文献速递眼底病与葡萄膜炎

人工智能与基于规则方法在DRCR视网膜网络光学相干断层扫描血管成像数据集中分割无灌注区的比较

Artificial Intelligence Versus Rules-Based Approach for Segmenting NonPerfusion Area in a DRCR Retina Network Optical Coherence Tomography Angiography Dataset.
PubMed 文献 · Invest Ophthalmol Vis Sci · 2025-01-01

本研究在101例有糖尿病视网膜病变(DR)和274例无DR的成人中,比较基于规则(RB)与深度学习人工智能(AI)算法在光学相干断层扫描血管成像上分割黄斑无灌注区(NPA)的表现。AI方法在交并比、召回率和F1分数上优于RB方法,RB方法在全部解剖分层中精确度更优(均P≤0.001);AI方法在多数分层中与DR严重程度秩相关更强、诊断需转诊DR的曲线下面积更大。

要点

  • AI方法在交并比、召回率和F1分数上优于RB方法,RB方法精确度更优(均P≤0.001)。
  • AI方法在全部解剖分层中与DR严重程度秩相关更强(均P<0.001)。
  • AI方法诊断需转诊DR的曲线下面积在浅层血管复合体、中间毛细血管丛和联合内层视网膜中更大(均P≤0.001),深部毛细血管丛无差异(P=0.92)。

核心结论

  • AI方法输出与人工分级一致性更好,且在多数血管丛中比RB方法更能区分临床相关DR严重程度。

研究设计与样本

  • 研究纳入来自美国20个中心的101例有DR和274例无DR的成人,采集3×3毫米黄斑扫描。
  • 使用基于规则(RB)和基于深度学习的人工智能(AI)算法将无灌注区(NPA)分割为四个解剖分层;对50例扫描子集进行人工分级作为参考。

主要发现

  • AI方法在交并比、召回率和F1分数上优于RB方法,但RB方法在全部解剖分层中相对人工分级的精确度更优(均P≤0.001)。
  • AI方法在全部解剖分层中与早期糖尿病视网膜病变治疗研究(ETDRS)DR严重程度的秩相关强于RB方法(均P<0.001)。
  • AI方法诊断需转诊DR的受试者工作特征曲线下面积在浅层血管复合体、中间毛细血管丛和联合内层视网膜中大于RB方法(均P≤0.001),在深部毛细血管丛中无显著差异(P=0.92)。

关键数值与推荐

  • AI方法在交并比、召回率和F1分数上优于RB方法(具体数值原文未报告)。
  • RB方法在全部解剖分层中精确度优于AI方法(均P≤0.001,具体数值原文未报告)。
  • AI方法在全部解剖分层中与ETDRS DR严重程度秩相关强于RB方法(均P<0.001,具体数值原文未报告)。
  • AI方法诊断需转诊DR的曲线下面积在浅层血管复合体、中间毛细血管丛和联合内层视网膜中大于RB方法(均P≤0.001),深部毛细血管丛无显著差异(P=0.92,具体数值原文未报告)。

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

PURPOSE: Loss of retinal perfusion is associated with both onset and worsening of diabetic retinopathy (DR). Optical coherence tomography angiography is a noninvasive method for measuring the nonperfusion area (NPA) and has promise as a scalable screening tool. This study compares two optical coherence tomography angiography algorithms for quantifying NPA. METHODS: Adults with (N = 101) and without (N = 274) DR were recruited from 20 U.S. sites. We collected 3 × 3-mm macular scans using an Optovue RTVue-XR. Rules-based (RB) and deep-learning-based artificial intelligence (AI) algorithms were used to segment the NPA into four anatomical slabs. For comparison, a subset of scans (n = 50) NPA was graded manually. RESULTS: The AI method outperformed the RB method in intersection over union, recall, and F1 score, but the RB method has better precision relative to manual grading in all anatomical slabs (all P ≤ 0.001). The AI method had a stronger rank correlation with Early Treatment of Diabetic Retinopathy Study DR severity than the RB method in all slabs (all P < 0.001). NPAs graded using the AI method had a greater area under the receiver operating characteristic curve for diagnosing referable DR than the RB method in the superficial vascular complex, intermediate capillary plexus, and combined inner retina (all P ≤ 0.001), but not in the deep capillary plexus (P = 0.92). CONCLUSIONS: Our results indicate that output from the AI-based method agrees better with manual grading and can better distinguish between clinically relevant DR severity levels than a RB approach using most plexuses.

来源:Invest Ophthalmol Vis Sci,查看原文。版权归原作者所有。

引用

  1. Hormel TT, Beaulieu WT, Wang J, et al. Artificial Intelligence Versus Rules-Based Approach for Segmenting NonPerfusion Area in a DRCR Retina Network Optical Coherence Tomography Angiography Dataset. Invest Ophthalmol Vis Sci. 2025. PMID: 40062815. doi:10.1167/iovs.66.3.22

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。