人工智能驱动的chatgpt - 40和DeepSeek-R1在HIV阴性塔氏菌病中的诊断和预后比较

IF 2.9 3区生物学 Q2 MYCOLOGY

Mycopathologia Pub Date : 2025-10-09 DOI:10.1007/s11046-025-01010-z

Haiyang He, Liuyang Cai, Yi Liu, Yusong Lin, Xingrui Zhu, Dongzhen Liu, Wanqing Liao, Xiaochun Xue, Weihua Pan

{"title":"人工智能驱动的chatgpt - 40和DeepSeek-R1在HIV阴性塔氏菌病中的诊断和预后比较","authors":"Haiyang He, Liuyang Cai, Yi Liu, Yusong Lin, Xingrui Zhu, Dongzhen Liu, Wanqing Liao, Xiaochun Xue, Weihua Pan","doi":"10.1007/s11046-025-01010-z","DOIUrl":null,"url":null,"abstract":"This study evaluates and compares the diagnostic and prognostic capabilities of ChatGPT-4o and DeepSeek-R1 in 56 HIV-negative talaromycosis cases. Clinical case fragments were de-identified and submitted to both models, with diagnostic accuracy and prognostic prediction rates statistically analyzed using chi-square tests, Fisher's exact tests, and logistic regression. Results showed DeepSeek-R1 achieved significantly higher diagnostic accuracy (66.1%) than ChatGPT-4o (3.6%) (χ2 = 48.2, p < 0.001), attributable to its regional data training focusing on Southeast Asia and southern China. Conversely, ChatGPT-4o demonstrated superior prognostic prediction accuracy (78.6% vs. 50.0%, p < 0.001), with 90.2% specificity for improved (survival) outcomes, while DeepSeek-R1 showed 86.7% sensitivity for mortality. Key diagnostic predictors included hilar lymphadenectasis (odds ratio [OR] = 6.8, 95% confidence interval [CI]: 2.1-22.3, P = 0.002) and chest pain (OR = 5.9, 95% CI: 1.4-25.6, P = 0.016). The findings highlight DeepSeek-R1's regional diagnostic advantage and ChatGPT-4o's prognostic utility, advocating for their collaborative use to enhance early detection and management of this neglected fungal infection in immunocompromised, non-HIV populations.","PeriodicalId":19017,"journal":{"name":"Mycopathologia","volume":"190 6","pages":"101"},"PeriodicalIF":2.9000,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Artificial Intelligence Driven Diagnosis and Prognosis Comparison of ChatGPT-4o and DeepSeek-R1 in HIV Negative Talaromycosis.\",\"authors\":\"Haiyang He, Liuyang Cai, Yi Liu, Yusong Lin, Xingrui Zhu, Dongzhen Liu, Wanqing Liao, Xiaochun Xue, Weihua Pan\",\"doi\":\"10.1007/s11046-025-01010-z\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This study evaluates and compares the diagnostic and prognostic capabilities of ChatGPT-4o and DeepSeek-R1 in 56 HIV-negative talaromycosis cases. Clinical case fragments were de-identified and submitted to both models, with diagnostic accuracy and prognostic prediction rates statistically analyzed using chi-square tests, Fisher's exact tests, and logistic regression. Results showed DeepSeek-R1 achieved significantly higher diagnostic accuracy (66.1%) than ChatGPT-4o (3.6%) (χ2 = 48.2, p < 0.001), attributable to its regional data training focusing on Southeast Asia and southern China. Conversely, ChatGPT-4o demonstrated superior prognostic prediction accuracy (78.6% vs. 50.0%, p < 0.001), with 90.2% specificity for improved (survival) outcomes, while DeepSeek-R1 showed 86.7% sensitivity for mortality. Key diagnostic predictors included hilar lymphadenectasis (odds ratio [OR] = 6.8, 95% confidence interval [CI]: 2.1-22.3, P = 0.002) and chest pain (OR = 5.9, 95% CI: 1.4-25.6, P = 0.016). The findings highlight DeepSeek-R1's regional diagnostic advantage and ChatGPT-4o's prognostic utility, advocating for their collaborative use to enhance early detection and management of this neglected fungal infection in immunocompromised, non-HIV populations.\",\"PeriodicalId\":19017,\"journal\":{\"name\":\"Mycopathologia\",\"volume\":\"190 6\",\"pages\":\"101\"},\"PeriodicalIF\":2.9000,\"publicationDate\":\"2025-10-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Mycopathologia\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.1007/s11046-025-01010-z\",\"RegionNum\":3,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MYCOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Mycopathologia","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1007/s11046-025-01010-z","RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MYCOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

