R. A. Welikala

Automated detection of proliferative diabetic retinopathy using a modified line operator and dual classification

Proliferative diabetic retinopathy (PDR) is a condition that carries a high risk of severe visual impairment. The hallmark of PDR is neovascularisation, the growth of abnormal new vessels. This paper describes an automated method for the detection of new vessels in retinal images. Two vessel segmentation approaches are applied, using the standard line operator and a novel modified line operator. The latter is designed to reduce false responses to non-vessel edges. Both generated binary vessel maps hold vital information which must be processed separately. This is achieved with a dual classification system. Local morphology features are measured from each binary vessel map to produce two separate feature sets. Independent classification is performed for each feature set using a support vector machine (SVM) classifier. The system then combines these individual classification outcomes to produce a final decision. Sensitivity and specificity results using a dataset of 60 images are 0.862 and 0.944 respectively on a per patch basis and 1.00 and 0.90 respectively on a per image basis.

Genetic algorithm based feature selection combined with dual classification for the automated detection of proliferative diabetic retinopathy

Proliferative diabetic retinopathy (PDR) is a condition that carries a high risk of severe visual impairment. The hallmark of PDR is the growth of abnormal new vessels. In this paper, an automated method for the detection of new vessels from retinal images is presented. This method is based on a dual classification approach. Two vessel segmentation approaches are applied to create two separate binary vessel map which each hold vital information. Local morphology features are measured from each binary vessel map to produce two separate 4-D feature vectors. Independent classification is performed for each feature vector using a support vector machine (SVM) classifier. The system then combines these individual outcomes to produce a final decision. This is followed by the creation of additional features to generate 21-D feature vectors, which feed into a genetic algorithm based feature selection approach with the objective of finding feature subsets that improve the performance of the classification. Sensitivity and specificity results using a dataset of 60 images are 0.9138 and 0.9600, respectively, on a per patch basis and 1.000 and 0.975, respectively, on a per image basis.

QUARTZ: Quantitative Analysis of Retinal Vessel Topology and size – An automated system for quantification of retinal vessels morphology

Abstract Retinal vessels are easily and non-invasively imaged using fundus cameras. Growing evidence including longitudinal evidence, suggests morphological changes in retinal vessels are early physio-markers of cardio-metabolic risk and outcome (as well as other disease processes). However, data from large population based studies are needed to examine the nature of these morphological associations. Several retinal image analysis (RIA) systems have been developed. While these provide a number of retinal vessel indices, they are often restricted in the area of analysis, and have limited automation, including the ability to distinguish between arterioles and venules. With the aim of developing reliable, automated, efficient retinal image analysis (RIA) software, generating a rich quantification of retinal vasculature in large volumes of fundus images, we present QUARTZ (Quantitative Analysis of Retinal Vessel Topology and size), a novel automated system for processing and analysing retinal images. QUARTZ consists of modules for vessel segmentation, width measurement and angular change at each vessel centreline pixel with sub-pixel accuracy, computing local vessel orientation, optic disc localisation, arteriole/venule classification, tortuosity measurement, and exporting the quantitative measurements in various output file formats. The performance metrics of the algorithms incorporated in QUARTZ are validated on a number of publically available retinal databases (including DRIVE, STARE, CHASE_DB1, INSPIRE-AVR, and DIARETDB1). QUARTZ performs well in terms of segmentation accuracy, calibre measurement, optic disc and arteriole/venule recognition. The system provides a rich quantification of retinal vessel morphology, which has potential medical applications in identifying those at high risk, so that prophylactic measure can be initiated before onset of overt disease.

