Fundus Image Quality Assessment Research Articles

VISTA: vision improvement via split and reconstruct deep neural network for fundus image quality assessment

AbstractWidespread eye conditions such as cataracts, diabetic retinopathy, and glaucoma impact people worldwide. Ophthalmology uses fundus photography for diagnosing these retinal disorders, but fundus images are prone to image quality challenges. Accurate diagnosis hinges on high-quality fundus images. Therefore, there is a need for image quality assessment methods to evaluate fundus images before diagnosis. Consequently, this paper introduces a deep learning model tailored for fundus images that supports large images. Our division method centres on preserving the original image’s high-resolution features while maintaining low computing and high accuracy. The proposed approach encompasses two fundamental components: an autoencoder model for input image reconstruction and image classification to classify the image quality based on the latent features extracted by the autoencoder, all performed at the original image size, without alteration, before reassembly for decoding networks. Through post hoc interpretability methods, we verified that our model focuses on key elements of fundus image quality. Additionally, an intrinsic interpretability module has been designed into the network that allows decomposing class scores into underlying concepts quality such as brightness or presence of anatomical structures. Experimental results in our model with EyeQ, a fundus image dataset with three categories (Good, Usable, and Rejected) demonstrate that our approach produces competitive outcomes compared to other deep learning-based methods with an overall accuracy of 0.9066, a precision of 0.8843, a recall of 0.8905, and an impressive F1-score of 0.8868. The code is publicly available at https://github.com/saifalkhaldiurv/VISTA_-Image-Quality-Assessment.

Enhancing Retinal Fundus Image Quality Assessment With Swin-Transformer-Based Learning Across Multiple Color-Spaces.

The assessment of retinal image (RI) quality holds significant importance in both clinical trials and large datasets, because suboptimal images can potentially conceal early signs of diseases, thereby resulting in inaccurate medical diagnoses. This study aims to develop an automatic method for Retinal Image Quality Assessment (RIQA) that incorporates visual explanations, aiming to comprehensively evaluate the quality of retinal fundus images (RIs). We developed an automatic RIQA system, named Swin-MCSFNet, utilizing 28,792 RIs from the EyeQ dataset, as well as 2000 images from the EyePACS dataset and an additional 1,000 images from the OIA-ODIR dataset. After preprocessing, including cropping black regions, data augmentation, and normalization, a Swin-MCSFNet classifier based on the Swin-Transformer for multiple color-space fusion was proposed to grade the quality of RIs. The generalizability of Swin-MCSFNet was validated across multiple data centers. Additionally, for enhanced interpretability, a Score-CAM-generated heatmap was applied to provide visual explanations. Experimental results reveal that the proposed Swin-MCSFNet achieves promising performance, yielding a micro-receiver operating characteristic (ROC) of 0.93 and ROC scores of 0.96, 0.81, and 0.96 for the "Good," "Usable," and "Reject" categories, respectively. These scores underscore the accuracy of RIQA based on Swin-MCSF in distinguishing among the three categories. Furthermore, heatmaps generated across different RIQA classification scores and various color spaces suggest that regions in the retinal images from multiple color spaces contribute significantly to the decision-making process of the Swin-MCSFNet classifier. Our study demonstrates that the proposed Swin-MCSFNet outperforms other methods in experiments conducted on multiple datasets, as evidenced by the superior performance metrics and insightful Score-CAM heatmaps. This study constructs a new retinal image quality evaluation system, which will contribute to the subsequent research of retinal images.

FundusQ-Net: A regression quality assessment deep learning algorithm for fundus images quality grading

Objective: Ophthalmological pathologies such as glaucoma, diabetic retinopathy and age-related macular degeneration are major causes of blindness and vision impairment. There is a need for novel decision support tools that can simplify and speed up the diagnosis of these pathologies. A key step in this process is to automatically estimate the quality of the fundus images to make sure these are interpretable by a human operator or a machine learning model. We present a novel fundus image quality scale and deep learning (DL) model that can estimate fundus image quality relative to this new scale.Methods: A total of 1245 images were graded for quality by two ophthalmologists within the range 1–10, with a resolution of 0.5. A DL regression model was trained for fundus image quality assessment. The architecture used was Inception-V3. The model was developed using a total of 89,947 images from 6 databases, of which 1245 were labeled by the specialists and the remaining 88,702 images were used for pre-training and semi-supervised learning. The final DL model was evaluated on an internal test set (n=209) as well as an external test set (n=194).Results: The final DL model, denoted FundusQ-Net, achieved a mean absolute error of 0.61 (0.54–0.68) on the internal test set. When evaluated as a binary classification model on the public DRIMDB database as an external test set the model obtained an accuracy of 99%.Significance: the proposed algorithm provides a new robust tool for automated quality grading of fundus images.

