Hirota Masakazu, Mizota Atsushi, Mimura Tatsuya, Hayashi Takao, Kotoku Junichi, Sawa Tomohiro, Inoue Kenji
Department of Orthoptics, Faculty of Medical Technology, Teikyo University, Itabashi, Tokyo, Japan.
Department of Ophthalmology, School of Medicine, Teikyo University, 2-11-1 Kaga, Itabashi-ku, Tokyo, 173-8605, Japan.
Int Ophthalmol. 2020 Nov;40(11):3013-3022. doi: 10.1007/s10792-020-01485-3. Epub 2020 Jun 27.
The purpose of this study was to evaluate the accuracy of the convolutional neural network (CNN) model in glaucoma identification with three primary colors (red, green, blue; RGB) and split color channels using fundus photographs with a small sample size.
The dataset was prepared using color fundus photographs captured with a fundus camera (VX-10i, Kowa Co., Ltd., Tokyo, Japan). The training dataset consisted of 200 images, and the validation dataset contained 60 images. In the preprocessing stage, the color channels for the fundus images were separated into red (red model), green (green model), and blue (blue model) using OpenCV on Windows. All images were resized to squares with a size of 512 × 512 pixels for preprocessing before input into the model, and the model was fine-tuned with VGG16.
The diagnostic performance was significantly higher in the green model [area under the curve (AUC) 0.946; 95% confidence interval (CI) 0.851-0.982] than in the RGB model (AUC 0.800; 95% CI 0.658-0.893; P = 0.006), red model (AUC 0.746; 95% CI 0.601-0.851; P = 0.002), and blue model (AUC 0.558; 95% CI 0.405-0.700; P < 0.001).
The present study showed that the green digital filter is useful for structuring CNN models for automatic discrimination of glaucoma using fundus photographs with a small sample size. The present findings suggest that preprocessing, when creating the CNN model, is an important step for the identification of a large number of retinal diseases using color fundus photographs.
本研究旨在使用小样本量的眼底照片，评估卷积神经网络（CNN）模型在通过三种原色（红、绿、蓝；RGB）和分离颜色通道进行青光眼识别中的准确性。
使用眼底相机（VX - 10i，日本东京兴和株式会社）拍摄的彩色眼底照片准备数据集。训练数据集由200张图像组成，验证数据集包含60张图像。在预处理阶段，使用Windows系统上的OpenCV将眼底图像的颜色通道分离为红色（红色模型）、绿色（绿色模型）和蓝色（蓝色模型）。在输入模型之前，所有图像都被调整为512×512像素的正方形进行预处理，并且使用VGG16对模型进行微调。
绿色模型的诊断性能[曲线下面积（AUC）0.946；95%置信区间（CI）0.851 - 0.982]显著高于RGB模型（AUC = 0.800；95% CI 0.658 - 0.893；P = 0.006）、红色模型（AUC 0.746；95% CI 0.601 - 0.851；P = 0.002）和蓝色模型（AUC 0.558；95% CI 0.405 - 0.700；P < 0.001）。
本研究表明，绿色数字滤波器有助于构建CNN模型，用于使用小样本量的眼底照片自动鉴别青光眼。本研究结果表明，在创建CNN模型时进行预处理是使用彩色眼底照片识别大量视网膜疾病的重要步骤。
