Deep learning on fundus images detects glaucoma beyond the optic disc
This provides the first irrefutable evidence that deep learning can detect glaucoma from regions outside the optic disc, advancing explainable AI in medical imaging for ophthalmology.
The study tackled the problem of limited decision transparency in deep learning models for glaucoma detection by proposing an explainable methodology using cropped fundus images, achieving an AUC of 0.94 for glaucoma detection and an R^2 of 77% for VCDR estimation with original images, and still obtaining significant performance (AUC 0.88, R^2 37%) when the optic nerve head was removed.
Although unprecedented sensitivity and specificity values are reported, recent glaucoma detection deep learning models lack in decision transparency. Here, we propose a methodology that advances explainable deep learning in the field of glaucoma detection and vertical cup-disc ratio (VCDR), an important risk factor. We trained and evaluated deep learning models using fundus images that underwent a certain cropping policy. We defined the crop radius as a percentage of image size, centered on the optic nerve head (ONH), with an equidistant spaced range from 10%-60% (ONH crop policy). The inverse of the cropping mask was also applied (periphery crop policy). Trained models using original images resulted in an area under the curve (AUC) of 0.94 [95% CI: 0.92-0.96] for glaucoma detection, and a coefficient of determination (R^2) equal to 77% [95% CI: 0.77-0.79] for VCDR estimation. Models that were trained on images with absence of the ONH are still able to obtain significant performance (0.88 [95% CI: 0.85-0.90] AUC for glaucoma detection and 37% [95% CI: 0.35-0.40] R^2 score for VCDR estimation in the most extreme setup of 60% ONH crop). Our findings provide the first irrefutable evidence that deep learning can detect glaucoma from fundus image regions outside the ONH.