Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, Japan.
Department of Ophthalmology, The University of Tokyo, Tokyo, Japan; Department of Ophthalmology, Seirei Hamamatsu General Hospital, Shizuoka, Hamamatsu, Japan; Seirei Christopher University, Shizuoka, Hamamatsu, Japan.
Am J Ophthalmol. 2020 Oct;218:304-313. doi: 10.1016/j.ajo.2020.04.037. Epub 2020 May 6.
To predict the visual field (VF) of glaucoma patients within the central 10° from optical coherence tomography (OCT) measurements using deep learning and tensor regression.
Cross-sectional study.
Humphrey 10-2 VFs and OCT measurements were carried out in 505 eyes of 304 glaucoma patients and 86 eyes of 43 normal subjects. VF sensitivity at each test point was predicted from OCT-measured thicknesses of macular ganglion cell layer + inner plexiform layer, retinal nerve fiber layer, and outer segment + retinal pigment epithelium. Two convolutional neural network (CNN) models were generated: (1) CNN-PR, which simply connects the output of the CNN to each VF test point; and (2) CNN-TR, which connects the output of the CNN to each VF test point using tensor regression. Prediction performance was assessed using 5-fold cross-validation through the root mean squared error (RMSE). For comparison, RMSE values were also calculated using multiple linear regression (MLR) and support vector regression (SVR). In addition, the absolute prediction error for predicting mean sensitivity in the whole VF was analyzed.
RMSE with the CNN-TR model averaged 6.32 ± 3.76 (mean ± standard deviation) dB. Significantly (P < .05) larger RMSEs were obtained with other models: CNN-PR (6.76 ± 3.86 dB), SVR (7.18 ± 3.87 dB), and MLR (8.56 ± 3.69 dB). The absolute mean prediction error for the whole VF was 2.72 ± 2.60 dB with the CNN-TR model.
The Humphrey 10-2 VF can be predicted from OCT-measured retinal layer thicknesses using deep learning and tensor regression.
利用深度学习和张量回归，从光学相干断层扫描（OCT）测量中预测青光眼患者中央 10°范围内的视野（VF）。
横断面研究。
对 304 例青光眼患者的 505 只眼和 43 例正常对照者的 86 只眼进行了 Humphrey 10-2 VF 和 OCT 测量。从 OCT 测量的黄斑神经节细胞层+内丛状层、视网膜神经纤维层和外节+视网膜色素上皮的厚度，预测每个测试点的 VF 敏感性。生成了 2 个卷积神经网络（CNN）模型：（1）CNN-PR，简单地将 CNN 的输出连接到每个 VF 测试点；（2）CNN-TR，使用张量回归将 CNN 的输出连接到每个 VF 测试点。通过均方根误差（RMSE）的 5 折交叉验证评估预测性能。为了比较，还使用多元线性回归（MLR）和支持向量回归（SVR）计算了 RMSE 值。此外，还分析了预测整个 VF 平均敏感性的绝对预测误差。
CNN-TR 模型的 RMSE 平均为 6.32±3.76（平均值±标准差）dB。其他模型的 RMSE 显著（P<.05）较大：CNN-PR（6.76±3.86 dB）、SVR（7.18±3.87 dB）和 MLR（8.56±3.69 dB）。CNN-TR 模型的整个 VF 的绝对平均预测误差为 2.72±2.60 dB。
可以利用深度学习和张量回归，从 OCT 测量的视网膜层厚度预测 Humphrey 10-2 VF。
