Combination of Global Features for the Automatic Quality Assessment of Retinal Images

Abstract

Diabetic retinopathy (DR) is one of the most common causes of visual loss in developed countries. Computer-aided diagnosis systems aimed at detecting DR can reduce the workload of ophthalmologists in screening programs. Nevertheless, a large number of retinal images cannot be analyzed by physicians and automatic methods due to poor quality. Automatic retinal image quality assessment (RIQA) is needed before image analysis. The purpose of this study was to combine novel generic quality features to develop a RIQA method. Several features were calculated from retinal images to achieve this goal. Features derived from the spatial and spectral entropy-based quality (SSEQ) and the natural images quality evaluator (NIQE) methods were extracted. They were combined with novel sharpness and luminosity measures based on the continuous wavelet transform (CWT) and the hue saturation value (HSV) color model, respectively. A subset of non-redundant features was selected using the fast correlation-based filter (FCBF) method. Subsequently, a multilayer perceptron (MLP) neural network was used to obtain the quality of images from the selected features. Classification results achieved 91.46% accuracy, 92.04% sensitivity, and 87.92% specificity. Results suggest that the proposed RIQA method could be applied in a more general computer-aided diagnosis system aimed at detecting a variety of retinal pathologies such as DR and age-related macular degeneration.

Magnetic Resonance Image Quality Assessment by Using Non-Maximum Suppression and Entropy Analysis

“…Entropy is the fundamental concept of Shannon information theory [ 23 , 24 ]. It is usually considered in the framework of measure theory.…”

Retinal Image Quality Assessment Using Morphological Operations

“…In some hospitals, the department that does image capturing is different from Figure 1. Image clarity versus visibility the ophthalmologist's department, so an ophthalmologist must request the patient to go back and have a clear image captured which is a tedious and costly process [8]. Using unclear images for the disease diagnosis can lead to wrong recommendation for medication and treatment causing severe problems including blindness.…”

A Teleophthalmology Support System Based on the Visibility of Retinal Elements Using the CNNs

“…Taking into account the above-mentioned situation, until now several automatic quality assessment methods for retinal images have been proposed, which are classified into three categories: generic featured-based methods [4][5][6][7], structural feature-based methods [8][9][10][11] and convolutional neural network (CNN)-based methods [12][13][14][15]. In the generic feature-based methods, some generic features of images, such as: sharpness, contrast, luminance and texture properties, are calculated and used to determine if an input image has good quality or inadequate quality for a reliable diagnosis [4][5][6][7]. In the structural feature-based methods, firstly some anatomical retinal structures, such as optic disc (OD), vessel tree and macula, are localized and/or segmented.…”