Deep learning for automated glaucomatous optic neuropathy detection from ultra-widefield fundus images

Abstract

Background/AimsTo develop a deep learning system for automated glaucomatous optic neuropathy (GON) detection using ultra-widefield fundus (UWF) images.MethodsWe trained, validated and externally evaluated a deep learning system for GON detection based on 22 972 UWF images from 10 590 subjects that were collected at 4 different institutions in China and Japan. The InceptionResNetV2 neural network architecture was used to develop the system. The area under the receiver operating characteristic curve (AUC), sensitivity and specificity were used to assess the performance of detecting GON by the system. The data set from the Zhongshan Ophthalmic Center (ZOC) was selected to compare the performance of the system to that of ophthalmologists who mainly conducted UWF image analysis in clinics.ResultsThe system for GON detection achieved AUCs of 0.983–0.999 with sensitivities of 97.5–98.2% and specificities of 94.3–98.4% in four independent data sets. The most common reasons for false-negative results were confounding optic disc characteristics caused by high myopia or pathological myopia (n=39 (53%)). The leading cause for false-positive results was having other fundus lesions (n=401 (96%)). The performance of the system in the ZOC data set was comparable to that of an experienced ophthalmologist (p>0.05).ConclusionOur deep learning system can accurately detect GON from UWF images in an automated fashion. It may be used as a screening tool to improve the accessibility of screening and promote the early diagnosis and management of glaucoma.

Telehealth and Screening Strategies in the Diagnosis and Management of Glaucoma

“…Eyes that truly had glaucoma were identified by glaucoma specialists who saw reproducible visual field scotomas that matched the appearance of the optic discs. Several other studies demonstrated that certain deep learning parameters can achieve high accuracy with AUC > 0.9 and sensitivity and specificity levels > 90%; false-positive and false-negative results were commonly due to pathologic myopia [65][66][67][68][69][70]. Even with the use of different fundus cameras, deep learning artificial intelligence was able to achieve an AUC > 0.9, provided that image augmentation was performed [71].…”

Artificial intelligence to detect malignant eyelid tumors from photographic images

“…Recently, artificial intelligence (AI) has been reported to attain a high level of accuracy in the automated detection of numerous diseases from clinical images 10 – 15 . In ophthalmology, a large number of studies developed deep learning-based systems that could accurately detect ocular diseases such as diabetic retinopathy, retinal detachment, and glaucoma 16 – 21 . However, eyelid tumors, particularly malignant ones, which need early detection and prompt referral, are not well investigated.…”

Comparison between Deep-Learning-Based Ultra-Wide-Field Fundus Imaging and True-Colour Confocal Scanning for Diagnosing Glaucoma

“…Some studies have shown that UWF fundus imaging is useful for diagnosing glaucoma [ 30 ], whereas others have shown that accurate glaucoma diagnosis can be achieved by combining UWF fundus imaging with DL methods [ 31 , 32 , 33 , 34 ]. However, no studies have shown the results of true-colour confocal scanning in the field of glaucoma, and this modality has only been used for the diagnosis of diabetic retinopathy and retinal diseases [ 35 , 36 , 37 ].…”

“…In a previous study conducted by our team, true-colour confocal scanning was found to be superior to UWF fundus imaging in detecting localised RNFL defects (early changes in glaucoma) when eye physicians manually evaluated images taken using both modalities [ 12 , 30 , 31 ]. It is widely known that a high false-positive rate is obtained when UWF fundus imaging is used to manually determine localised RNFL defects.…”