Digital fundus image quality assessment

Diabetic retinopathy (DR) is a disease caused by complications of diabetes. It starts asymptomatically and can end in blindness. To detect it, doctors use special fundus cameras that allow them to register images of the retina in the visible range of the spectrum. On these images one can see features, which determine the presence of DR and its grade. Researchers around the world are developing systems for the automated analysis of fundus images. At present, the level of accuracy of classification of diseases caused by DR by systems based on machine learning is comparable to the level of qualified medical doctors.The article shows variants for representation of the retina in digital images by different cameras. We define the task to develop a universal approach for the image quality assessment of a retinal image obtained by an arbitrary fundus camera. It is solved in the first block of any automated retinal image analysis system. The quality assessment procedure is carried out in several stages. At the first stage, it is necessary to perform binarization of the original image and build a retinal mask. Such a mask is individual for each image, even among the images recorded by one camera. For this, a new universal retinal image binarization algorithm is proposed. By analyzing result of the binarization, it is possible to identify and remove imagesoutliers, which show not the retina, but other objects. Further, the problem of no-reference image quality assessment is solved and images are classified into two classes: satisfactory and unsatisfactory for analysis. Contrast, sharpness and possibility of segmentation of the vascular system on the retinal image are evaluated step by step. It is shown that the problem of no-reference image quality assessment of an arbitrary fundus image can be solved.Experiments were performed on a variety of images from the available retinal image databases.

Topological Data Analysis for Eye Fundus Image Quality Assessment.

The objective of this work is to perform image quality assessment (IQA) of eye fundus images in the context of digital fundoscopy with topological data analysis (TDA) and machine learning methods. Eye health remains inaccessible for a large amount of the global population. Digital tools that automize the eye exam could be used to address this issue. IQA is a fundamental step in digital fundoscopy for clinical applications; it is one of the first steps in the preprocessing stages of computer-aided diagnosis (CAD) systems using eye fundus images. Images from the EyePACS dataset were used, and quality labels from previous works in the literature were selected. Cubical complexes were used to represent the images; the grayscale version was, then, used to calculate a persistent homology on the simplex and represented with persistence diagrams. Then, 30 vectorized topological descriptors were calculated from each image and used as input to a classification algorithm. Six different algorithms were tested for this study (SVM, decision tree, k-NN, random forest, logistic regression (LoGit), MLP). LoGit was selected and used for the classification of all images, given the low computational cost it carries. Performance results on the validation subset showed a global accuracy of 0.932, precision of 0.912 for label “quality” and 0.952 for label “no quality”, recall of 0.932 for label “quality” and 0.912 for label “no quality”, AUC of 0.980, F1 score of 0.932, and a Matthews correlation coefficient of 0.864. This work offers evidence for the use of topological methods for the process of quality assessment of eye fundus images, where a relatively small vector of characteristics (30 in this case) can enclose enough information for an algorithm to yield classification results useful in the clinical settings of a digital fundoscopy pipeline for CAD.

Fundus image quality assessment for an eye model with cataract using adapted illumination patterns

PurposeIn recent years, different methods have been developed to improve fundus image quality in eyes with cataract. For stray light reduction, we have introduced a segmented LED light source. To analyze the effect of different illumination patterns at different cataract characteristics, we developed an optical cataract eye model. Accordingly, we performed fundus imaging using a fundus camera equipped with a LED light source controlled by a Raspberry Pi.MethodsA customized 24‐fold segmented LED light source was connected to a fundus camera. The eye model consists of a plano‐convex lens, a cover glass, and a test chart with five black‐white test fields. The pupil diameter was set to 7 mm. To test the influence of variable illumination patterns, four different cortical cataract patterns were produced by sand blasting the cover glasses. Three spoke‐like (1Spoke, 4S, 7S) patterns and one crescent‐shaped/CS pattern were realized. To analyze the effect of adapted illumination conditions, the gray‐level co‐occurrence matrix (GLCM) contrasts were calculated. At that, one image was taken at standard annular illumination (all LEDs on) as baseline followed by seven defined adapted configurations, for each type of cataract.ResultsFor the four types of cataracts, the baseline GLCM contrast values were 0.21 (1S), 0.18 (4S), 0.17 (7S), and 0.22 (CS). In each cataract type, the best‐fitting illumination pattern resulted in 0.23 (1S), 0.29 (4S), 0.22 (7S), and 0.23 (CS). Compared to the baseline values, the increases of the GLCM contrasts were 9.5% (1S), 61% (4S), 29% (7S), and 4.5% (CS).ConclusionsWe successfully performed structured LED illumination for different cortical cataract types using an optical eye model. An improvement of fundus image quality could be determined in all investigated cases using the gray‐level co‐occurrence matrix contrast.

