文献速递眼底病与葡萄膜炎

基于OCT图像的深度学习模型检测糖尿病黄斑水肿形态学模式

DETECTION OF MORPHOLOGIC PATTERNS OF DIABETIC MACULAR EDEMA USING A DEEP LEARNING APPROACH BASED ON OPTICAL COHERENCE TOMOGRAPHY IMAGES.
PubMed 文献 · Retina · 2021-01-01

该研究开发了一个深度学习模型,基于OCT图像检测糖尿病黄斑水肿的三种形态学模式:弥漫性视网膜增厚、囊样黄斑水肿和浆液性视网膜脱离。模型在内部验证中表现出高AUC(0.971、0.974、0.994)和高准确率,外部验证中AUC分别为0.970、0.997、0.997,准确率分别为90.2%、95.4%、95.9%。遮挡测试显示模型能识别关键病理区域,表明人工智能在辅助DME临床决策方面具有潜力。

要点

  • 深度学习模型可高精度检测DME的三种OCT形态模式。
  • 外部验证中浆液性视网膜脱离检测AUC达0.997。
  • 遮挡测试证实模型关注病理区域,具有透明度。

核心结论

  • 深度学习模型在检测DME的OCT形态模式方面表现出高准确性和透明度。
  • 结果强调人工智能在辅助DME患者临床决策过程中的潜力。

研究设计与样本

  • 训练集包含来自公共数据集和一家眼科中心的12,365张OCT图像。
  • 外部验证集包含来自另一家眼科中心的656张OCT图像。
  • 对三种DME OCT模式(弥漫性视网膜增厚、囊样黄斑水肿、浆液性视网膜脱离)的存在与否分别标记为1或0。
  • 训练深度学习模型检测三种OCT模式,并应用遮挡测试进行可视化。

主要发现

  • 内部验证采用5折交叉验证,三种模式(弥漫性视网膜增厚、囊样黄斑水肿、浆液性视网膜脱离)的AUC分别为0.971、0.974、0.994。
  • 内部验证准确率分别为93.0%、95.1%、98.8%,敏感度分别为93.5%、94.5%、96.7%,特异度分别为92.3%、95.6%、99.3%。
  • 外部验证AUC分别为0.970、0.997、0.997,准确率分别为90.2%、95.4%、95.9%,敏感度分别为80.1%、93.4%、94.9%,特异度分别为97.6%、97.2%、96.5%。
  • 遮挡测试显示深度学习模型能成功识别对检测最关键的病理区域。

关键数值与推荐

  • 内部验证中弥漫性视网膜增厚检测AUC为0.971,准确率93.0%,敏感度93.5%,特异度92.3%。
  • 内部验证中囊样黄斑水肿检测AUC为0.974,准确率95.1%,敏感度94.5%,特异度95.6%。
  • 内部验证中浆液性视网膜脱离检测AUC为0.994,准确率98.8%,敏感度96.7%,特异度99.3%。
  • 外部验证中弥漫性视网膜增厚检测AUC为0.970,准确率90.2%,敏感度80.1%,特异度97.6%。
  • 外部验证中囊样黄斑水肿检测AUC为0.997,准确率95.4%,敏感度93.4%,特异度97.2%。

临床怎么用

  • 该深度学习模型可作为辅助工具,帮助临床医生基于OCT图像识别DME的形态学模式。
  • 高敏感度和特异度提示模型在筛查和诊断中具有潜在应用价值。
  • 遮挡测试提供的可视化有助于理解模型决策,增强临床信任。

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: To develop a deep learning (DL) model to detect morphologic patterns of diabetic macular edema (DME) based on optical coherence tomography (OCT) images. METHODS: In the training set, 12,365 OCT images were extracted from a public data set and an ophthalmic center. A total of 656 OCT images were extracted from another ophthalmic center for external validation. The presence or absence of three OCT patterns of DME, including diffused retinal thickening, cystoid macular edema, and serous retinal detachment, was labeled with 1 or 0, respectively. A DL model was trained to detect three OCT patterns of DME. The occlusion test was applied for the visualization of the DL model. RESULTS: Applying 5-fold cross-validation method in internal validation, the area under the receiver operating characteristic curve for the detection of three OCT patterns (i.e., diffused retinal thickening, cystoid macular edema, and serous retinal detachment) was 0.971, 0.974, and 0.994, respectively, with an accuracy of 93.0%, 95.1%, and 98.8%, respectively, a sensitivity of 93.5%, 94.5%, and 96.7%, respectively, and a specificity of 92.3%, 95.6%, and 99.3%, respectively. In external validation, the area under the receiver operating characteristic curve was 0.970, 0.997, and 0.997, respectively, with an accuracy of 90.2%, 95.4%, and 95.9%, respectively, a sensitivity of 80.1%, 93.4%, and 94.9%, respectively, and a specificity of 97.6%, 97.2%, and 96.5%, respectively. The occlusion test showed that the DL model could successfully identify the pathologic regions most critical for detection. CONCLUSION: Our DL model demonstrated high accuracy and transparency in the detection of OCT patterns of DME. These results emphasized the potential of artificial intelligence in assisting clinical decision-making processes in patients with DME.

来源:Retina,查看原文。版权归原作者所有。

引用

  1. Wu Q, Zhang B, Hu Y, et al. DETECTION OF MORPHOLOGIC PATTERNS OF DIABETIC MACULAR EDEMA USING A DEEP LEARNING APPROACH BASED ON OPTICAL COHERENCE TOMOGRAPHY IMAGES. Retina. 2021. PMID: 33031250. doi:10.1097/IAE.0000000000002992

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。