In-Depth Evaluation of Saliency Maps for Interpreting Convolutional Neural Network Decisions in the Diagnosis of Glaucoma Based on Fundus Imaging.

Glaucoma, a leading cause of blindness, damages the optic nerve, making early diagnosis challenging due to no initial symptoms. Fundus eye images taken with a non-mydriatic retinograph help diagnose glaucoma by revealing structural changes, including the optic disc and cup. This research aims to thoroughly analyze saliency maps in interpreting convolutional neural network decisions for diagnosing glaucoma from fundus images. These maps highlight the most influential image regions guiding the network's decisions. Various network architectures were trained and tested on 739 optic nerve head images, with nine saliency methods used. Some other popular datasets were also used for further validation. The results reveal disparities among saliency maps, with some consensus between the folds corresponding to the same architecture. Concerning the significance of optic disc sectors, there is generally a lack of agreement with standard medical criteria. The background, nasal, and temporal sectors emerge as particularly influential for neural network decisions, showing a likelihood of being the most relevant ranging from 14.55% to 28.16% on average across all evaluated datasets. We can conclude that saliency maps are usually difficult to interpret and even the areas indicated as the most relevant can be very unintuitive. Therefore, its usefulness as an explanatory tool may be compromised, at least in problems such as the one addressed in this study, where the features defining the model prediction are generally not consistently reflected in relevant regions of the saliency maps, and they even cannot always be related to those used as medical standards.

Contrast based circular approximation for accurate and robust optic disc segmentation in retinal images.

A new method for automatic optic disc localization and segmentation is presented. The localization procedure combines vascular and brightness information to provide the best estimate of the optic disc center which is the starting point for the segmentation algorithm. A detection rate of 99.58% and 100% was achieved for the Messidor and ONHSD databases, respectively. A simple circular approximation to the optic disc boundary is proposed based on the maximum average contrast between the inner and outer ring of a circle centered on the estimated location. An average overlap coefficient of 0.890 and 0.865 was achieved for the same datasets, outperforming other state of the art methods. The results obtained confirm the advantages of using a simple circular model under non-ideal conditions as opposed to more complex deformable models.

Estimating the Amount of Hemoglobin in the Neuroretinal Rim Using Color Images and OCT.

PURPOSE: To calculate the amount of hemoglobin (Hb) in the optic nerve head (ONH), using superimposed color fundus images with disc, rim and cup boundaries obtained by OCT-Cirrus. MATERIAL AND METHODS: We examined 100 healthy and 121 glaucomatous eyes using Oculus-Spark perimetry, Cirrus-OCT and Visucam (Zeiss) ONH color images. The Laguna ONhE program was then used to calculate the amount of Hb in the cup and six sectors of the rim. Receiver operating characteristic (ROC) analysis was performed and correlations between parameters were calculated. RESULTS: In suspected and confirmed glaucoma, Hb was significantly lower than controls in all rim sectors, especially the inferior and superonasal (p < 0.0001). Mean deviation (MD) of visual field regions showed greater correlation with the amount of Hb in the superior and inferior sectors of the rim than with rim area (p = 0.02) or nerve fiber layer thickness (p < 0.0001). On ROC analysis, the best diagnostic indicators were OCT rim area, vertical cup/disc ratio (C/D) and Glaucoma Discriminant Function (GDF) of Laguna ONhE, without significant differences. CONCLUSIONS: The amount of Hb in the ONH seems to have an important relationship with glaucomatous visual field sensitivity. The remaining rim has insufficient perfusion in many cases of glaucoma.

Estimation of the relative amount of hemoglobin in the cup and neuroretinal rim using stereoscopic color fundus images.

PURPOSE: To calculate the relative amount of hemoglobin (Hb) in sectors of the optic nerve head (ONH) from stereoscopic color fundus images using the Laguna ONhE method and compare the results with the visual field evaluation and optical coherence tomography (OCT). METHODS: Healthy eyes (n = 87) and glaucoma eyes (n = 71) underwent reliable Oculus Spark perimetry and Cirrus OCT. Optical nerve head color images were acquired with a nonmydriatic stereoscopic Wx Kowa fundus camera. Laguna ONhE program was applied to these images to calculate the relative Hb amount in the cup and six sectors of the rim. Receiver operating characteristic (ROC) analysis and correlations between parameters were calculated. RESULTS: We did not observe any variations in the relative amount of Hb in relation to age in healthy subjects (R(2) = 0.033, P > 0.05). Maximum ROC area confidence intervals were observed for a combination between perimetric indices and the Laguna ONhE Glaucoma discriminant function (0.970-0.899) followed by rim area (0.960-0.883), and mean deviation (MD; 0.944-0.857). In glaucoma cases, relative Hb amount presented significant reduction in all rim sectors, especially 231° to 270° and 81° to 120° (P < 0.001), except in the temporal 311° to 40° (P = 0.11). Perimetry mean sensitivity by sectors was better correlated with respective Hb levels than with rim areas or the corresponding nerve fiber thickness, especially the superior and inferior sectors (P < 0.05). CONCLUSIONS: Visual field sensitivity was better correlated with Hb levels than with rim sector areas or the corresponding nerve fiber thickness. In many cases the remaining rim show low perfusion, especially in the superior and inferior sectors.

