 difficult cases. Both models were accessed via their official platforms with default parameters, and all prompts and evaluation criteria were kept identical for cross-model comparison. Two senior nuclear medicine physicians independently rated the model responses using a 4-point standardized scale (assessing appropriateness, helpfulness, inter-trial consistency, and reference validity) and a binary scale for empathy; Cohen\u2019s Kappa coefficient was used to evaluate inter-rater agreement. McNemar\u2019s test was used to compare paired proportions of appropriateness, empathy, and response inconsistency between the two models.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Across the 39 tasks, DeepSeek-R1 achieved 94.9% appropriateness and 100% helpfulness. Specifically, 91.7% of responses to follow-up inquiries about tumor staging or treatment were rated empathetic. However, 7.7% of regenerated responses showed substantial inconsistencies, primarily in tumor staging, and only 37% of cited references were fully valid, with 11.1% being invalid. GPT-5.3 exhibited equivalent core performance to DeepSeek-R1 with 94.9% appropriateness and 100% helpfulness, a slightly lower substantial inconsistency rate (5.1%), favorable reference validity (33% fully valid, 7.4% invalid), but a notably lower empathy score (66.7%) for follow-up inquiries. McNemar tests showed identical appropriateness (\n                      <jats:italic>p<\/jats:italic>\n                      \u202f=\u202f1.00) and no significant difference in inconsistency (\n                      <jats:italic>p<\/jats:italic>\n                      \u202f=\u202f1.00, 95% CI 0.60\u201314.80) between models. DeepSeek-R1 had higher empathy, the difference was not significant (\n                      <jats:italic>p<\/jats:italic>\n                      \u202f=\u202f0.25, 95% CI 0.09\u20130.66). For the 10 identical difficult cases, both models reached 10% primary diagnosis accuracy and 60% differential diagnosis accuracy.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>DeepSeek-R1 and GPT-5.3 have complementary strengths but similar reference hallucination issues and cannot replace clinicians. DeepSeek-R1 is a cost-effective auxiliary tool, with future optimization needed for consistency, diagnostic accuracy and reference validity.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1785671","type":"journal-article","created":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T04:37:45Z","timestamp":1781152665000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Preliminary evaluation of DeepSeek-R1 and GPT-5.3 in selected PET\/CT clinical scenarios: patient preparation, report interpretation, and diagnostic reasoning"],"prefix":"10.3389","volume":"13","author":[{"given":"Runze","family":"Duan","sequence":"first","affiliation":[{"name":"Hebei Medical University","place":["Shijiazhuang, China"]},{"name":"Department of Nuclear Medicine, Hebei General Hospital","place":["Shijiazhuang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jing","family":"Pang","sequence":"additional","affiliation":[{"name":"Hebei Medical University","place":["Shijiazhuang, China"]},{"name":"Department of Nuclear Medicine, Hebei General Hospital","place":["Shijiazhuang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lu","family":"Zheng","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Hebei General Hospital","place":["Shijiazhuang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ziyu","family":"Guo","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Hebei General Hospital","place":["Shijiazhuang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianyue","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Hebei General Hospital","place":["Shijiazhuang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanzhu","family":"Bian","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Hebei General Hospital","place":["Shijiazhuang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yujing","family":"Hu","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Hebei General Hospital","place":["Shijiazhuang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,6,11]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"286","DOI":"10.1053\/j.semnuclmed.2021.10.005","article-title":"Advances in PET\/CT technology: an update","volume":"52","author":"Aide","year":"2022","journal-title":"Semin Nucl Med"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"3058","DOI":"10.1007\/s00259-025-07159-5","article-title":"Clinical benefits of LAFOV PET\/CT in growing demand for molecular imaging","volume":"52","author":"Vogel","year":"2025","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1093\/inthealth\/ihaa007","article-title":"Artificial intelligence: opportunities and implications for the health workforce","volume":"12","author":"Hazarika","year":"2020","journal-title":"Int Health"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"238","DOI":"10.1055\/s-0045-1807251","article-title":"Global challenges in the provision of Theranostic services the International Atomic Energy Agency (IAEA) perspective","volume":"24","author":"Brink","year":"2025","journal-title":"World J Nucl Med"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"641","DOI":"10.1016\/S0140-6736(23)00216-7","article-title":"The promise of large language models in health care","volume":"401","author":"Arora","year":"2023","journal-title":"Lancet"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"1930","DOI":"10.1038\/s41591-023-02448-8","article-title":"Large language models in medicine","volume":"29","author":"Thirunavukarasu","year":"2023","journal-title":"Nat Med"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1038\/s41586-023-06291-2","article-title":"Large language models encode clinical