Development and application of a fundus image quality assessment system based on computer vision technology

Objective: To develop a fundus image quality assessment system based on computer vision technology and to verify its accuracy by comparing the results of artificial discrimination and using this system. Methods: The process of image evaluation was divided into four modules: fundus image preprocessing, fundus image quality evaluation, fundus image content detection and evaluation result output. The system was designed to automatically evaluate the image quality of each fundus image, identify the optic disc and macula, and judge whether the image was qualified or not according to the image quality discrimination rules. A total of 2 397 fundus images of 787 type 2 diabetes patients were selected as the test data set. The average age of the patients, including 384 males and 403 females, was (69.65±19.09) years old. The images were taken by the staff of community health service centers in Shanghai with a fundus camera. The fundus image quality assessment system was used to conduct quality control and classification of the data set. At the same time, 12 professional fundus picture readers were employed to conduct manual quality control and classification of this data set. The system quality control results and artificial quality discrimination results were compared and analyzed. Results: The fundus image quality assessment system automatically recognized left and right eyes and eye positions on the input fundus images. The quality control interface included four indicator lights, which respectively corresponded to the images with the optic disc or macula as the center of the left or right eye. Evaluation of each fundus image was completed within 1 second, and the results were automatically displayed on the user interface. The 2 397 fundus photos were identified manually as 1 846 qualified photos and 551 unqualified photos. Among the unqualified images, 62 (11.27%) were too dark, 51 (9.27%) were too bright, 59 (10.73%) were not clear in the macular area, 36 (6.54%) showed no macula or optic disc, 125 (22.73%) could not present the fundus structure, 175 (31.82%) were blurred, and 42 (7.64%) were blocked. The results of the system and manual assessment were consistent in 1 788 qualified images (96.86%) and 550 unqualified images (99.82%), with an overall consistency rate of 97.54%. Conclusion: The fundus image quality assessment system can achieve highly consistent results with the professional judgment of ophthalmologists and has the characteristics of objectivity. (Chin J Ophthalmol, 2020, 56:920-927).

Automatic fundus image quality assessment on a continuous scale

Fundus photography is commonly used for screening, diagnosis, and monitoring of various diseases affecting the eye. In addition, it has shown promise in the diagnosis of brain diseases and evaluation of cardiovascular risk factors. Good image quality is important if diagnosis is to be accurate and timely. Here, we propose a method that automatically grades image quality on a continuous scale which is more flexible than binary quality classification. The method utilizes random forest regression models trained on image features discovered automatically by combining basic image filters using simulated annealing as well as features extracted with the discrete Fourier transform. The method was developed and tested on images from two different fundus camera models. The quality of those images was rated on a continuous scale from 0.0 to 1.0 by five experts. In addition, the method was tested on DRIMDB, a publicly available dataset with binary quality ratings. On the DRIMDB dataset the method achieves an accuracy of 0.981, sensitivity of 0.993 and specificity of 0.958 which is consistent with the state of the art. When evaluating image quality on a continuous scale the method outperforms human raters.

Small sample color fundus image quality assessment based on gcforest

Color fundus image quality greatly influence the doctors’ diagnostic accuracy. However, the problems of imbalance data and small sample are the key issues of the color fundus images quality assessment. Hence, this paper purposes a small sample color fundus image quality assessment based on gcforest to solve these problems. Firstly, this paper extracts color and texture features to represent the quality of color fundus image. Next, re-sampling process is used to re-balance training data. Thirdly, the training data after re-balanced is sent to train gcforest which is a forest integration model. Finally, the trained gcforest which is good for small sample problem is used to evaluate color fundus images quality. Experiments demonstrate that the proposed method not only in color fundus image quality assessment but also in glaucoma classification task get good results.

