A deep learning approach for detection of shallow anterior chamber depth based on the hidden features of fundus photographs. (June 2022)

• Record Type: Journal Article
• Authors: Yoo, Tae Keun; Ryu, Ik Hee; Kim, Jin Kuk; Lee, In Sik; Kim, Hong Kyu
• Is Part Of: Computer methods and programs in biomedicine. Volume 219(2022)
• Publication Date: 2022-06
• Subjects: Shallow anterior chamber depth -- Fundus photographs -- Deep learning -- CycleGAN -- Characteristic feature map; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2022.106735
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: Shallow anterior chamber depth (ACD) is a significant risk factor of angle-closure glaucoma. We propose an application of deep learning with fundus photos for detecting a shallow ACD. The cyclegan-based feature map generation provides new insights into hidden patterns. Our framework can link the posterior segment information (from the retina) to the anterior segment of the eye. The hidden features of a shallow ACD were the brightened surroundings of the macula and optic disk, which were difficult to be detected with the naked eye. Abstract: Background and Objectives: Patients with angle-closure glaucoma (ACG) are asymptomatic until they experience a painful attack. Shallow anterior chamber depth (ACD) is considered a significant risk factor for ACG. We propose a deep learning approach to detect shallow ACD using fundus photographs and to identify the hidden features of shallow ACD. Methods: This retrospective study assigned healthy subjects to the training ( n = 1188 eyes) and test ( n = 594) datasets (prospective validation design). We used a deep learning approach to estimate ACD and build a classification model to identify eyes with a shallow ACD. The proposed method, including subtraction of the input and output images of CycleGAN and a thresholding algorithm, was adopted to visualize the characteristic features of fundus photographs with a shallow ACD. Results: The deep learning model integrating fundus photographs and clinical variables achieved areas under the receiver operating characteristic curve of 0.978 (95% confidence interval [CI], 0.963–0.988) for an ACD ≤ 2.60 mm and 0.895 (95% CI, 0.868–0.919) for an ACD ≤ 2.80 mm, and outperformed the regression model using only clinical variables. However, the difference between shallow and deep ACD classes on fundus photographs was difficult to be detected with the naked eye. We were unable to identify the features of shallow ACD using the Grad-CAM. The CycleGAN-based feature images showed that area around the macula and optic disk significantly contributed to the classification of fundus photographs with a shallow ACD. Conclusions: We demonstrated the feasibility of a novel deep learning model to detect a shallow ACD as a screening tool for ACG using fundus photographs. The CycleGAN-based feature map showed the hidden characteristic features of shallow ACD that were previously undetectable by conventional techniques and ophthalmologists. This framework will facilitate the early detection of shallow ACD to prevent overlooking the risks associated with ACG. Graphical abstract: Image, graphical abstract