A Deep Learning Approach for Classification of the Primary Angle-closure Disease Spectrum Based on Anterior Segment Optical Coherence Tomography.

PRCIS: We developed a deep learning-based classifier that can discriminate primary angle-closure suspects, primary angle-closure/primary angle-closure glaucoma, and also control eyes with open-angle with acceptable accuracy.

PURPOSE: To develop a deep learning (DL) based classifier for differentiating subtypes of primary angle closure disease (PACD), including primary angle-closure suspect (PACS) and primary angle-closure/primary angle-closure glaucoma (PAC/PACG) and also normal control eyes.

MATERIALS AND METHODS: Anterior segment optical coherence tomography (AS-OCT) images were used for analysis with five 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 above-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.

RESULTS: 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 male. 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 dataset.

CONCLUSION: The MobileNet-based classifier can detect normal, PACS, and PAC/PACG eyes with acceptable accuracy based on AS-OCT images.