{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T06:27:57Z","timestamp":1786429677635,"version":"build-2736575974"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T00:00:00Z","timestamp":1786406400000},"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. Digit. Health"],"abstract":"<jats:p>\n                    In regions with high thalassemia prevalence, such as southern China and Southeast Asia, chronic shortages of professional genetic counseling resources have driven interest in large language models (LLMs) as auxiliary tools, yet their performance and safety boundaries in this setting remain uncharacterized. This single-center retrospective study evaluated four LLMs (ChatGPT-5.2 Thinking, DeepSeek-V3.2 chat, Gemini 3 Flash, and Grok 4.1 Fast) using 1,080 standardized knowledge questions administered across five independent sessions and 150 real-world clinical cases scored by six senior experts across eight dimensions. All models exceeded 90% accuracy on single-choice and true-false questions. Between-model differences were most pronounced in multiple-choice questions, where ChatGPT-5.2 Thinking achieved the highest accuracy (87.28%\u2009\u00b1\u20091.69%), significantly outperforming Grok 4.1 Fast (72.06%\u2009\u00b1\u20091.69%,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&amp;lt;\u20090.001). DeepSeek-V3.2 chat showed lower cross-session consistency than the other three models. In clinical case analysis, all models scored below human expert levels overall, with ChatGPT-5.2 Thinking performing closest to experts. Test report interpretation was generally adequate, whereas larger gaps emerged in genetic risk estimation, phenotype prediction, and counseling recommendations (all\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&amp;lt;\u20090.001 vs. human experts, with limited exceptions for ChatGPT-5.2 Thinking in\n                    <jats:italic>\u03b2<\/jats:italic>\n                    -thalassemia subtypes). All 145 severe errors were confined to\n                    <jats:italic>\u03b1<\/jats:italic>\n                    -thalassemia and\n                    <jats:italic>\u03b1<\/jats:italic>\n                    -combined-\n                    <jats:italic>\u03b2<\/jats:italic>\n                    -thalassemia subtypes, concentrated in phenotype prediction (76\/145, 52.4%) and risk estimation (40\/145, 27.6%). LLMs may support knowledge retrieval and structured report interpretation in thalassemia genetic counseling but should not be used independently for risk assessment or final counseling decisions, particularly in complex\n                    <jats:italic>\u03b1<\/jats:italic>\n                    -related cases.\n                  <\/jats:p>","DOI":"10.3389\/fdgth.2026.1902480","type":"journal-article","created":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T05:35:55Z","timestamp":1786426555000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Performance and limitations of four large language models in genetic counseling for thalassemia"],"prefix":"10.3389","volume":"8","author":[{"given":"Wenfu","family":"Zhong","sequence":"first","affiliation":[{"name":"Department of Clinical Laboratory, Guigang Clinical Medical Research Center for Medical Laboratory, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingwen","family":"Huang","sequence":"additional","affiliation":[{"name":"Department of Respiratory and Critical Care Medicine, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengsi","family":"Wei","sequence":"additional","affiliation":[{"name":"Department of Clinical Laboratory, Guigang Clinical Medical Research Center for Medical Laboratory, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qingpeng","family":"Liang","sequence":"additional","affiliation":[{"name":"Department of Clinical Laboratory, Guigang Clinical Medical Research Center for Medical Laboratory, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuanwu","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Clinical Laboratory, Guigang Clinical Medical Research Center for Medical Laboratory, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinrong","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Clinical Laboratory, Guigang Clinical Medical Research Center for Medical Laboratory, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinjiang","family":"Mao","sequence":"additional","affiliation":[{"name":"Department of Obstetrics, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ying","family":"Qin","sequence":"additional","affiliation":[{"name":"Department of Obstetrics, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yifan","family":"Sun","sequence":"additional","affiliation":[{"name":"Department of Clinical Laboratory, Guigang Clinical Medical Research Center for Medical Laboratory, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yishan","family":"Liang","sequence":"additional","affiliation":[{"name":"Department of Clinical Laboratory, Guigang Clinical Medical Research Center for Medical Laboratory, Guigang City People's