Detection of Hard Exudates in Colour Fundus Images Using Fuzzy Support Vector Machine-Based Expert System

Diabetic retinopathy is a major cause of vision loss in diabetic patients. Currently, there is a need for making decisions using intelligent computer algorithms when screening a large volume of data. This paper presents an expert decision-making system designed using a fuzzy support vector machine (FSVM) classifier to detect hard exudates in fundus images. The optic discs in the colour fundus images are segmented to avoid false alarms using morphological operations and based on circular Hough transform. To discriminate between the exudates and the non-exudates pixels, colour and texture features are extracted from the images. These features are given as input to the FSVM classifier. The classifier analysed 200 retinal images collected from diabetic retinopathy screening programmes. The tests made on the retinal images show that the proposed detection system has better discriminating power than the conventional support vector machine. With the best combination of FSVM and features sets, the area under the receiver operating characteristic curve reached 0.9606, which corresponds to a sensitivity of 94.1 % with a specificity of 90.0 %. The results suggest that detecting hard exudates using FSVM contribute to computer-assisted detection of diabetic retinopathy and as a decision support system for ophthalmologists.     

Ensemble selection for feature-based classification of diabetic maculopathy images

As diabetic maculopathy (DM) is a prevalent cause of blindness in the world, it is increasingly important to use automated techniques for the early detection of the disease. In this paper, we propose a decision system to classify DM fundus images into normal, clinically significant macular edema (CMSE), and non-clinically significant macular edema (non-CMSE) classes. The objective of the proposed decision system is three fold namely, to automatically extract textural features (both region specific and global), to effectively choose subset of discriminatory features, and to classify DM fundus images to their corresponding class of disease severity. The system uses a gamut of textural features and an ensemble classifier derived from four popular classifiers such as the hidden naïve Bayes, naïve Bayes, sequential minimal optimization (SMO), and the tree-based J48 classifiers. We achieved an average classification accuracy of 96.7% using five-fold cross validation.     

Prognostic factors and retinopathy in type 1 diabetics in Iceland

Prognostic factors, particularly those related to metabolic control, were analysed individually over a period of 12 months prior to fundus photography in 149 type 1 (insulin-dependent, ketosis-prone) diabetics in Iceland. Patients without retinopathy in their first 20 years of diabetes visited the Diabetic Clinic significantly more frequently than those with retinopathy (p less than 0.05). Mean blood sugar values and mean per cent body weight did not differ between patients with or without retinopathy. Few between patients with or without retinopathy. Few patients were obese. In patients with 5-9 years' duration of diabetes, retinopathy was first seen after 7.7 +/- 0.3 years (mean +/- SEM). Those with retinopathy were significantly older at the time of the diagnosis of diabetes than those without eye lesions (p less than 0.025), a tendency also noted in those with 10-19 years' duration (p less than 0.10). Conversely, patients without retinopathy after diabetes for 20 years or more were significantly older at the time of diagnosis (p less than 0.02). They used significantly less insulin daily than those with retinal lesions (p less than 0.005) and 70% were females. Subsequently, a significant (p less than 0.047) male excess was found to characterize a group of patients with proliferative retinopathy.     

2型糖尿病视网膜病变患者血清sVCAM-1水平的变化和意义

研究糖尿病视网膜病变不同时期血清可溶性血管内皮细胞粘附分子-1(sVCAM-1)与视网膜病变严重程度、诊断病程、血清胰岛素和血糖的相关性.采用ELISA法检测85例2型糖尿病视网膜病变患者血清sVCAM-1含量, 由同一眼科医师通过眼底镜或荧光造影检查, 将患者分为无视网膜病变组(NDR)、背景期视网膜病变组(BDR)和增殖期视网膜病变组(PDR).结果显示, 三组糖尿病患者血清sVCAM-1水平均高于对照组, 其中PDR组、BDR组血清sVCAM-1水平与对照组和NDR组比较, 均有显著性差异(P＜0.01), NDR组与对照组比较也存在显著性差异(P＜0.01).糖尿病患者血清sVCAM-1水平与血糖、血清胰岛素、诊断病程均无相关性(P＞0.05).研究表明, 糖尿病视网膜病变不同时期血清sVCAM-1水平的变化可作为判断视网膜病变发展和严重程度的指标, 为临床早期发现和治疗糖尿病视网膜病变提供较好的依据.     

