{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Lin Z"],"funding":["Multicenter Investigator-Initiated Trial Project of Mianyang Central Hospital (Project Number: 2024KY-FZC033),Project of the Mianyang Municipal Health Commission(Grant No.202207),Incubation Project of Mianyang Central Hospital (Grant No.2019FH14)"],"pagination":["213"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12181206"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["25(1)"],"pubmed_abstract":["This study aimed to compare the performance of three large language models (ChatGPT-4o, OpenAI O1, and OpenAI O3 mini) in delivering accurate and guideline compliant recommendations for pneumonia management. By assessing both general and guideline-focused questions, the investigation sought to elucidate each model's strengths, limitations, and capacity to self-correct in response to expert feedback. Fifty pneumonia-related questions (30 general, 20 guideline-based) were posed to the three models. Ten infectious disease specialists independently scored responses for accuracy using a 5-point scale. The two chain-of-thought models (OpenAI O1 and OpenAI O3 mini) were further tested for self-correction when initially rated \"poor,\" with re-evaluations conducted one week later to reduce recall bi"],"journal":["Clinical and experimental medicine"],"pubmed_title":["Performance analysis of large language models Chatgpt-4o, OpenAI O1, and OpenAI O3 mini in clinical treatment of pneumonia: a comparative study."],"pmcid":["PMC12181206"],"funding_grant_id":["2019FH14"],"pubmed_authors":["Liu H","Li Y","Wu M","Yu Q","Xie J","Xiao G","Lin Z","Song X"],"additional_accession":[]},"is_claimable":false,"name":"Performance analysis of large language models Chatgpt-4o, OpenAI O1, and OpenAI O3 mini in clinical treatment of pneumonia: a comparative study.","description":"This study aimed to compare the performance of three large language models (ChatGPT-4o, OpenAI O1, and OpenAI O3 mini) in delivering accurate and guideline compliant recommendations for pneumonia management. By assessing both general and guideline-focused questions, the investigation sought to elucidate each model's strengths, limitations, and capacity to self-correct in response to expert feedback. Fifty pneumonia-related questions (30 general, 20 guideline-based) were posed to the three models. Ten infectious disease specialists independently scored responses for accuracy using a 5-point scale. The two chain-of-thought models (OpenAI O1 and OpenAI O3 mini) were further tested for self-correction when initially rated \"poor,\" with re-evaluations conducted one week later to reduce recall bi","dates":{"release":"2025-01-01T00:00:00Z","publication":"2025 Jun","modification":"2026-06-05T23:02:51.143Z","creation":"2026-05-23T03:10:45.468Z"},"accession":"S-EPMC12181206","cross_references":{"pubmed":["40540092"],"doi":["10.1007/s10238-025-01743-7"]}}