人机对比:深度学习算法与人类分级在眼底照片上检测青光眼的比较
本研究比较了经光学相干断层扫描视网膜神经纤维层厚度训练的深度学习算法与2名青光眼专家对490张眼底照片的分级,发现该算法预测的视网膜神经纤维层厚度与视野平均偏差的相关性(rho=0.54)显著强于人类分级给出的青光眼性视神经病变概率(rho=0.48),且在临床有意义特异度区域的部分AUC显著更高(0.529 vs 0.411)。
要点
- 深度学习算法预测的视网膜神经纤维层厚度与视野平均偏差的绝对相关性(rho=0.54)显著强于人类分级(rho=0.48)。
- 在特异度85%-100%区域,深度学习算法的部分AUC(0.529)显著高于人类分级(0.411)。
- 深度学习算法在检测可重复性青光眼性视野缺损方面表现不劣于甚至优于人类分级。
核心结论
- 深度学习算法在检测可重复性青光眼性视野缺损方面表现不劣于甚至优于人类分级。
- 该深度学习算法可能替代人类分级用于青光眼人群筛查。
研究设计与样本
- 研究为机器学习算法评估,纳入370名受试者的490只眼的490张眼底照片,由2名青光眼专家分级。
主要发现
- 深度学习算法预测的视网膜神经纤维层厚度与视野平均偏差的绝对相关系数(rho=0.54)显著高于人类分级给出的青光眼性视神经病变概率(rho=0.48),P < .001。
- 在特异度85%-100%区域,深度学习算法的部分AUC为0.529,显著高于人类分级的0.411,P = .016。
关键数值与推荐
- 以可重复性青光眼性视野缺损为标准,比较深度学习算法与人类分级的判别能力,采用受试者工作特征曲线下面积及特异度85%-100%区域的部分AUC。
临床怎么用
- 作者认为该深度学习算法可能替代人类分级用于青光眼人群筛查。
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 compare the diagnostic performance of human gradings vs predictions provided by a machine-to-machine (M2M) deep learning (DL) algorithm trained to quantify retinal nerve fiber layer (RNFL) damage on fundus photographs. DESIGN: Evaluation of a machine learning algorithm. METHODS: An M2M DL algorithm trained with RNFL thickness parameters from spectral-domain optical coherence tomography was applied to a subset of 490 fundus photos of 490 eyes of 370 subjects graded by 2 glaucoma specialists for the probability of glaucomatous optical neuropathy (GON), and estimates of cup-to-disc (C/D) ratios. Spearman correlations with standard automated perimetry (SAP) global indices were compared between the human gradings vs the M2M DL-predicted RNFL thickness values. The area under the receiver operating characteristic curves (AUC) and partial AUC for the region of clinically meaningful specificity (85%-100%) were used to compare the ability of each output to discriminate eyes with repeatable glaucomatous SAP defects vs eyes with normal fields. RESULTS: The M2M DL-predicted RNFL thickness had a significantly stronger absolute correlation with SAP mean deviation (rho=0.54) than the probability of GON given by human graders (rho=0.48; P < .001). The partial AUC for the M2M DL algorithm was significantly higher than that for the probability of GON by human graders (partial AUC = 0.529 vs 0.411, respectively; P = .016). CONCLUSION: An M2M DL algorithm performed as well as, if not better than, human graders at detecting eyes with repeatable glaucomatous visual field loss. This DL algorithm could potentially replace human graders in population screening efforts for glaucoma.
引用
- Jammal AA, Thompson AC, Mariottoni EB, et al. Human Versus Machine: Comparing a Deep Learning Algorithm to Human Gradings for Detecting Glaucoma on Fundus Photographs. Am J Ophthalmol. 2019. PMID: 31730838. doi:10.1016/j.ajo.2019.11.006
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。