DM2患者血糖偏移与DR的关系

目的:探讨血糖偏移和糖尿病视网膜病变(diabetic retinopathy,DR)的关系.方法:采用动态血糖监测系统监测24例新发2型糖尿病(DM2)患者日内血糖的偏移情况,并行眼底照相,做回归分析.结果:DR与平均血糖偏移幅度差异无统计学意义(P＞0.05),而与早餐后2h和22:00～0:00的血糖偏移差异有统...     

Diagnosis, therapy and follow-up of diabetic eye diseases

Diabetes mellitus causes diabetic retinopathy and maculopathy, optical nerve neuropathy, cataract and defects of the eye muscles. The incidence of these defects correlates with duration and quality of the metabolic control. The recommendations of the Austrian Diabetes Association for the diagnosis, the therapeutic measures and requirements for adequate follow-up depending on the stages of the different forms of diabetic eye diseases are summarized.     

糖尿病视网膜病变图像数据库建立及应用

目的评价基于糖尿病视网膜病变(DR)图像数据库的临床应用技术,探讨其在临床DR早期诊断中的作用。方法采集1100例DR患者的眼底图像,经过初选和归纳分析后选择300只眼,建立微型DR图像数据库及应用软件技术。对100例糖尿病患者的眼底照相和荧光素眼底血管造影(FFA)检查图像进行人工直接诊断,并进行基于DR图像数据库技术的虚拟FFA诊断,对两者各层诊断符合情况进行统计分析,确定其诊断一致性。结果在DR患者不同分期的层次中,FFA检查诊断相符程度均大于80%,人工直接诊断和虚拟FFA诊断的总符合率为90%。结论虚拟FFA诊断与DR眼底照相诊断有较高的一致性,该图像数据库的建立和应用技术可指导和规范临床DR的早期诊疗,也为有创性FFA检查在DR诊治中合理应用提供依据和参考。     

Optic disc detection in color fundus images using ant colony optimization

Diabetic retinopathy has been revealed as the most common cause of blindness among people of working age in developed countries. However, loss of vision could be prevented by an early detection of the disease and, therefore, by a regular screening program to detect retinopathy. Due to its characteristics, the digital color fundus photographs have been the easiest way to analyze the eye fundus. An important prerequisite for automation is the segmentation of the main anatomical features in the image, particularly the optic disc. Currently, there are many works reported in the literature with the purpose of detecting and segmenting this anatomical structure. Though, none of them performs as needed, especially when dealing with images presenting pathologies and a great variability. Ant colony optimization (ACO) is an optimization algorithm inspired by the foraging behavior of some ant species that has been applied in image processing with different purposes. In this paper, this algorithm preceded by anisotropic diffusion is used for optic disc detection in color fundus images. Experimental results demonstrate the good performance of the proposed approach as the optic disc was detected in most of all the images used, even in the images with great variability.     

Rapid Grading of Fundus Photographs for Diabetic Retinopathy Using Crowdsourcing

Background

Screening for diabetic retinopathy is both effective and cost-effective, but rates of screening compliance remain suboptimal. As screening improves, new methods to deal with screening data may help reduce the human resource needs. Crowdsourcing has been used in many contexts to harness distributed human intelligence for the completion of small tasks including image categorization.

Objective

Our goal was to develop and validate a novel method for fundus photograph grading.

Methods

An interface for fundus photo classification was developed for the Amazon Mechanical Turk crowdsourcing platform. We posted 19 expert-graded images for grading by Turkers, with 10 repetitions per photo for an initial proof-of-concept (Phase I). Turkers were paid US $0.10 per image. In Phase II, one prototypical image from each of the four grading categories received 500 unique Turker interpretations. Fifty draws of 1-50 Turkers were then used to estimate the variance in accuracy derived from randomly drawn samples of increasing crowd size to determine the minimum number of Turkers needed to produce valid results. In Phase III, the interface was modified to attempt to improve Turker grading.