Hospital","place":["Guigang, Guangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,11]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"102619","DOI":"10.1016\/j.eclinm.2024.102619","article-title":"Global, regional, and national burden of thalassemia, 1990\u20132021: a systematic analysis for the global burden of disease study 2021","volume":"72","author":"Tuo","year":"2024","journal-title":"EClinicalMed"},{"key":"B2","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1186\/s13690-025-01575-7","article-title":"Prevalence, mutation distribution, and economic burden of thalassemia in China: a systematic review and regional analysis","volume":"83","author":"Cai","year":"2025","journal-title":"Arch Public Health"},{"key":"B3","doi-asserted-by":"publisher","first-page":"101074","DOI":"10.1016\/j.blre.2023.101074","article-title":"Thalassaemia in China","volume":"60","author":"Wang","year":"2023","journal-title":"Blood Rev"},{"key":"B4","doi-asserted-by":"publisher","first-page":"e732","DOI":"10.1097\/HS9.0000000000000732","article-title":"2021 thalassaemia international federation guidelines for the management of transfusion-dependent thalassemia","volume":"6","author":"Farmakis","year":"2022","journal-title":"Hemasphere"},{"key":"B5","first-page":"577","article-title":"Expert consensus on the management of thalassemia during pregnancy","volume":"23","year":"2020","journal-title":"Chin J Perinat Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1793","DOI":"10.1038\/s41436-021-01203-z","article-title":"Screening for autosomal recessive and X-linked conditions during pregnancy and preconception: a practice resource of the American college of medical genetics and genomics (ACMG)","volume":"23","author":"Gregg","year":"2021","journal-title":"Genet Med"},{"key":"B7","first-page":"243","article-title":"Clinical practice guidelines for \u03b2-thalassemia","volume":"37","year":"2020","journal-title":"Chin J Med Genet"},{"key":"B8","doi-asserted-by":"publisher","first-page":"235","DOI":"10.3760\/cma.j.issn.1003-9406.2020.03.003","article-title":"Clinical practice guidelines for \u03b1-thalassemia","volume":"37","year":"2020","journal-title":"Chin J Med Genet"},{"key":"B9","doi-asserted-by":"publisher","first-page":"e24696","DOI":"10.1002\/jcla.24696","article-title":"Analysis of genotype-phenotype correlation in patients with \u03b1-thalassemia from fujian province, southeastern China","volume":"36","author":"Pan","year":"2022","journal-title":"J Clin Lab Anal"},{"key":"B10","doi-asserted-by":"publisher","first-page":"440","DOI":"10.1002\/jha2.923","article-title":"Alpha globin gene alterations modifying the phenotype of homozygous beta thalassaemia","volume":"5","author":"Shaw","year":"2024","journal-title":"EJHaem"},{"key":"B11","doi-asserted-by":"publisher","first-page":"632","DOI":"10.3324\/haematol.2023.283211","article-title":"Quantitative evaluation of the clinical severity of hemoglobin H disease in a cohort of 591 patients using a scoring system based on regression analysis","volume":"109","author":"Liu","year":"2024","journal-title":"Haematologica"},{"key":"B12","doi-asserted-by":"publisher","first-page":"101887","DOI":"10.1016\/j.gimo.2024.101887","article-title":"The global status of genetic counselors in 2023: what has changed in the past 5 years?","volume":"2","author":"Ormond","year":"2024","journal-title":"Genet Med Open"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1239817","DOI":"10.3389\/fgene.2023.1239817","article-title":"The growing needs of genetic counselling-feasibility in utilization of tele-genetic counselling in Asia and Hong Kong","volume":"14","author":"Chu","year":"2023","journal-title":"Front Genet"},{"key":"B14","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1001\/jama.2024.21700","article-title":"Testing and evaluation of health care applications of large language models: a systematic review","volume":"333","author":"Bedi","year":"2025","journal-title":"JAMA"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1233","DOI":"10.1056\/NEJMsr2214184","article-title":"Benefits, limits, and risks of GPT-4 as an AI chatbot for medicine","volume":"388","author":"Lee","year":"2023","journal-title":"N Engl J Med"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1233","DOI":"10.1038\/s41591-024-03456-y","article-title":"GPT-4 assistance for improvement of physician performance on patient care tasks: a randomized controlled trial","volume":"31","author":"Goh","year":"2025","journal-title":"Nat Med"},{"key":"B17","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1016\/j.ygyno.2024.04.006","article-title":"ChatGPT accurately performs genetic counseling for gynecologic cancers","volume":"183","author":"Patel","year":"2024","journal-title":"Gynecol Oncol"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1685","DOI":"10.1111\/bjh.19738","article-title":"Exploring the role of large language models in haematology: a focused review of applications, benefits and limitations","volume":"205","author":"Mudrik","year":"2024","journal-title":"Br J Haematol"},{"key":"B19","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1186\/s12911-024-02439-w","article-title":"An