Diagnostic ability of a new method for measuring haemoglobin levels in the optic nerve head in multiple sclerosis patients.

AIMS: To evaluate a new method for measuring haemoglobin (Hb) levels and quantifying the colour changes in the optic nerve head of multiple sclerosis (MS) patients to detect axonal loss and consequently optic disc atrophy. MATERIAL AND METHODS: 40 MS patients and 40 age and sex-matched healthy subjects were included in this prospective cross-sectional study and underwent a full ophthalmological examination, including three photographs of the optic disc. The Laguna ONhE ('optic nerve hemoglobin'; Insoft SL, Tenerife, Spain) software was used to obtain the Hb analysis in each of the 24 sectors and average Hb of optic disc photographs acquired. Reproducibility of measurements provided by Laguna ONhE program was analysed. RESULTS: MS patients showed significant reduction of optic disc Hb percentages in average Hb (58.99% in MS, 65.39% in healthy subjects; p<0.001) and in almost all analysed sectors with the largest differences in temporal sectors. Laguna ONhE program showed good reproducibility measuring Hb percentages in MS patients and healthy subjects. CONCLUSIONS: Measurements of optic disc Hb levels obtained with Laguna ONhE software had good ability detecting optic atrophy and axonal loss in MS patients. This method had good reliability and is easy to implement in routine clinical practice.

Measuring hemoglobin levels in the optic nerve head: comparisons with other structural and functional parameters of glaucoma.

PURPOSE: We evaluated and compared the ability of a new method for measuring hemoglobin (Hb) levels at the optic nerve head (ONH) to that of visual field evaluation, scanning laser ophthalmoscopy (HRT), scanning laser polarimetry (GDx), and optical coherence tomography (OCT) for diagnosing glaucoma. METHODS: Healthy eyes (n = 102) and glaucomatous eyes (n = 101) underwent reliable Oculus Spark perimetry, and imaging with the HRT, GDx, and Cirrus OCT. In addition, ONH color images were acquired with a non-mydriatic fundus camera. The Laguna ON(h)E program then was used to calculate the Hb amount in each of 24 sectors of the ONH. Sensitivities at 95% fixed specificity, diagnostic agreement, and linear correlations between parameters with the best diagnostic ability were calculated. RESULTS: The glaucoma discriminant function (GDF) of the Laguna program, evaluating Hb in the vertical intermediate sectors and center/periphery Hb amount slope, yielded an 89.1% sensitivity and 95.1% specificity, which was superior or similar to the other tests. The best GDF diagnostic agreement was for the OCT-vertical cup-to-disc (C/D) ratio (kappa = 0.772) and the final phase Spark pattern SD (kappa = 0.672). Hb levels correlated strongly with the Spark mean sensitivity (first phase 0.70, final phase 0.71). Hb also correlated well with the Reinhard OW Burk discriminant function of the HRT (0.56), nerve fiber indicator of GDx (-0.64), and vertical C/D ratio of OCT (0.71). CONCLUSIONS: Hb levels evaluated by color analysis of ONH photographs had high reproducibility, a high sensitivity-specificity balance, and moderate to strong agreement with other structural and functional tests.

Iris center corneal reflection method for gaze tracking using visible light.

Very few attempts, if any, have been made to use visible light in corneal reflection approaches to the problem of gaze tracking. The reasons usually given to justify the limited application of this type of illumination are that the required image features are less accurately depicted, and that visible light may disturb the user. The aim of this paper is to show that it is possible to overcome these difficulties and build an accurate and robust gaze tracker under these circumstances. For this purpose, visible light is used to obtain the corneal reflection or glint in a way analogous to the well-known pupil center corneal reflection technique. Due to the lack of contrast, the center of the iris is tracked instead of the center of the pupil. The experiments performed in our laboratory have shown very satisfactory results, allowing free-head movement and no need of recalibration.