文献速递眼底病与葡萄膜炎

人工智能在视网膜疾病临床决策中的应用:从筛查、诊断到治疗与纵向管理

Artificial intelligence for clinical decision-making in retinal disorders: From screening and diagnosis to treatment and longitudinal management.
PubMed 文献 · Prog Retin Eye Res · 2026-09-29

本文为叙述性综述,围绕糖尿病视网膜病变(DR)筛查与转诊、糖尿病黄斑水肿(DME)治疗评估、新生血管性年龄相关性黄斑变性(nAMD)再治疗与纵向监测、遗传性视网膜疾病(IRD)诊断与试验富集四个临床场景,梳理AI在视网膜疾病临床决策中的证据。证据最强的是DR筛查,有前瞻性和真实世界验证支持;DME和nAMD的治疗导向应用及IRD的多模态诊断或预后应用验证尚不一致。作者强调模型性能与临床效用之间的差距,并提出强调外部验证、校准、不确定性处理、工作流整合、前瞻性评估和可问责部署的审慎转化路线图。

要点

  • AI在视网膜影像中的应用已从图像分类扩展到多模态解读、纵向预测和临床决策支持。
  • DR筛查证据最强,有前瞻性和真实世界验证支持;DME、nAMD治疗应用及IRD多模态应用验证尚不一致。
  • 需关注模型性能与临床效用的差距,转化需外部验证、校准、不确定性处理和工作流整合。

核心结论

  • AI在视网膜疾病临床决策中证据最强的是DR筛查,治疗导向及IRD多模态应用验证尚不一致。
  • 模型性能与临床效用存在差距,需审慎转化路线图推动可问责部署。

研究设计与样本

  • 叙述性综述,聚焦DR筛查与转诊、DME治疗评估、nAMD再治疗与纵向监测、IRD诊断与试验富集四个临床场景。
  • 围绕疾病特异性决策点、相关数据模态及多模态与纵向整合要求组织证据。

主要发现

  • AI在视网膜影像中已从图像分类扩展到多模态解读、纵向预测和临床决策支持。
  • DR筛查证据最强,有前瞻性和真实世界验证支持。
  • DME和nAMD的治疗导向应用及IRD的多模态诊断或预后应用验证尚不一致。

临床怎么用

  • DR筛查可优先考虑AI辅助,但需外部验证、校准和不确定性处理。
  • 治疗导向的DME和nAMD应用及IRD多模态应用需前瞻性评估和工作流整合后再推广。

局限性

  • 为叙述性综述,未报告系统检索范围、样本量或定量合并结果。
  • 治疗导向及IRD多模态应用验证尚不一致,模型性能与临床效用存在差距。

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

Artificial intelligence (AI) in retinal imaging has expanded from image classification to multimodal interpretation, longitudinal prediction, and clinically oriented decision support. This narrative review focuses on four complementary clinical settings: diabetic retinopathy (DR) screening and referral, diabetic macular edema (DME) treatment assessment, neovascular age-related macular degeneration (nAMD) retreatment and longitudinal monitoring, and inherited retinal disease (IRD) diagnosis, genotype-phenotype support, progression modeling, and trial enrichment. We organize the evidence around disease-specific decision points, relevant data modalities, and the requirements for multimodal and longitudinal integration. Evidence is strongest for DR screening, supported by prospective and real-world validation, whereas treatment-oriented applications in DME and nAMD and multimodal diagnostic or prognostic applications in IRDs remain less consistently validated. We highlight the gap between model performance and clinical utility and propose a cautious translational roadmap emphasizing external validation, calibration, uncertainty handling, workflow integration, prospective evaluation, and accountable deployment.

来源:Prog Retin Eye Res,查看原文。版权归原作者所有。

引用

  1. Jin K, Zhao K, Agrawal R, et al. Artificial intelligence for clinical decision-making in retinal disorders: From screening and diagnosis to treatment and longitudinal management. Prog Retin Eye Res. 2026. PMID: 42810675. doi:10.1016/j.preteyeres.2026.101533

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。