evaluation of GPT models for phenotype concept recognition","volume":"24","author":"Groza","year":"2024","journal-title":"BMC Med Inform Decis Mak"},{"key":"B20","doi-asserted-by":"publisher","first-page":"2190","DOI":"10.1016\/j.ajhg.2024.08.010","article-title":"Assessing the utility of large language models for phenotype-driven gene prioritization in the diagnosis of rare genetic disease","volume":"111","author":"Kim","year":"2024","journal-title":"Am J Hum Genet"},{"key":"B21","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1038\/s41746-024-01366-4","article-title":"Probabilistic medical predictions of large language models","volume":"7","author":"Gu","year":"2024","journal-title":"NPJ Digit Med"},{"key":"B22","doi-asserted-by":"publisher","first-page":"672","DOI":"10.3390\/diagnostics15060672","article-title":"Evaluating the use of generative artificial intelligence to support genetic counseling for rare diseases","volume":"15","author":"Jeon","year":"2025","journal-title":"Diagnostics (Basel)"},{"key":"B23","doi-asserted-by":"publisher","first-page":"281","DOI":"10.1038\/s41431-024-01782-w","article-title":"Artificial intelligence in clinical genetics","volume":"33","author":"Duong","year":"2025","journal-title":"Eur J Hum Genet"},{"key":"B24","volume-title":"Prevention and Treatment of Thalassemia","year":"2025"},{"key":"B25","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1007\/s44163-026-00863-7","article-title":"Comparative evaluation of large language models in generating clinical insights for HIV associated oral Kaposi sarcoma","volume":"6","author":"Babalola","year":"2026","journal-title":"Discov Artif Intell"},{"key":"B26","doi-asserted-by":"publisher","first-page":"2613","DOI":"10.1038\/s41591-024-03097-1","article-title":"Evaluation and mitigation of the limitations of large language models in clinical decision-making","volume":"30","author":"Hager","year":"2024","journal-title":"Nat Med"},{"key":"B27","doi-asserted-by":"publisher","first-page":"1681979","DOI":"10.3389\/frai.2025.1681979","article-title":"ChatGPT and reference intervals: a comparative analysis of repeatability in GPT-3.5 turbo, GPT-4, and GPT-4o","volume":"8","author":"Meyer","year":"2025","journal-title":"Front Artif Intell"},{"key":"B28","doi-asserted-by":"publisher","first-page":"e65047","DOI":"10.2196\/65047","article-title":"Evaluating and enhancing Japanese large language models for genetic counseling support: comparative study of domain adaptation and the development of an expert-evaluated dataset","volume":"13","author":"Fukushima","year":"2025","journal-title":"JMIR Med Inform"},{"key":"B29","doi-asserted-by":"publisher","first-page":"307","DOI":"10.1186\/s12967-025-06345-1","article-title":"Targeted long-read sequencing facilitates effective carrier screening for complex monogenic diseases including spinal muscular atrophy, \u03b1-\/\u03b2-thalassemia, 21-hydroxylase deficiency, and fragile-X syndrome","volume":"23","author":"Li","year":"2025","journal-title":"J Transl Med"},{"key":"B30","doi-asserted-by":"publisher","first-page":"330","DOI":"10.1038\/s43856-025-01021-3","article-title":"Multi-model assurance analysis showing large language models are highly vulnerable to adversarial hallucination attacks during clinical decision support","volume":"5","author":"Omar","year":"2025","journal-title":"Commun Med (Lond)"},{"key":"B31","doi-asserted-by":"publisher","first-page":"e65848","DOI":"10.2196\/65848","article-title":"Chatbot for the return of positive genetic screening results for hereditary cancer syndromes: prompt engineering project","volume":"11","author":"Coen","year":"2025","journal-title":"JMIR Cancer"},{"key":"B32","doi-asserted-by":"publisher","first-page":"368504251412703","DOI":"10.1177\/00368504251412703","article-title":"AI-driven patient-centered care: a digital transformation framework for gynecologic cancer genetic counseling","volume":"109","author":"Yang","year":"2026","journal-title":"Sci Prog"},{"key":"B33","doi-asserted-by":"publisher","first-page":"e2432143","DOI":"10.1001\/jamanetworkopen.2024.32143","article-title":"Uptake of cancer genetic services for chatbot vs standard-of-care delivery models: the BRIDGE randomized clinical trial","volume":"7","author":"Kaphingst","year":"2024","journal-title":"JAMA Netw Open"},{"key":"B34","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1038\/s41439-026-00344-4","article-title":"ChatTogovar: a TogoVar-based retrieval-augmented generation system for precise genomic variant interpretation","volume":"13","author":"Mitsuhashi","year":"","journal-title":"Hum Genome Var"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1902480\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T05:35:57Z","timestamp":1786426557000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1902480\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,11]]},"references-count":34,"alternative-id":["10.3389\/fdgth.2026.1902480"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1902480","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,11]]},"article-number":"1902480"}}