Automated fundus image quality assessment and segmentation of optic disc using convolutional neural networks

An automated fundus image analysis is used as a tool for the diagnosis of common retinal diseases. A good quality fundus image results in better diagnosis and hence discarding the degraded fundus images at the time of screening itself provides an opportunity to retake the adequate fundus photographs, which save both time and resources. In this paper, we propose a novel fundus image quality assessment (IQA) model using the convolutional neural network (CNN) based on the quality of optic disc (OD) visibility. We localize the OD by transfer learning with Inception v-3 model. Precise segmentation of OD is done using the GrabCut algorithm. Contour operations are applied to the segmented OD to approximate it to the nearest circle for finding its center and diameter. For training the model, we are using the publicly available fundus databases and a private hospital database. We have attained excellent classification accuracy for fundus IQA on DRIVE, CHASE-DB, and HRF databases. For the OD segmentation, we have experimented our method on DRINS-DB, DRISHTI-GS, and RIM-ONE v.3 databases and compared the results with existing state-of-the-art methods. Our proposed method outperforms existing methods for OD segmentation on Jaccard index and F-score metrics.

Domain-invariant interpretable fundus image quality assessment.

Objective and quantitative assessment of fundus image quality is essential for the diagnosis of retinal diseases. The major factors in fundus image quality assessment are image artifact, clarity, and field definition. Unfortunately, most of existing quality assessment methods focus on the quality of overall image, without interpretable quality feedback for real-time adjustment. Furthermore, these models are often sensitive to the specific imaging devices, and cannot generalize well under different imaging conditions. This paper presents a new multi-task domain adaptation framework to automatically assess fundus image quality. The proposed framework provides interpretable quality assessment with both quantitative scores and quality visualization for potential real-time image recapture with proper adjustment. In particular, the present approach can detect optic disc and fovea structures as landmarks, to assist the assessment through coarse-to-fine feature encoding. The framework also exploit semi-tied adversarial discriminative domain adaptation to make the model generalizable across different data sources. Experimental results demonstrated that the proposed algorithm outperforms different state-of-the-art approaches and achieves an area under the ROC curve of 0.9455 for the overall quality classification.

Multivariate Regression-Based Convolutional Neural Network Model for Fundus Image Quality Assessment

Objectively assessing the perceptual quality of an ocular fundus image is essential for the reliable diagnosis of various ocular diseases. A fair amount of work has been done in this field to date. However, the generalizability of the current work is limited, as the existing quality models were developed and evaluated with data-sets built with limited subjective inputs. This paper aims at addressing this limitation with the following two contributions. First, a new fundus image quality assessment (FIQuA) data-set is presented, containing 1500 fundus images with three classes of quality: Good, Fair, and Poor. Also, for each image, subjective scores (in the range [0-10]) were collected for six quality parameters, including structural and generic properties of the fundus images. Second, a new multivariate regression based convolutional neural network (CNN) model is proposed to predict the fundus image quality. The proposed model consists of two individually trained blocks. The first block consists of four pre-trained models, trained against the subjective scores for the six quality parameters, and aims at deriving the optimized features for classification. Next, the optimized features from each of the four models are ensembled together and transferred to the second block for final classification. The proposed model achieves a strong correlation with the subjective scores, with the values 0.941 , 0.954 , 0.853 , and 0.401 obtained for SROCC, LCC, KCC, and RMSE respectively. Its classification accuracy is 95.66% over the FIQuA data-set, and 98.96% and 88.43% respectively over the two publicly available data-sets DRIMDB and EyeQ.

Fundus image quality assessment: survey, challenges, and future scope

Various ocular diseases, such as cataract, diabetic retinopathy, and glaucoma have affected a large proportion of the population worldwide. In ophthalmology, fundus photography is used for the diagnosis of such retinal disorders. Nowadays, the set-up of fundus image acquisition has changed from a fixed position to portable devices, making acquisition more vulnerable to distortions. However, a trustworthy diagnosis solely relies upon the quality of the fundus image. In recent years, fundus image quality assessment (IQA) has drawn much attention from researchers. This study presents a detailed survey of the fundus IQA research. The survey covers a comprehensive discussion on the factors affecting the fundus image quality and the real-time distortions. The fundus IQA algorithms have been analysed on the basis of the methodologies used and divided into three classes, namely: (i) similarity-based, (ii) segmentation-based, and (iii) machine learning based. In addition, limitations of state of the art in this research field are also presented with the possible solutions. The objective of this study is to provide a detailed information about the fundus IQA research with its significance, present status, limitations, and future scope. To the best of the authors' knowledge, this is the first survey paper on the fundus IQA research.

