基于深度学习模型利用超广角图像预测近视患者屈光不正
Prediction of Refractive Error Based on Ultrawide Field Images With Deep Learning Models in Myopia Patients.
作者信息
Yang Danjuan, Li Meiyan, Li Weizhen, Wang Yunzhe, Niu Lingling, Shen Yang, Zhang Xiaoyu, Fu Bo, Zhou Xingtao
机构信息
Eye Institute and Department of Ophthalmology, Eye & ENT Hospital, Fudan University, Shanghai, China.
NHC Key Laboratory of Myopia, Fudan University, Shanghai, China.
出版信息
Front Med (Lausanne). 2022 Mar 30;9:834281. doi: 10.3389/fmed.2022.834281. eCollection 2022.
SUMMARY
Ultrawide field fundus images could be applied in deep learning models to predict the refractive error of myopic patients. The predicted error was related to the older age and greater spherical power.
PURPOSE
To explore the possibility of predicting the refractive error of myopic patients by applying deep learning models trained with ultrawide field (UWF) images.
METHODS
UWF fundus images were collected from left eyes of 987 myopia patients of Eye and ENT Hospital, Fudan University between November 2015 and January 2019. The fundus images were all captured with Optomap Daytona, a 200° UWF imaging device. Three deep learning models (ResNet-50, Inception-v3, Inception-ResNet-v2) were trained with the UWF images for predicting refractive error. 133 UWF fundus images were also collected after January 2021 as an the external validation data set. The predicted refractive error was compared with the "true value" measured by subjective refraction. Mean absolute error (MAE), mean absolute percentage error (MAPE) and coefficient ( ) value were calculated in the test set. The Spearman rank correlation test was applied for univariate analysis and multivariate linear regression analysis on variables affecting MAE. The weighted heat map was generated by averaging the predicted weight of each pixel.
RESULTS
ResNet-50, Inception-v3 and Inception-ResNet-v2 models were trained with the UWF images for refractive error prediction with of 0.9562, 0.9555, 0.9563 and MAE of 1.72(95%CI: 1.62-1.82), 1.75(95%CI: 1.65-1.86) and 1.76(95%CI: 1.66-1.86), respectively. 29.95%, 31.47% and 29.44% of the test set were within the predictive error of 0.75D in the three models. 64.97%, 64.97%, and 64.47% was within 2.00D predictive error. The predicted MAE was related to older age ( < 0.01) and greater spherical power( < 0.01). The optic papilla and macular region had significant predictive power in the weighted heat map.
CONCLUSIONS
It was feasible to predict refractive error in myopic patients with deep learning models trained by UWF images with the accuracy to be improved.
摘要
超广角眼底图像可应用于深度学习模型，以预测近视患者的屈光不正。预测误差与年龄较大和球镜度数较高有关。
目的
探讨应用经超广角（UWF）图像训练的深度学习模型预测近视患者屈光不正的可能性。
方法
收集2015年11月至2019年1月复旦大学附属眼耳鼻喉科医院987例近视患者左眼的UWF眼底图像。所有眼底图像均使用Optomap Daytona 200°超广角成像设备采集。使用UWF图像训练三种深度学习模型（ResNet-50、Inception-v3、Inception-ResNet-v2）来预测屈光不正。2021年1月以后还收集了133张UWF眼底图像作为外部验证数据集。将预测的屈光不正与主观验光测量的“真实值”进行比较。在测试集中计算平均绝对误差（MAE）、平均绝对百分比误差（MAPE）和相关系数（ ）值。对影响MAE的变量进行Spearman等级相关检验和多变量线性回归分析。通过对每个像素的预测权重求平均值生成加权热图。
结果
使用UWF图像训练ResNet-50、Inception-v3和Inception-ResNet-v2模型用于屈光不正预测，相关系数分别为0.9562、0.9555、0.9563，MAE分别为1.72（95%CI：1.62-1.82）、1.75（95%CI：1.65-1.86）和1.76（95%CI：1.66-1.86）。在三个模型中，测试集分别有29.95%、31.47%和29.44%在预测误差0.75D范围内。64.97%、64.97%和64.47%在预测误差2.00D范围内。预测的MAE与年龄较大（<0.01）和球镜度数较高（<0.01）有关。在加权热图中，视乳头和黄斑区具有显著的预测能力。
结论
用UWF图像训练的深度学习模型预测近视患者的屈光不正具有可行性，但其准确性有待提高。
Zotero 插件