Advances in Image Processing Techniques for Drusens Detection and Quantification in Fundus Images

Abstract

Age-Related Macular Degeneration (ARMD) is considered the leading cause of irreversible blindness in developed countries. One of its risk factors is the presence of drusens, which are retina abnormalities appearing as yellowish spots in fundus images. In this article a methodology using image processing techniques for the quantification of drusens is presented. The method uses splines combined with a contrast normalization to correct uneven illumination, followed by a drusen detection and modelling algorithm. The detection uses a gradient based segmentation algorithm that isolates drusens. They are then fitted by Gaussian functions, producing a model that is used to compute the area affected. To validate the methodology, 22 images were marked by three ophthalmologists and compared to the automated method. The sensitivity and specificity for the automated process (0.664 and 0.963) were comparable to that obtained among the specialists (0.656 and 0.971). Also, the Intraclass Correlation Coefficient showed an agreement of 74.9% between the processed images and the specialists' analysis.

Automated pathologies detection in retina digital images based on complex continuous wavelet transform phase angles

“…As shown in Table 1, our results compare favourably with the results obtained by other algorithms found in the literature to detect exudates [1, 2], drusens [3, 4] and MA [5, 6], especially for MA pathology. The main advantages of our approach are that it is simple, fast and effective.…”

“…We have presented a new automated system for pathologies detection in retina digital images based on the analysis of the image by means of the CCWT to obtain phase angle information which is fed to SVM. The simulation results show that our method gives results that compare favourably with the results reported in [1–6]. This could be explained mainly by the fact that phase angles contain valuable information that allows discriminating between normal and abnormal retina images.…”

“…Automatic digital image processing systems are becoming increasingly important in health care, since they allow efficient processing of images and objective medical diagnostics. In recent years, a particular interest has been devoted to the design of an automated computer‐based system for the screening of retina digital images [1] to detect the presence of exudates [2, 3], drusen [4, 5] and microaneurysm (MA) [6, 7]. In the problem of exudates detection [2]; contrast enhancement fuzzy C‐means and support vector machines (SVMs) were used to detect and classify bright lesions.…”

“…The proposed system achieved a sensitivity of 83%. A gradient‐based segmentation algorithm based on labelling the maximum gradient path was adopted in [5] to isolate drusens. Then, they were fitted by Gaussian functions to compute the area affected.…”

Analysis of gabor filter based features with PCA and GA for the detection of drusen in fundus images

“…Drusen affected area detection using intensity gradient based segmentation. Gaussian function model has been suggested by Mora A et al [6]. Wavelet transforms in combination with Fuzzy C Means clustering has been reported by Rama Prasath A et al [7].…”