Fully Automated Colorimetric Analysis of the Optic Nerve Aided by Deep Learning and Its Association with Perimetry and OCT for the Study of Glaucoma

Abstract

Background: Laguna-ONhE is an application for the colorimetric analysis of optic nerve images, which topographically assesses the cup and the presence of haemoglobin. Its latest version has been fully automated with five deep learning models. In this paper, perimetry in combination with Laguna-ONhE or Cirrus-OCT was evaluated. Methods: The morphology and perfusion estimated by Laguna ONhE were compiled into a “Globin Distribution Function” (GDF). Visual field irregularity was measured with the usual pattern standard deviation (PSD) and the threshold coefficient of variation (TCV), which analyses its harmony without taking into account age-corrected values. In total, 477 normal eyes, 235 confirmed, and 98 suspected glaucoma cases were examined with Cirrus-OCT and different fundus cameras and perimeters. Results: The best Receiver Operating Characteristic (ROC) analysis results for confirmed and suspected glaucoma were obtained with the combination of GDF and TCV (AUC: 0.995 and 0.935, respectively. Sensitivities: 94.5% and 45.9%, respectively, for 99% specificity). The best combination of OCT and perimetry was obtained with the vertical cup/disc ratio and PSD (AUC: 0.988 and 0.847, respectively. Sensitivities: 84.7% and 18.4%, respectively, for 99% specificity). Conclusion: Using Laguna ONhE, morphology, perfusion, and function can be mutually enhanced with the methods described for the purpose of glaucoma assessment, providing early sensitivity.

Optic disc area frequency distribution in a large sample of retinographic images

“…The neural network that has been trained and used to detect these types of situations is described in the paper referenced in Ref. 23 . In the first group of limiting motifs, the user has the option to retake the image to obtain a new, higher quality image.…”

“…In the second case, anomalies such as papillary oedema, colobomas, myelin fibres and other dysmorphia. The neural network that has been trained and used to detect these types of situations is described in the paper referenced in Ref 23. .…”

“…The optic disc segmentation was performed using a previously described neural network, trained using deep learning U-Net architecture, identifying the inner edge of Elschnig’s scleral ring (figure 1). 22 23 These previous papers explain how the data were prepared, how the network was trained and detail the results of the evaluation of the optical disc segmentation in comparison with those described in the literature up to that time. For example, it is described that ‘The Sorensen-Dice similarity index between manual and automatic segmentations was 0.993".…”

Literature Review on Artificial Intelligence Methods for Glaucoma Screening, Segmentation, and Classification

“…The authors of [95] evaluated the perimetry with a combination of Laguna-ONhE and Cirrus-OCT, analyzing a total of 477 normal eyes, 235 confirmed, and 98 suspected glaucoma cases. The best results were verified with the combination of the "Globin Distribution Function" (GDF) and the threshold coefficient of variation (TCV).…”

Updates in Clinical and Translational Glaucoma Research

“…González-Hernández et al [ 23 ] addressed artificial intelligence tools by means of their innovative article entitled “Fully automated colorimetric analysis of the optic nerve aided by Deep Learning and its association with perimetry for the study of glaucoma”. These authors designed the Laguna-ONhE, an application for the colorimetric analysis of optic nerve images, which is capable of topographically assessing the cup and the presence of hemoglobin.…”