大型语言模型在为患者提供基于证据的乳疾病教育方面的有效性:比较分析
Luisa Bertin1,2, Federica Branchi3, Carolina Ciacci4
1Department of Surgery, Oncology, Gastroenterology, University of Padua, 35128 Padua, Italy.
Nutrients
|December 31, 2025
概括
与其他大型语言模型 (LLM) 相比,Gemini 2.0为患有乳病的患者教育提供了更好的准确性和清晰度. 然而,由于持续的错误信息风险,所有人工智能模型都需要医疗保健提供者的监督.
科学领域:
- 人工智能在医学中的应用
- 医疗信息学 医疗信息学
- 患者教育 患者教育
背景情况:
- 大型语言模型 (LLM) 显示出患者教育的潜力,但它们在慢性疾病管理中的安全性和有效性尚未得到充分证实.
- 结核病需要终身管理,使得可靠的患者教育至关重要.
研究的目的:
- 进行首次对领先的LLMs (ChatGPT-4,Claude 3.7,Gemini 2.0) 进行比较评价,以教育患有腹腔疾病的患者.
- 评估AI模型的科学准确性,清晰性,错误信息和可读性.
主要方法:
- 使用盲目临床专家 (胃肠科医生,营养学家) 对三个LLM进行横截面评估.
- 四个领域的20个问题:一般理解,症状/诊断,饮食/营养和生活方式管理.
- 评估科学准确性,清晰度 (利克特尺度),错误信息率和可读性 (弗莱什-金凯德,SMOG).
主要成果:
- 与ChatGPT-4和Claude 3.7.7相比,双子座2.0显示出更高的科学准确性 (p=0.015) 和清晰度 (p=0.011).
- 双子座2.0表现出明显更好的可读性,需要2-3年少的教育才能理解 (p<0.001).
- 在所有模型中,错误信息率从13.3%到24.2%不等,没有统计学上显著的差异.
结论:
- 双子座2.0显示在精度,清晰度和可读性方面具有统计学上的显著优势,用于乳疾病教育.
- 关于错误信息率 (13.3-24.2%) 需要医疗保健提供者的监督直接患者应用人工智能工具.
- 人工智能可以支持患者教育,但需要进一步提高准确性以减轻风险.
相关概念视频
Inflammatory Bowel Disease III: Diagnostic Studies and Management I-Nutritional Therapy
609
Various diagnostic tests are employed in the diagnostic process for Inflammatory Bowel Disease (IBD), particularly to differentiate between Crohn's disease and ulcerative colitis.
Diagnostic studies
A colonoscopy is the definitive screening test, distinguishing ulcerative colitis from other colon diseases with similar symptoms. During a colonoscopy test, inflamed mucosa with exudate ulcerations can be observed, and biopsies are taken to determine the histologic characteristics of the...
Diagnostic studies
A colonoscopy is the definitive screening test, distinguishing ulcerative colitis from other colon diseases with similar symptoms. During a colonoscopy test, inflamed mucosa with exudate ulcerations can be observed, and biopsies are taken to determine the histologic characteristics of the...
609
Cancer Survival Analysis
626
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
626


