基于彩色眼底照片检测青光眼性视神经病变的深度学习系统效能
本研究回顾性纳入48116张眼底照片开发并验证深度学习算法,以垂直杯盘比≥0.7及其他典型改变定义可转诊青光眼性视神经病变(GON),由21名眼科医师分级建立参考标准。在8000张可分级照片的独立验证集中,系统检测可转诊GON的AUC为0.986,敏感度95.6%,特异度92.0%。
要点
- 验证集8000张可分级眼底照片中,深度学习系统AUC 0.986,敏感度95.6%,特异度92.0%。
- 假阴性最常见原因为GON合并其他眼病(44/87,50.6%),其中病理性或高度近视37例(42.6%)。
- 假阳性最常见原因为其他眼病(458/480,95.4%),主要为生理性视杯267例(55.6%)。
核心结论
- 深度学习系统能以高敏感度和高特异度检测可转诊青光眼性视神经病变。
- 高度近视或病理性近视共存是导致假阴性结果的最常见原因。
- 生理性视杯和病理性近视是假阳性结果最常见的原因。
研究设计与样本
- 研究回顾性纳入48116张眼底照片用于深度学习算法的开发与验证。
- 研究招募21名受训眼科医师对照片进行分类,参考标准由3名分级者达成一致后确定。
- 可转诊GON定义为垂直杯盘比≥0.7并伴有其他典型GON改变。
- 使用8000张完全可分级的眼底照片作为独立验证数据集评估算法性能。
- 主要结局指标为受试者工作特征曲线下面积(AUC)及敏感度、特异度。
主要发现
- 验证数据集中深度学习系统AUC为0.986,敏感度95.6%,特异度92.0%。
- 假阴性分级(n=87)最常见原因为GON合并其他眼病(n=44,50.6%),包括病理性或高度近视(n=37,42.6%)、糖尿病视网膜病变(n=4,4.6%)和年龄相关性黄斑变性(n=3,3.4%)。
- 假阳性结果(n=480)首要原因为其他眼病(n=458,95.4%),主要包括生理性视杯(n=267,55.6%)。
- 外观正常眼底被误判为假阳性的仅22眼(4.6%)。
关键数值与推荐
- 可转诊GON的参考标准定义为垂直杯盘比≥0.7并伴有其他典型GON改变。
- 验证数据集样本量为8000张完全可分级的眼底照片。
- 深度学习系统在验证集的AUC为0.986,敏感度95.6%,特异度92.0%。
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: To assess the performance of a deep learning algorithm for detecting referable glaucomatous optic neuropathy (GON) based on color fundus photographs. DESIGN: A deep learning system for the classification of GON was developed for automated classification of GON on color fundus photographs. PARTICIPANTS: We retrospectively included 48 116 fundus photographs for the development and validation of a deep learning algorithm. METHODS: This study recruited 21 trained ophthalmologists to classify the photographs. Referable GON was defined as vertical cup-to-disc ratio of 0.7 or more and other typical changes of GON. The reference standard was made until 3 graders achieved agreement. A separate validation dataset of 8000 fully gradable fundus photographs was used to assess the performance of this algorithm. MAIN OUTCOME MEASURES: The area under receiver operator characteristic curve (AUC) with sensitivity and specificity was applied to evaluate the efficacy of the deep learning algorithm detecting referable GON. RESULTS: In the validation dataset, this deep learning system achieved an AUC of 0.986 with sensitivity of 95.6% and specificity of 92.0%. The most common reasons for false-negative grading (n = 87) were GON with coexisting eye conditions (n = 44 [50.6%]), including pathologic or high myopia (n = 37 [42.6%]), diabetic retinopathy (n = 4 [4.6%]), and age-related macular degeneration (n = 3 [3.4%]). The leading reason for false-positive results (n = 480) was having other eye conditions (n = 458 [95.4%]), mainly including physiologic cupping (n = 267 [55.6%]). Misclassification as false-positive results amidst a normal-appearing fundus occurred in only 22 eyes (4.6%). CONCLUSIONS: A deep learning system can detect referable GON with high sensitivity and specificity. Coexistence of high or pathologic myopia is the most common cause resulting in false-negative results. Physiologic cupping and pathologic myopia were the most common reasons for false-positive results.
引用
- Li Z, He Y, Keel S, et al. Efficacy of a Deep Learning System for Detecting Glaucomatous Optic Neuropathy Based on Color Fundus Photographs. Ophthalmology. 2018. PMID: 29506863. doi:10.1016/j.ophtha.2018.01.023
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。