文献速递医学人工智能与信息学临床试验与 GCP

眼科基础模型辅助临床:一项随机对照试验

An eyecare foundation model for clinical assistance: a randomized controlled trial.
PubMed 文献 · Nat Med · 2025-08-28

在中国高危人群中,668名参与者随机分配至16名眼科医生,使用EyeFM助手组正确诊断率92.2%对比标准护理组75.4%(P<0.001),转诊率92.2%对比80.5%(P<0.001),报告标准化评分中位数33对比37(P<0.001),干预组自我管理依从性70.1%对比49.1%(P<0.001),转诊建议依从性33.7%对比20.2%(P<0.001)。

要点

  • EyeFM助手使眼科医生正确诊断率从75.4%提升至92.2%(P<0.001)。
  • 转诊率从80.5%提升至92.2%(P<0.001),报告标准化评分中位数33对比37(P<0.001)。
  • 干预组自我管理依从性70.1%对比49.1%,转诊建议依从性33.7%对比20.2%(均P<0.001)。

核心结论

  • 实施EyeFM助手可改善眼科医生的表现和患者的结局。

研究设计与样本

  • 开展多层面评估,包括回顾性验证、多国效力验证(北美、欧洲、亚洲和非洲44名眼科医生参与)以及一项双盲随机对照试验。
  • 随机对照试验为平行、单中心、双盲研究,在中国高危人群中评估EyeFM作为视网膜疾病筛查的临床助手。
  • 共668名参与者(平均年龄57.5岁,79.5%男性)随机分配至16名眼科医生,等分为干预组(使用EyeFM助手)和对照组(标准护理)。

主要发现

  • 主要终点显示,使用EyeFM助手的眼科医生正确诊断率更高(92.2%对比75.4%,P<0.001),转诊率更高(92.2%对比80.5%,P<0.001)。
  • 次要结局显示临床报告标准化评分改善(中位数33对比37,P<0.001)。
  • 参与者对筛查的满意度两组相似,而干预组在随访时自我管理依从性更高(70.1%对比49.1%,P<0.001),转诊建议依从性更高(33.7%对比20.2%,P<0.001)。
  • 部署后评估显示用户接受度高。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

In the context of an increasing need for clinical assessments of foundation models, we developed EyeFM, a multimodal vision-language eyecare copilot, and conducted a multifaceted evaluation, including retrospective validations, multicountry efficacy validation as a clinical copilot and a double-masked randomized controlled trial (RCT). EyeFM was pretrained on 14.5 million ocular images from five imaging modalities paired with clinical texts from global, multiethnic datasets. Efficacy validation invited 44 ophthalmologists across North America, Europe, Asia and Africa in primary and specialty care settings, highlighting its utility as a clinical copilot. The RCT-a parallel, single-center, double-masked study-assessed EyeFM as a clinical copilot in retinal disease screening among a high-risk population in China. A total of 668 participants (mean age 57.5 years, 79.5% male) were randomized to 16 ophthalmologists, equally allocated into intervention (with EyeFM copilot) and control (standard care) groups. The primary endpoint indicated that ophthalmologists with EyeFM copilot achieved higher correct diagnostic rate (92.2% versus 75.4%, P < 0.001) and referral rate (92.2% versus 80.5%, P < 0.001). Secondary outcome indicated improved standardization score of clinical reports (median 33 versus 37, P < 0.001). Participant satisfaction with the screening was similar between groups, whereas the intervention group demonstrated higher compliance with self-management (70.1% versus 49.1%, P < 0.001) and referral suggestions (33.7% versus 20.2%, P < 0.001) at follow-up. Post-deployment evaluations indicated strong user acceptance. Our study provided evidence that implementing EyeFM copilot can improve the performance of ophthalmologists and the outcome of patients. Chinese Clinical Trial Registry registration: ChiCTR2500095518 .

来源:Nat Med,查看原文。版权归原作者所有。

引用

  1. Wu Y, Qian B, Li T, et al. An eyecare foundation model for clinical assistance: a randomized controlled trial. Nat Med. 2025. PMID: 40877476. doi:10.1038/s41591-025-03900-7

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。