Results

Across 230 grading instances in the normal versus abnormal arm of Phase I, 187 images (81.3%) were correctly classified by Turkers. Average time to grade each image was 25 seconds, including time to review training images. With the addition of grading categories, time to grade each image increased and percentage of images graded correctly decreased. In Phase II, area under the curve (AUC) of the receiver-operator characteristic (ROC) indicated that sensitivity and specificity were maximized after 7 graders for ratings of normal versus abnormal (AUC=0.98) but was significantly reduced (AUC=0.63) when Turkers were asked to specify the level of severity. With improvements to the interface in Phase III, correctly classified images by the mean Turker grade in four-category grading increased to a maximum of 52.6% (10/19 images) from 26.3% (5/19 images). Throughout all trials, 100% sensitivity for normal versus abnormal was maintained.

Conclusions

With minimal training, the Amazon Mechanical Turk workforce can rapidly and correctly categorize fundus photos of diabetic patients as normal or abnormal, though further refinement of the methodology is needed to improve Turker ratings of the degree of retinopathy. Images were interpreted for a total cost of US $1.10 per eye. Crowdsourcing may offer a novel and inexpensive means to reduce the skilled grader burden and increase screening for diabetic retinopathy.     

Non-proliferative diabetic retinopathy symptoms detection and classification using neural network

Diabetic retinopathy (DR) causes blindness in the working age for people with diabetes in most countries. The increasing number of people with diabetes worldwide suggests that DR will continue to be major contributors to vision loss. Early detection of retinopathy progress in individuals with diabetes is critical for preventing visual loss. Non-proliferative DR (NPDR) is an early stage of DR. Moreover, NPDR can be classified into mild, moderate and severe. This paper proposes a novel morphology-based algorithm for detecting retinal lesions and classifying each case. First, the proposed algorithm detects the three DR lesions, namely haemorrhages, microaneurysms and exudates. Second, we defined and extracted a set of features from detected lesions. The set of selected feature emulates what physicians looked for in classifying NPDR case. Finally, we designed an artificial neural network (ANN) classifier with three layers to classify NPDR to normal, mild, moderate and severe. Bayesian regularisation and resilient backpropagation algorithms are used to train ANN. The accuracy for the proposed classifiers based on Bayesian regularisation and resilient backpropagation algorithms are 96.6 and 89.9, respectively. The obtained results are compared with results of the recent published classifier. Our proposed classifier outperforms the best in terms of sensitivity and specificity.     

妊娠期糖尿病合并视网膜病变患者21例眼科随访观察

目的:了解妊娠期糖尿病(GDM)患者视网膜病变的进展及预后.方法:经孕妇常规糖尿病筛查确诊的GDM患者132例,其中合并DRP 21例(A组),自GDM患者中随机抽取20例(B组),选正常孕妇20人为对照组(C组).分别于妊娠6个月、9个月及产后6个月统计并比较3组对象的矫正视力、眼压及眼底检查结果.结果:妊娠晚期,微血管瘤数量增加者8例(38.1%),但所有病例仍为单纯性DRP;产后6个月,微血管瘤数量减少者15例.DRP组患者的矫正视力在妊娠9个月时与妊娠6个月和产后6个月比较有统计学差异(P＜0.05),与对照组比较视力最差(P＜0.05).结论:在妊娠期眼底正常或仅有单纯性糖尿病视网膜病变患者,其具有自愈性,预后较好,无需特殊处理,但需定期随诊眼底.如发现有增殖性病变应及时进行激光治疗.     

糖尿病视网膜病MTHFR基因多态性及其与血浆同型半胱氨酸水平的关系

目的：研究亚甲基四氢叶酸还原酶（methylenetetrahydrofolate reductase,MTHFR）基因多态性及血浆同型半胱氨酸水平与2型糖尿病视网膜病的关系。方法：应用聚合酶链反应－限制性内切酶片段长度多态性技术检测208例2型糖尿病患者（其中110例伴视网膜病）及57名正常对照的MTHFR C677T基因型，采用高效液相色谱法测定血浆同型半胱氨酸水平。结果：糖尿病视网膜病组MTHFR基因TT纯合基因型、CT杂合基因型及T等位基因频率（分别为28．18％、41．82％、49．09％）均明显高于糖尿病不伴视网膜病组（分别为18．37％、29．59％、33．16％）及正常对照组（分别为17．54％、28．07％、31．58％），基因型和等位基因频率分布差异均有显著性（P＜0．01），而MTHFR基因多态性在糖尿病不伴视网膜病组与正常对照组之间差异无显著性（P＞0．05），T等位基因与糖尿病视网膜病的发生密切相关（OR＝1．94，95％CI；1．31－2．88）。糖尿病视网膜病组、糖尿病不伴视网膜病组及正常对照组中，MTHFR基因有C677T突变者血浆同型半胱氨酸水平均显著高于无基因突变者。结论：MTHFR基因C677T位碱基突变致血浆同型半胱氨酸水平升高可能是糖尿病视网膜病发病的重要遗传因素。     

