Features extraction using encoded local binary pattern for detection and grading diabetic retinopathy

Abstract

Introduction Reliable computer diagnosis of diabetic retinopathy (DR) is needed to rescue many with diabetes who may be under threat of blindness. This research aims to detect the presence of diabetic retinopathy in fundus images and grade the disease severity without lesion segmentation. Methods To ensure that the fundus images are in a standard state of brightness, a series of preprocessing steps have been applied to the green channel image using histogram matching and a median filter. Then, contrast-limited adaptive histogram equalisation is performed, followed by the unsharp filter. The preprocessed image is divided into small blocks, and then each block is processed to extract uniform local binary patterns (LBPs) features. The extracted features are encoded, and the feature size is reduced to 3.5 percent of its original size. Classifiers like Support Vector Machine (SVM) and a proposed CNN model were used to classify retinal fundus images. The classification is abnormal or normal and to grade the severity of DR. Results Our feature extraction method was tested on a binary classifier and resulted in an accuracy of 98.37% and 98.84% on the Messidor2 and EyePACS databases, respectively. The proposed system could grade DR severity into three grades (0: no DR, 1: mild DR, and 5: moderate, severe NPDR, and PDR). It obtains an F1-score of 0.9617 and an accuracy of 95.37% on the EyePACS database, and an F1-score of 0.9860 and an accuracy of 97.57% on the Messidor2 database. The resultant values are dependent on the selection of (neighbours, radius) pairs during the extraction of LBP features. Conclusions This study’s results proved that the preprocessing steps are significant and had a great effect on highlighting image features. The novel method of stacking and encoding the LBP values in the feature vector greatly affects results when using SVM or CNN for classification. The proposed system outperforms the state of the artwork. The proposed CNN model performs better than SVM.

UC-stack: a deep learning computer automatic detection system for diabetic retinopathy classification

“…The preprocessed images are then fed into a custom deep convolutional model for feature extraction. Green channel enhancement (Berbar 2022) is a commonly used image enhancement technique that adjusts the color information in the image to improve contrast and visualization. In DR images, vessels and lesions typically exhibit different color intensities and contrasts.…”

“…Feng Li et al (2021a) evaluated the improved Inception-v4 method on the Messidor-2 dataset, achieving an AUC of 0.994. Berbar (2022) extracted local binary pattern features from fundus images and evaluated the method on the Messidor-2 and EyePACS databases, obtaining satisfactory results. Abdelmaksoud et al (2021) performed segmentation of exudates, microaneurysms, hemorrhages, and vessels using U-Net to highlight DR pathological changes and assist ophthalmologists in diagnosis and utilized support vector machine (SVM) for differentiating DR levels, achieving excellent performance on nine publicly available benchmark datasets.…”

Research on smoke edge detection algorithm based on structure similarity index measure

“…LBP (Local Binary Pattern) [6] is an algorithm that describes the local texture features of an image by reflecting the texture situation through the contrast between the target pixel and the surrounding pixels. LBP has significant advantages such as rotation invariance and grayscale invariance.…”

Multi-label classification of retinal disease via a novel vision transformer model

“…These methods include SVM and naive Bayes (NB) as the classifiers as well as the LBP for feature extraction. To discriminate the presence of DR and grade the severity of DR in retinal images without lesion segmentation, Berbar ( 2022 ) first employed the pre-processing techniques, including histogram matching and median filter, to the green channels of retinal images. Then, the contrast-limited adaptive histogram equalization was leveraged as well as the unsharp filter, to note that each image was segmented into small patches, from which the LBP features were generated.…”

“…In addition, an SVM was taken as the classifier to implement the retinal image classification. In general, the study of Berbar ( 2022 ) can grade the severity of DR into three different levels. Recently, the study of Reddy and Ravindran ( 2022 ) presented an automatic screening platform to recognize DR in retinal images.…”