Deep Learning Approaches Predict Glaucomatous Visual Field Damage from OCT Optic Nerve Head En Face Images and Retinal Nerve Fiber Layer Thickness Maps.
Deep Learning Approaches Predict Glaucomatous Visual Field Damage from OCT Optic Nerve Head En Face Images and Retinal Nerve Fiber Layer Thickness Maps.
复制标题
DOI：
10.1016/j.ophtha.2019.09.036
复制
发表时间：
2020-03
期刊：
影响因子：
13.7
通讯作者：
Zangwill LM
中科院分区：
文献类型：
--
作者：
Christopher M;Bowd C;Belghith A;Goldbaum MH;Weinreb RN;Fazio MA;Girkin CA;Liebmann JM;Zangwill LM
开发和评估一种深度学习系统，用于区分有和无青光眼性视野损伤（GVFD）的眼睛，并根据谱域光学相干断层扫描（SDOCT）视神经乳头图像预测GFVD的严重程度。诊断技术的评价9，765个视野（VF）-SDOCT对，收集自1，194名有和没有GVFD的参与者（1909只眼）。训练深度学习模型，以使用SDOCT视网膜神经纤维层（RNFL）厚度图、RNFL表面图像和共焦扫描激光检眼镜（CSLO）图像来识别患有GVFD的眼睛，并根据SDOCT数据预测定量VF平均偏差（MD）、模式标准偏差（PSD）和平均VF扇形模式偏差（PD）。将深度学习模型与平均RNFL厚度进行比较，以使用受试者工作特征曲线下面积（AUC）、灵敏度和特异性识别GVFD。为了预测MD、PSD和平均部门PD，使用R2和平均绝对误差（MAE）评价模型。在独立测试数据集中，基于RNFL表面图像的深度学习模型在识别患有GVFD的眼睛时实现了0.88的AUC，在检测轻度GVFD时实现了0.82的AUC，显著（p < 0.001）优于使用平均RNFL厚度测量（AUC分别为0.82和0.73）。深度学习模型在预测所有定量VF指标方面优于标准RNFL厚度测量。在预测MD时，基于RNFL表面图像的深度学习模型实现了0.70的R2和2.5 dB的MAE，而RNFL厚度测量值为0.45和3.7 dB。在预测平均VF扇区PD时，深度学习模型在下鼻（R2 = 0.60）和上级鼻（R2 = 0.67）扇区中实现了高准确性，在下（R2 = 0.26）和上级（R2 = 0.35）扇区中实现了中等准确性，在中央（R2 = 0.15）和颞（R2 = 0.12）扇区中实现了较低准确性。深度学习模型在识别患有GFVD的眼睛和预测图像功能丧失的严重程度方面具有很高的准确性。从SDOCT成像中准确预测GFVD的严重程度可以帮助临床医生更有效地个性化VF测试的频率。应用于OCT表面视神经乳头图像的深度学习算法实现了高诊断准确性，用于区分有和没有昏迷性视野损伤的眼睛，并预测视野损伤的严重程度（平均偏差）。
To develop and evaluate a deep learning system for differentiating between eyes with and without glaucomatous visual field damage (GVFD) and predicting the severity of GFVD from spectral domain optical coherence tomography (SDOCT) optic nerve head images. Evaluation of a diagnostic technology 9,765 visual field (VF)–SDOCT pairs collected from 1,194 participants with and without GVFD (1909 eyes). Deep learning models were trained to use SDOCT retinal nerve fiber layer (RNFL) thickness maps, RNFL enface images, and confocal scanning laser ophthalmoscopy (CSLO) images to identify eyes with GVFD and predict quantitative VF mean deviation (MD), pattern standard deviation (PSD), and mean VF sectoral pattern deviation (PD) from SDOCT data. Deep learning models were compared to mean RNFL thickness for identifying GVFD using area under receiver operating characteristic curve (AUC), sensitivity, and specificity. For predicting MD, PSD and mean sectoral PD, models were evaluated using R2 and mean absolute error (MAE). In the independent test dataset, the deep learning models based on RNFL enface images achieved an AUC of 0.88 for identifying eyes with GVFD and 0.82 for detecting mild GVFD, significantly (p < 0.001) better than using mean RNFL thickness measurements (AUC = 0.82 and 0.73, respectively). Deep learning models outperformed standard RNFL thickness measurements in predicting all quantitative VF metrics. In predicting MD, deep learning models based on RNFL enface images achieved an R2 of 0.70 and MAE of 2.5 dB compared to 0.45 and 3.7 dB for RNFL thickness measurements. In predicting mean VF sectoral PD, deep learning models achieved high accuracy in the inferior nasal (R2 = 0.60) and superior nasal (R2 = 0.67) sectors, moderate accuracy in inferior (R2 = 0.26) and superior (R2 = 0.35) sectors, and lower accuracy in the central (R2 = 0.15) and temporal (R2 = 0.12) sectors. Deep learning models had high accuracy in identifying eyes with GFVD and predicting the severity of functional loss from images. Accurately predicting the severity of GFVD from SDOCT imaging can help clinicians more effectively individualize the frequency of VF testing to the individual patient. Deep learning algorithms applied to OCT enface optic nerve head images achieved high diagnostic accuracy for differentiating between eyes with and without glaucomatous visual field damage and predicting severity of visual field damage (mean deviation).
登录
查看更多内容
影响因子：
2
作者：
Muhammad H;Fuchs TJ;De Cuir N;De Moraes CG;Blumberg DM;Liebmann JM;Ritch R;Hood DC
通讯作者：
Hood DC
影响因子：
2
作者：
Lopes, Flavio S.;Matsubara, Igor;Prata, Tiago S.
通讯作者：
Prata, Tiago S.
影响因子：
3
作者：
Hood, Donald C.;Chen, Monica F.;Chui, Toco Y. P.
通讯作者：
Chui, Toco Y. P.
影响因子：
4.4
作者：
Guo Z;Kwon YH;Lee K;Wang K;Wahle A;Alward WLM;Fingert JH;Bettis DI;Johnson CA;Garvin MK;Sonka M;Abràmoff MD
通讯作者：
Abràmoff MD
影响因子：
--
作者：
Rao, Harsha L.;Zangwill, Linda M.;Medeiros, Felipe A.
通讯作者：
Medeiros, Felipe A.