Automated retinal image quality assessment on the UK Biobank dataset for epidemiological studies

Morphological changes in the retinal vascular network are associated with future risk of many systemic and vascular diseases. However, uncertainty over the presence and nature of some of these associations exists. Analysis of data from large population based studies will help to resolve these uncertainties. The QUARTZ (QUantitative Analysis of Retinal vessel Topology and siZe) retinal image analysis system allows automated processing of large numbers of retinal images. However, an image quality assessment module is needed to achieve full automation. In this paper, we propose such an algorithm, which uses the segmented vessel map to determine the suitability of retinal images for use in the creation of vessel morphometric data suitable for epidemiological studies. This includes an effective 3-dimensional feature set and support vector machine classification. A random subset of 800 retinal images from UK Biobank (a large prospective study of 500,000 middle aged adults; where 68,151 underwent retinal imaging) was used to examine the performance of the image quality algorithm. The algorithm achieved a sensitivity of 95.33% and a specificity of 91.13% for the detection of inadequate images. The strong performance of this image quality algorithm will make rapid automated analysis of vascular morphometry feasible on the entire UK Biobank dataset (and other large retinal datasets), with minimal operator involvement, and at low cost.

Differing matched filter responsivity for the detection of proliferative diabetic retinopathy

Diabetic retinopathy (DR) is a retinal vascular disease and is one of the most common causes of blindness worldwide. Proliferative diabetic retinopathy (PDR) is the most advanced stage of the disease and poses a high risk of severe visual impairment. PDR is characterised by the growth of abnormal new vessels known as neovascularisation. In this paper, we propose the use of the matched filter (MF) technique for vessel segmentation with emphasis on using two different sets of parameters to allow for the detection of new vessels. Parameters are selected to first increase and then decrease the MF response to new vessels, followed by thresholding to produce two separate binary vessel maps. The difference image removes most normal vessels and retains all possible new vessels, therefore making further analysis a much simpler task. Several steps are also included to reduce the detection of non vessel objects (dark and bright lesions). Five local features associated with the morphology of the vasculature are used to create a feature set. Based on these features, regions of the retina are categorized as normal or abnormal using a k-nearest neighbour classifier. Sensitivity and specificity results were 100% and 70% respectively on a per image basis.

Automated arteriole and venule classification using deep learning for retinal images from the UK Biobank cohort

The morphometric characteristics of the retinal vasculature are associated with future risk of many systemic and vascular diseases. However, analysis of data from large population based studies is needed to help resolve uncertainties in some of these associations. This requires automated systems that extract quantitative measures of vessel morphology from large numbers of retinal images. Associations between retinal vessel morphology and disease precursors/outcomes may be similar or opposing for arterioles and venules. Therefore, the accurate detection of the vessel type is an important element in such automated systems. This paper presents a deep learning approach for the automatic classification of arterioles and venules across the entire retinal image, including vessels located at the optic disc. This comprises of a convolutional neural network whose architecture contains six learned layers: three convolutional and three fully-connected. Complex patterns are automatically learnt from the data, which avoids the use of hand crafted features. The method is developed and evaluated using 835,914 centreline pixels derived from 100 retinal images selected from the 135,867 retinal images obtained at the UK Biobank (large population-based cohort study of middle aged and older adults) baseline examination. This is a challenging dataset in respect to image quality and hence arteriole/venule classification is required to be highly robust. The method achieves a significant increase in accuracy of 8.1% when compared to the baseline method, resulting in an arteriole/venule classification accuracy of 86.97% (per pixel basis) over the entire retinal image.

Automated retinal vessel recognition and measurements on large datasets

The characteristics of the retinal vascular network have been prospectively associated with many systemic and vascular diseases. QUARTZ is a fully automated software that has been developed to localize and quantify the morphological characteristics of blood vessels in retinal images for use in epidemiological studies. This software was used to analyse a dataset containing 16,000 retinal images from the EPIC-Norfolk cohort study. The objective of this paper is to both assess the suitability of this dataset for computational analysis and to further evaluate the QUARTZ software.