本研究评估并比较了chatgpt - 40和DeepSeek-R1在56例hiv阴性talaromylosis病例中的诊断和预后能力。临床病例片段被去识别并提交给两个模型，使用卡方检验、Fisher精确检验和逻辑回归对诊断准确性和预后预测率进行统计分析。结果显示，DeepSeek-R1的诊断准确率（66.1%）显著高于chatgpt - 40 (3.6%) (χ2 = 48.2, p

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Artificial Intelligence Driven Diagnosis and Prognosis Comparison of ChatGPT-4o and DeepSeek-R1 in HIV Negative Talaromycosis.

This study evaluates and compares the diagnostic and prognostic capabilities of ChatGPT-4o and DeepSeek-R1 in 56 HIV-negative talaromycosis cases. Clinical case fragments were de-identified and submitted to both models, with diagnostic accuracy and prognostic prediction rates statistically analyzed using chi-square tests, Fisher's exact tests, and logistic regression. Results showed DeepSeek-R1 achieved significantly higher diagnostic accuracy (66.1%) than ChatGPT-4o (3.6%) (χ² = 48.2, p < 0.001), attributable to its regional data training focusing on Southeast Asia and southern China. Conversely, ChatGPT-4o demonstrated superior prognostic prediction accuracy (78.6% vs. 50.0%, p < 0.001), with 90.2% specificity for improved (survival) outcomes, while DeepSeek-R1 showed 86.7% sensitivity for mortality. Key diagnostic predictors included hilar lymphadenectasis (odds ratio [OR] = 6.8, 95% confidence interval [CI]: 2.1-22.3, P = 0.002) and chest pain (OR = 5.9, 95% CI: 1.4-25.6, P = 0.016). The findings highlight DeepSeek-R1's regional diagnostic advantage and ChatGPT-4o's prognostic utility, advocating for their collaborative use to enhance early detection and management of this neglected fungal infection in immunocompromised, non-HIV populations.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Mycopathologia 生物-真菌学

CiteScore

6.80

自引率

3.60%

发文量

审稿时长

3 months

期刊介绍： Mycopathologia is an official journal of the International Union of Microbiological Societies (IUMS). Mycopathologia was founded in 1938 with the mission to ‘diffuse the understanding of fungal diseases in man and animals among mycologists’. Many of the milestones discoveries in the field of medical mycology have been communicated through the pages of this journal. Mycopathologia covers a diverse, interdisciplinary range of topics that is unique in breadth and depth. The journal publishes peer-reviewed, original articles highlighting important developments concerning medically important fungi and fungal diseases. The journal highlights important developments in fungal systematics and taxonomy, laboratory diagnosis of fungal infections, antifungal drugs, clinical presentation and treatment, and epidemiology of fungal diseases globally. Timely opinion articles, mini-reviews, and other communications are usually invited at the discretion of the editorial board. Unique case reports highlighting unprecedented progress in the diagnosis and treatment of fungal infections, are published in every issue of the journal. MycopathologiaIMAGE is another regular feature for a brief clinical report of potential interest to a mixed audience of physicians and laboratory scientists. MycopathologiaGENOME is designed for the rapid publication of new genomes of human and animal pathogenic fungi using a checklist-based, standardized format.