Department of Ophthalmology, Seoul National University College of Medicine.
Department of Ophthalmology, Seoul National University Hospital, Seoul, Korea.
J Glaucoma. 2020 Apr;29(4):287-294. doi: 10.1097/IJG.0000000000001458.
PRéCIS:: A spectral-domain optical coherence tomography (SD-OCT) based deep learning system detected glaucomatous structural change with high sensitivity and specificity. It outperformed the clinical diagnostic parameters in discriminating glaucomatous eyes from healthy eyes.
The purpose of this study was to assess the performance of a deep learning classifier for the detection of glaucomatous change based on SD-OCT.
Three hundred fifty image sets of ganglion cell-inner plexiform layer (GCIPL) and retinal nerve fiber layer (RNFL) SD-OCT for 86 glaucomatous eyes and 307 SD-OCT image sets of 196 healthy participants were recruited and split into training (197 eyes) and test (85 eyes) datasets based on a patient-wise split. The bottleneck features extracted from the GCIPL thickness map, GCIPL deviation map, RNFL thickness map, and RNFL deviation map were used as predictors for the deep learning classifier. The area under the receiver operating characteristic curve (AUC) was calculated and compared with those of conventional glaucoma diagnostic parameters including SD-OCT thickness profile and standard automated perimetry (SAP) to evaluate the accuracy of discrimination for each algorithm.
In the test dataset, this deep learning system achieved an AUC of 0.990 [95% confidence interval (CI), 0.975-1.000] with a sensitivity of 94.7% and a specificity of 100.0%, which was significantly larger than the AUCs with all of the optical coherence tomography and SAP parameters: 0.949 (95% CI, 0.921-0.976) with average GCIPL thickness (P=0.006), 0.938 (95% CI, 0.905-0.971) with average RNFL thickness (P=0.003), and 0.889 (0.844-0.934) with mean deviation of SAP (P<0.001; DeLong test).
An SD-OCT-based deep learning system can detect glaucomatous structural change with high sensitivity and specificity.
目的：本研究旨在评估基于光谱域光学相干断层扫描（SD-OCT）的深度学习分类器检测青光眼结构变化的性能。
方法：纳入 86 只青光眼眼和 196 只健康参与者的 355 组节细胞内丛状层（GCIPL）和视网膜神经纤维层（RNFL）SD-OCT 图像集，并根据患者分组将其分为训练集（197 只眼）和测试集（85 只眼）。从 GCIPL 厚度图、GCIPL 偏差图、RNFL 厚度图和 RNFL 偏差图中提取瓶颈特征作为深度学习分类器的预测因子。计算受试者工作特征曲线（ROC）下面积（AUC），并与包括 SD-OCT 厚度图谱和标准自动视野计（SAP）在内的传统青光眼诊断参数进行比较，以评估每种算法的判别准确性。
结果：在测试集中，该深度学习系统的 AUC 为 0.990（95%置信区间[CI]：0.975-1.000），灵敏度为 94.7%，特异性为 100.0%，明显优于所有 OCT 和 SAP 参数的 AUC：平均 GCIPL 厚度的 AUC 为 0.949（95%CI：0.921-0.976）（P=0.006），平均 RNFL 厚度的 AUC 为 0.938（95%CI：0.905-0.971）（P=0.003），SAP 平均偏差的 AUC 为 0.889（0.844-0.934）（P<0.001；DeLong 检验）。
结论：基于 SD-OCT 的深度学习系统可以高灵敏度和特异性检测青光眼结构变化。