Retinal vasculometry associations with cardiometabolic risk factors in the European Prospective Investigation of Cancer Norfolk study

Purpose To examine associations between retinal vessel morphometry and cardiometabolic risk factors in older British men and women. Design Retinal imaging examination as part of the European Prospective Investigation into Cancer—Norfolk Eye Study. Participants Retinal imaging and clinical assessments were carried out in 7411 participants. Retinal images were analyzed using a fully automated validated computerized system that provides novel measures of vessel morphometry. Methods Associations between cardiometabolic risk factors, chronic disease, and retinal markers were analyzed using multilevel linear regression, adjusted for age, gender, and within-person clustering, to provide percentage differences in tortuosity and absolute differences in width. Main Outcomes Measures Retinal arteriolar and venular tortuosity and width. Results In all, 279 802 arterioles and 285 791 venules from 5947 participants (mean age, 67.6 years; standard deviation [SD], 7.6 years; 57% female) were analyzed. Increased venular tortuosity was associated with higher body mass index (BMI; 2.5%; 95% confidence interval [CI], 1.7%–3.3% per 5 kg/m2), hemoglobin A1c (HbA1c) level (2.2%; 95% CI, 1.0%–3.5% per 1%), and prevalent type 2 diabetes (6.5%; 95% CI, 2.8%–10.4%); wider venules were associated with older age (2.6 μm; 95% CI, 2.2–2.9 μm per decade), higher triglyceride levels (0.6 μm; 95% CI, 0.3–0.9 μm per 1 mmol/l), BMI (0.7 μm; 95% CI, 0.4–1.0 per 5 kg/m2), HbA1c level (0.4 μm; 95% CI, –0.1 to 0.9 per 1%), and being a current smoker (3.0 μm; 95% CI, 1.7–4.3 μm); smoking also was associated with wider arterioles (2.1 μm; 95% CI, 1.3–2.9 μm). Thinner venules were associated with high-density lipoprotein (HDL) (1.4 μm; 95% CI, 0.7–2.2 per 1 mmol/l). Arteriolar tortuosity increased with age (5.4%; 95% CI, 3.8%–7.1% per decade), higher systolic blood pressure (1.2%; 95% CI, 0.5%–1.9% per 10 mmHg), in females (3.8%; 95% CI, 1.4%–6.4%), and in those with prevalent stroke (8.3%; 95% CI, –0.6% to 18%); no association was observed with prevalent myocardial infarction. Narrower arterioles were associated with age (0.8 μm; 95% CI, 0.6–1.0 μm per decade), higher systolic blood pressure (0.5 μm; 95% CI, 0.4–0.6 μm per 10 mmHg), total cholesterol level (0.2 μm; 95% CI, 0.0–0.3 μm per 1 mmol/l), and HDL (1.2 μm; 95% CI, 0.7–1.6 μm per 1 mmol/l). Conclusions Metabolic risk factors showed a graded association with both tortuosity and width of retinal venules, even among people without clinical diabetes, whereas atherosclerotic risk factors correlated more closely with arteriolar width, even excluding those with hypertension and cardiovascular disease. These noninvasive microvasculature measures should be evaluated further as predictors of future cardiometabolic disease.

Microaneurysm detection in retinal images using an ensemble classifier

Diabetic Retinopathy (DR) is one of the leading causes of blindness amongst the working age population. The presence of microaneurysms (MA) in retinal images is a pathognomonic sign of DR. In this work we have presented a novel combination of algorithms applied to a public dataset for automated detection of MA in colour fundus images of the retina. The proposed technique first detects an initial set of candidates using a Gaussian Matched filter and then classifies the initial set of candidates in order to reduce the number of false positives. A Random Forest ensemble classifier using a set of 79 features (the most common features used within literature) was used for classification. Our proposed algorithm was evaluated on a subset of 20 images from the MESSIDOR dataset. We show that the use of the Random Forest classifier with the 79 features improves the sensitivity of the detection, compared to using a K-Nearest Neighbours classifier that has been proposed in other techniques. In addition, the Random Forest is capable of ranking features according to their importance. We have ranked the 79 features according to their importance. This ranking provides an insight into the most important features that are necessary for discriminating true MA candidates from spurious objects. Eccentricity, aspect ratio and moments are found to be among the important features.