knowledge","volume":"620","author":"Singhal","year":"2023","journal-title":"Nature"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"2550","DOI":"10.1038\/s41591-025-03726-3","article-title":"Comparative benchmarking of the DeepSeek large language model on medical tasks and clinical reasoning","volume":"31","author":"Tordjman","year":"2025","journal-title":"Nat Med"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1038\/d41586-025-00229-6","article-title":"China's cheap, open AI model DeepSeek thrills scientists","volume":"638","author":"Gibney","year":"2025","journal-title":"Nature"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1638904","DOI":"10.3389\/frai.2025.1638904","article-title":"The potential of DeepSeek for AI-aided diagnosis of antibody-positive autoimmune encephalitis: a single-center, retrospective, observational study","volume":"8","author":"Meng","year":"2025","journal-title":"Front Artif Intell"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1876","DOI":"10.2967\/jnumed.123.266114","article-title":"ChatGPT: can you prepare my patients for [(18)F]FDG PET\/CT and explain my reports?","volume":"64","author":"Rogasch","year":"2023","journal-title":"J Nucl Med"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"120","DOI":"10.1053\/j.semnuclmed.2020.08.001","article-title":"Ethical and legal challenges of artificial intelligence in nuclear medicine","volume":"51","author":"Currie","year":"2021","journal-title":"Semin Nucl Med"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"188","DOI":"10.2967\/jnumed.121.263703","article-title":"Artificial intelligence in nuclear medicine: opportunities, challenges, and responsibilities toward a trustworthy ecosystem","volume":"64","author":"Saboury","year":"2023","journal-title":"J Nucl Med"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"189","DOI":"10.1016\/j.aopr.2025.05.001","article-title":"DeepSeek-R1 outperforms Gemini 2.0 pro, OpenAI O1, and O3-mini in bilingual complex ophthalmology reasoning","volume":"5","author":"Xu","year":"2025","journal-title":"Adv Ophthalmol Pract Res"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"871","DOI":"10.1186\/s12903-025-06267-w","article-title":"Comparative evaluation of responses from DeepSeek-R1, ChatGPT-o1, ChatGPT-4, and dental GPT chatbots to patient inquiries about dental and maxillofacial prostheses","volume":"25","author":"\u00d6zcivelek","year":"2025","journal-title":"BMC Oral Health"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"638","DOI":"10.1186\/s12903-025-06034-x","article-title":"Can deepseek and ChatGPT be used in the diagnosis of oral pathologies?","volume":"25","author":"Kaygisiz","year":"2025","journal-title":"BMC Oral Health"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"1581","DOI":"10.1186\/s12909-025-08184-6","article-title":"DeepSeek-R1 for automated scoring in radiology residency examinations: an agreement and test-retest reliability study","volume":"25","author":"Niu","year":"2025","journal-title":"BMC Med Educ"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"366","DOI":"10.1016\/j.hrtlng.2025.08.007","article-title":"Comparing DeepSeek and GPT-4o in ECG interpretation: is AI improving over time?","volume":"75","author":"G\u00fcnay","year":"2026","journal-title":"Heart Lung"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1136\/svn-2017-000101","article-title":"Artificial intelligence in healthcare: past, present and future","volume":"2","author":"Jiang","year":"2017","journal-title":"Stroke Vasc Neurol"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"1744389","DOI":"10.3389\/fcell.2026.1744389","article-title":"Comparative performance of GPT-4, GPT-o3, GPT-5, Gemini-3-flash, and DeepSeek-R1 in ophthalmology question answering","volume":"14","author":"Zhang","year":"2026","journal-title":"Front Cell Dev Biol"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"e43313","DOI":"10.7759\/cureus.43313","article-title":"Artificial hallucinations by Google bard: think before you leap","volume":"15","author":"Kumar","year":"2023","journal-title":"Cureus."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"e37432","DOI":"10.7759\/cureus.37432","article-title":"Exploring the boundaries of reality: investigating the phenomenon of artificial intelligence hallucination in scientific writing through ChatGPT references","volume":"15","author":"Athaluri","year":"2023","journal-title":"Cureus"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"163","DOI":"10.1038\/s41746-025-01475-8","article-title":"Large language model agents can use tools to perform clinical calculations","volume":"8","author":"Goodell","year":"2025","journal-title":"NPJ Digit Med"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"60","DOI":"10.3390\/medicina60030445","article-title":"Integrating retrieval-augmented generation with large language models in nephrology: advancing practical applications","volume":"60","author":"Miao","year":"2024","journal-title":"Medicina (Kaunas)"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"12897","DOI":"10.1609\/aaai.v40i15.38288","article-title":"PET2Rep: towards vision-language model-drived automated radiology report generation for positron emission tomography Proc AAAI Conf","volume":"40","author":"Zhang","year":"2026","journal-title":"Artif Intell"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1785671\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T04:37:54Z","timestamp":1781152674000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1785671\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,11]]},"references-count":25,"alternative-id":["10.3389\/fmed.2026.1785671"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1785671","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,11]]},"article-number":"1785671"}}