Automated Fundus Image Quality Assessment in Retinopathy of Prematurity Using Deep Convolutional Neural Networks

Accurate image-based ophthalmic diagnosis relies on fundus image clarity. This has important implications for the quality of ophthalmic diagnoses and for emerging methods such as telemedicine and computer-based image analysis. The purpose of this study was to implement a deep convolutional neural network (CNN) for automated assessment of fundus image quality in retinopathy of prematurity (ROP). Experimental study. Retinal fundus images were collected from preterm infants during routine ROP screenings. Six thousand one hundred thirty-nine retinal fundus images were collected from 9 academic institutions. Each image was graded for quality (acceptable quality [AQ], possibly acceptable quality [PAQ], or not acceptable quality [NAQ]) by 3 independent experts. Quality was defined as the ability to assess an image confidently for the presence of ROP. Of the 6139 images, NAQ, PAQ, and AQ images represented 5.6%, 43.6%, and 50.8% of the image set, respectively. Because of low representation of NAQ images in the data set, images labeled NAQ were grouped into the PAQ category, and a binary CNN classifier was trained using 5-fold cross-validation on 4000 images. A test set of 2109 images was held out for final model evaluation. Additionally, 30 images were ranked from worst to best quality by 6 experts via pairwise comparisons, and the CNN's ability to rank quality, regardless of quality classification, was assessed. The CNN performance was evaluated using area under the receiver operating characteristic curve (AUC). A Spearman's rank correlation was calculated to evaluate the overall ability of the CNN to rank images from worst to best quality as compared with experts. The mean AUC for 5-fold cross-validation was 0.958 (standard deviation, 0.005) for the diagnosis of AQ versus PAQ images. The AUC was 0.965 for the test set. The Spearman's rank correlation coefficient on the set of 30 images was 0.90 as compared with the overall expert consensus ranking. This model accurately assessed retinal fundus image quality in a comparable manner with that of experts. This fully automated model has potential for application in clinical settings, telemedicine, and computer-based image analysis in ROP and for generalizability to other ophthalmic diseases.

Computer-aided diagnosis based on enhancement of degraded fundus photographs.

Retinal imaging is an important and effective tool for detecting retinal diseases. However, degraded images caused by the aberrations of the eye can disguise lesions, so that a diseased eye can be mistakenly diagnosed as normal. In this work, we propose a new image enhancement method to improve the quality of degraded images. A new method is used to enhance degraded-quality fundus images. In this method, the image is converted from the input RGB colour space to LAB colour space and then each normalized component is enhanced using contrast-limited adaptive histogram equalization. Human visual system (HVS)-based fundus image quality assessment, combined with diagnosis by experts, is used to evaluate the enhancement. The study included 191 degraded-quality fundus photographs of 143 subjects with optic media opacity. Objective quality assessment of image enhancement (range: 0-1) indicated that our method improved colour retinal image quality from an average of 0.0773 (variance 0.0801) to an average of 0.3973 (variance 0.0756). Following enhancement, area under curves (AUC) were 0.996 for the glaucoma classifier, 0.989 for the diabetic retinopathy (DR) classifier, 0.975 for the age-related macular degeneration (AMD) classifier and 0.979 for the other retinal diseases classifier. The relatively simple method for enhancing degraded-quality fundus images achieves superior image enhancement, as demonstrated in a qualitative HVS-based image quality assessment. This retinal image enhancement may, therefore, be employed to assist ophthalmologists in more efficient screening of retinal diseases and the development of computer-aided diagnosis.

Human Visual System-Based Fundus Image Quality Assessment of Portable Fundus Camera Photographs.

Telemedicine and the medical "big data" era in ophthalmology highlight the use of non-mydriatic ocular fundus photography, which has given rise to indispensable applications of portable fundus cameras. However, in the case of portable fundus photography, non-mydriatic image quality is more vulnerable to distortions, such as uneven illumination, color distortion, blur, and low contrast. Such distortions are called generic quality distortions. This paper proposes an algorithm capable of selecting images of fair generic quality that would be especially useful to assist inexperienced individuals in collecting meaningful and interpretable data with consistency. The algorithm is based on three characteristics of the human visual system--multi-channel sensation, just noticeable blur, and the contrast sensitivity function to detect illumination and color distortion, blur, and low contrast distortion, respectively. A total of 536 retinal images, 280 from proprietary databases and 256 from public databases, were graded independently by one senior and two junior ophthalmologists, such that three partial measures of quality and generic overall quality were classified into two categories. Binary classification was implemented by the support vector machine and the decision tree, and receiver operating characteristic (ROC) curves were obtained and plotted to analyze the performance of the proposed algorithm. The experimental results revealed that the generic overall quality classification achieved a sensitivity of 87.45% at a specificity of 91.66%, with an area under the ROC curve of 0.9452, indicating the value of applying the algorithm, which is based on the human vision system, to assess the image quality of non-mydriatic photography, especially for low-cost ophthalmological telemedicine applications.