文献速递医学人工智能与信息学基础与转化研究

用于检测视网膜眼底照片中糖尿病视网膜病变的深度学习算法的开发与验证

Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs.
PubMed 文献 · JAMA · 2016-01-01

本研究利用128175张视网膜图像训练深度卷积神经网络,并在EyePACS-1和Messidor-2两个独立数据集上验证其检测需转诊糖尿病视网膜病变(RDR)的性能;算法在EyePACS-1的AUC为0.991,Messidor-2为0.990,高特异性操作点下敏感度分别为90.3%和87.0%,特异度分别为98.1%和98.5%。

要点

  • 深度学习算法在EyePACS-1和Messidor-2数据集检测RDR的AUC分别为0.991和0.990。
  • 高特异性操作点下,EyePACS-1敏感度90.3%、特异度98.1%,Messidor-2敏感度87.0%、特异度98.5%。
  • 高敏感度操作点下,EyePACS-1敏感度97.5%、特异度93.4%,Messidor-2敏感度96.1%、特异度93.9%。

核心结论

  • 在成人糖尿病患者视网膜眼底照片评估中,基于深度机器学习的算法检测需转诊糖尿病视网膜病变具有高敏感度和特异度。

研究设计与样本

  • 使用2015年5月至12月由54名美国执业眼科医师和眼科高年住院医师对128175张视网膜图像进行3至7次分级(糖尿病视网膜病变、糖尿病黄斑水肿和图像可分级性)的回顾性开发数据集训练深度卷积神经网络。
  • 算法于2016年1月和2月在两个独立数据集(EyePACS-1和Messidor-2)中验证,两者均由至少7名美国委员会认证眼科医师分级且评分者内一致性高。

主要发现

  • EyePACS-1数据集包含4997例患者的9963张图像(平均年龄54.4岁;62.2%女性;RDR患病率为683/8878张完全可分级图像[7.8%])。
  • Messidor-2数据集包含874例患者的1748张图像(平均年龄57.6岁;42.6%女性;RDR患病率为254/1745张完全可分级图像[14.6%])。
  • 检测RDR的受试者工作特征曲线下面积:EyePACS-1为0.991(95%CI,0.988-0.993),Messidor-2为0.990(95%CI,0.986-0.995)。
  • 高特异性操作点下,EyePACS-1敏感度90.3%(95%CI,87.5%-92.7%)、特异度98.1%(95%CI,97.8%-98.5%);Messidor-2敏感度87.0%(95%CI,81.1%-91.0%)、特异度98.5%(95%CI,97.7%-99.1%)。
  • 高敏感度操作点下,EyePACS-1敏感度97.5%、特异度93.4%;Messidor-2敏感度96.1%、特异度93.9%。

关键数值与推荐

  • 需转诊糖尿病视网膜病变(RDR)定义为中度及更差糖尿病视网膜病变、需转诊糖尿病黄斑水肿或两者兼有,以眼科专家小组多数决策为参考标准。
  • 算法在从开发集选出的两个操作点进行评估,一个选为高特异性,另一个选为高敏感度。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文指出需进一步研究以确定该算法在临床环境中应用的可行性,并确定与当前眼科评估相比,使用该算法是否能改善诊疗和结局。

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

IMPORTANCE: Deep learning is a family of computational methods that allow an algorithm to program itself by learning from a large set of examples that demonstrate the desired behavior, removing the need to specify rules explicitly. Application of these methods to medical imaging requires further assessment and validation. OBJECTIVE: To apply deep learning to create an algorithm for automated detection of diabetic retinopathy and diabetic macular edema in retinal fundus photographs. DESIGN AND SETTING: A specific type of neural network optimized for image classification called a deep convolutional neural network was trained using a retrospective development data set of 128 175 retinal images, which were graded 3 to 7 times for diabetic retinopathy, diabetic macular edema, and image gradability by a panel of 54 US licensed ophthalmologists and ophthalmology senior residents between May and December 2015. The resultant algorithm was validated in January and February 2016 using 2 separate data sets, both graded by at least 7 US board-certified ophthalmologists with high intragrader consistency. EXPOSURE: Deep learning-trained algorithm. MAIN OUTCOMES AND MEASURES: The sensitivity and specificity of the algorithm for detecting referable diabetic retinopathy (RDR), defined as moderate and worse diabetic retinopathy, referable diabetic macular edema, or both, were generated based on the reference standard of the majority decision of the ophthalmologist panel. The algorithm was evaluated at 2 operating points selected from the development set, one selected for high specificity and another for high sensitivity. RESULTS: The EyePACS-1 data set consisted of 9963 images from 4997 patients (mean age, 54.4 years; 62.2% women; prevalence of RDR, 683/8878 fully gradable images [7.8%]); the Messidor-2 data set had 1748 images from 874 patients (mean age, 57.6 years; 42.6% women; prevalence of RDR, 254/1745 fully gradable images [14.6%]). For detecting RDR, the algorithm had an area under the receiver operating curve of 0.991 (95% CI, 0.988-0.993) for EyePACS-1 and 0.990 (95% CI, 0.986-0.995) for Messidor-2. Using the first operating cut point with high specificity, for EyePACS-1, the sensitivity was 90.3% (95% CI, 87.5%-92.7%) and the specificity was 98.1% (95% CI, 97.8%-98.5%). For Messidor-2, the sensitivity was 87.0% (95% CI, 81.1%-91.0%) and the specificity was 98.5% (95% CI, 97.7%-99.1%). Using a second operating point with high sensitivity in the development set, for EyePACS-1 the sensitivity was 97.5% and specificity was 93.4% and for Messidor-2 the sensitivity was 96.1% and specificity was 93.9%. CONCLUSIONS AND RELEVANCE: In this evaluation of retinal fundus photographs from adults with diabetes, an algorithm based on deep machine learning had high sensitivity and specificity for detecting referable diabetic retinopathy. Further research is necessary to determine the feasibility of applying this algorithm in the clinical setting and to determine whether use of the algorithm could lead to improved care and outcomes compared with current ophthalmologic assessment.

来源:JAMA,查看原文。版权归原作者所有。

引用

  1. Gulshan V, Peng L, Coram M, et al. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA. 2016. PMID: 27898976. doi:10.1001/jama.2016.17216

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。