Glaucoma Service, Farabi Eye Hospital, Tehran University of Medical Sciences, Tehran, Iran.
School of ECE, College of Engineering, University of Tehran, Tehran.
J Glaucoma. 2023 Jun 1;32(6):540-547. doi: 10.1097/IJG.0000000000002194. Epub 2023 Mar 3.
We developed a deep learning-based classifier that can discriminate primary angle closure suspects (PACS), primary angle closure (PAC)/primary angle closure glaucoma (PACG), and also control eyes with open angle with acceptable accuracy.
To develop a deep learning-based classifier for differentiating subtypes of primary angle closure disease, including PACS and PAC/PACG, and also normal control eyes.
Anterior segment optical coherence tomography images were used for analysis with 5 different networks including MnasNet, MobileNet, ResNet18, ResNet50, and EfficientNet. The data set was split with randomization performed at the patient level into a training plus validation set (85%), and a test data set (15%). Then 4-fold cross-validation was used to train the model. In each mentioned architecture, the networks were trained with original and cropped images. Also, the analyses were carried out for single images and images grouped on the patient level (case-based). Then majority voting was applied to the determination of the final prediction.
A total of 1616 images of normal eyes (87 eyes), 1055 images of PACS (66 eyes), and 1076 images of PAC/PACG (66 eyes) eyes were included in the analysis. The mean ± SD age was 51.76 ± 15.15 years and 48.3% were males. MobileNet had the best performance in the model, in which both original and cropped images were used. The accuracy of MobileNet for detecting normal, PACS, and PAC/PACG eyes was 0.99 ± 0.00, 0.77 ± 0.02, and 0.77 ± 0.03, respectively. By running MobileNet in a case-based classification approach, the accuracy improved and reached 0.95 ± 0.03, 0.83 ± 0.06, and 0.81 ± 0.05, respectively. For detecting the open angle, PACS, and PAC/PACG, the MobileNet classifier achieved an area under the curve of 1, 0.906, and 0.872, respectively, on the test data set.
The MobileNet-based classifier can detect normal, PACS, and PAC/PACG eyes with acceptable accuracy based on anterior segment optical coherence tomography images.
我们开发了一种基于深度学习的分类器，能够准确地区分原发性闭角型青光眼疑似患者（PACS）、原发性闭角型青光眼（PAC）/原发性闭角型青光眼（PACG）以及正常的对照组眼睛。
开发一种基于深度学习的分类器，用于区分原发性闭角型疾病的亚型，包括 PACS 和 PAC/PACG 以及正常的对照组眼睛。
使用眼前节光学相干断层扫描图像进行分析，包括 5 种不同的网络，包括 MnasNet、MobileNet、ResNet18、ResNet50 和 EfficientNet。数据集通过在患者水平上进行随机化分割为训练加验证集（85%）和测试数据集（15%）。然后使用 4 折交叉验证来训练模型。在每个提到的架构中，网络使用原始和裁剪的图像进行训练。此外，还对单张图像和按患者水平分组的图像（基于病例的）进行了分析。然后应用多数投票来确定最终预测。
共纳入正常眼（87 眼）、PAC 疑似患者（66 眼）和 PACG 疑似患者（66 眼）的 1616 张图像。平均年龄（±标准差）为 51.76±15.15 岁，其中 48.3%为男性。MobileNet 在使用原始和裁剪图像的模型中表现最佳。MobileNet 检测正常、PACS 和 PACG 眼睛的准确率分别为 0.99±0.00、0.77±0.02 和 0.77±0.03。通过以病例为基础的分类方法运行 MobileNet，准确率提高到 0.95±0.03、0.83±0.06 和 0.81±0.05。在检测开角、PACS 和 PACG 时，MobileNet 分类器在测试数据集上的曲线下面积分别为 1、0.906 和 0.872。
基于眼前节光学相干断层扫描图像，基于 MobileNet 的分类器可以以可接受的准确性检测正常、PACS 和 PACG 眼睛。
