Deep Learning-Based Glaucoma Detection With Cropped Optic Cup and Disc and Blood Vessel Segmentation

Abstract

Glaucoma is an irreversible neurodegenerative disease, where intraocular hypertension is developed due to the increased aqueous humor and blockage of the drainage system between the iris and cornea. As a result, the optic nerve head, which sends visual stimulus from our eyes to the brain, is damaged, causing visual field loss and ultimately blindness. Glaucoma is considered as the sneak thief of vision because it is difficult to diagnose early, and its regular screening is highly recommended to distinguish the neurological disorder. The detection of glaucoma is costly and time-consuming and not only there always remains a good possibility of human error but also this detection method is dependent upon the availability of the resources (experienced ophthalmologists and expensive instruments). In this work, an automatic glaucoma classification technique has been developed by utilizing multiple deep learning approaches. First, a new private dataset of 634 color fundus images has been collected and annotated by two eye specialists, a pediatric ophthalmologist and a glaucoma and refractive surgeon, from Bangladesh Eye Hospital, Bangladesh. Next, various deep learning models (EfficientNet, MobileNet, DenseNet, and GoogLeNet) have been used to detect glaucoma from fundus images. The model with EfficientNet-b3 architecture achieved the best results with test accuracy, F1-score, and ROC AUC of 0.9652, 0.9512, and 0.9574, respectively, for the cropped optic cup and disc fundus photographs. We also constructed a new dataset by segmenting the blood vessels from retinal fundus images employing a U-net model trained on High-Resolution Fundus Image Database. The MobileNet v3 model trained on this dataset achieved a satisfactory test accuracy of 0.8348 and an F1-score of 0.7957. This impressive result suggests that blood vessel segmentation of fundus images can be utilized as an alternative to detect glaucoma automatically.

Detecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions: Transformer for Improved Generalization

“…In general, diagnostic accuracy using artificial intelligence for the detection of glaucoma from fundus photographs and optical coherence tomography images is worse in external datasets than in test sets from the original data source 4 . Others have reported better diagnostic accuracy on some of the external fundus photograph test sets used in the current study 6,7,[31][32][33][34] . However, these reports trained and tested the datasets, so the diagnostic accuracy is expected to be higher than when the fundus photographs are independent external test sets as in the current study.…”

Detecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions

“… 5 Others have reported better diagnostic accuracy on some of the external fundus photograph test sets used in the current study. 7 , 8 , 29 , 30 , 31 , 32 However, these reports trained and tested the datasets, so the diagnostic accuracy is expected to be higher than when the fundus photographs are independent external test sets as in the current study. Different glaucoma definitions between the external datasets may also explain the differences in performance (e.g., VF vs. expert photograph review, cup-to-disc ratio, etc.).…”

Development of preprocessing methods and revised EfficientNet for diabetic retinopathy detection

“…Even when we resized the fundus image, the image was deformed (Figure 5B). There are two types of comment methods to create a crop around the image center of a fundus image: square crop 29 method and circular crop methods 30 . This study attempted to determine which method was suitable for processing fundus images, as discussed in the Results section.…”

“…The square crop method 29 adopted the short edge length of the fundus image as the length to crop the square image through the size property after removal of the black border of the image (Figure 5A). Although the square crop method could scale the original fundus image to the same proportion (Figure 6B), it still could not restore the fundus image to a circular state.…”