Learning ensemble classifiers for diabetic retinopathy assessment

Diabetic retinopathy is one of the most common comorbidities of diabetes. Unfortunately, the recommended annual screening of the eye fundus of diabetic patients is too resource-consuming. Therefore, it is necessary to develop tools that may help doctors to determine the risk of each patient to attain this condition, so that patients with a low risk may be screened less frequently and the use of resources can be improved. This paper explores the use of two kinds of ensemble classifiers learned from data: fuzzy random forest and dominance-based rough set balanced rule ensemble. These classifiers use a small set of attributes which represent main risk factors to determine whether a patient is in risk of developing diabetic retinopathy. The levels of specificity and sensitivity obtained in the presented study are over 80%. This study is thus a first successful step towards the construction of a personalized decision support system that could help physicians in daily clinical practice.     

基于Gabor滤波器的糖尿病视网膜新生血管检测

目的:采用Gabor滤波器实现眼底图像中新生血管检测,帮助医生准确确定糖尿病视网膜病变的分期。 方法:对眼底图像进行预处理,并使用不同尺度参数和方向参数Gabor滤波器作用于预处理图像,并在尺度参数确定的情况下取各方向输出结果的最大值作为最后Gabor滤波器的输出。 结果:对比分析不同尺度参数的Gabor滤波器的结果,发现小尺度参数的Gabor滤波器在新生血管部分具有较强的输出。 结论:本研究提出的Gabor滤波器可以很好地区分眼底图像中正常血管与新生血管结构。     

Telemedicine screening of diabetic retinopathy using a hand-held fundus camera.

The objective was to evaluate digital images of the retina from a handheld fundus camera (Nidek NM-100) for suitability in telemedicine screening of diabetic retinopathy. A handheld fundus camera (Nidek) and a standard fundus camera (Zeiss) were used to photograph 49 eyes from 25 consecutive patients attending our diabetic clinic. One patient had cataracts, making it impossible to get a quality image of one of the eyes (retina). The Nidek images were digitized, compressed, and stored in a Fujix DF-10M digitizer supplied with the camera. The digital images and the photographs were presented separately in a random order to three ophthalmologists. The quality of the images was ranked as good, acceptable or unacceptable for diabetic retinopathy diagnosis. The images were also evaluated for the presence of microaneurysms, blot hemorrhages, exudates, fibrous tissue, previous photocoagulation, and new vessel formation. kappa Values were computed for agreement between the photographs and digital images. Overall agreement between the photographs and digital images was poor (kappa < 0.30). On average, only 24% of the digital images were graded as being good quality and 56% as having an acceptable quality. However, 93% of the photographs were graded as good-quality images for diagnosis. The results indicate that the digital images from the handheld fundus camera may not be suitable for diagnosis of diabetic retinopathy. The images shown on the liquid crystal display (LCD) screen of the camera were of good quality. However, the images produced by the digitizer (Fujix DF-10M) attached to the camera were not as good as the images shown on the LCD screen. A better digitizing system may produce better quality images from the Nidek camera.     

Progression of diabetic retinopathy correlated with muscle perfusion disturbances of the lower limbs,with clinically important diagnostic recommendations

Introduction

The objective of this study was to investigate the relationship between microvascular and macrovascular disturbances in patients with type 2 diabetes, as shown by results of ophthalmological examination and by vascular and perfusion examinations of the lower limbs.

Material and methods

A total of 85 patients with type 2 diabetes and an additional cardiovascular risk factor were enrolled in the study. All patients underwent complex ophthalmological examination, including fundus colour photography and fluorescein angiography, and were divided into two groups: group I with signs of diabetic non-proliferative retinopathy (NPDR), and group II with signs of diabetic proliferative retinopathy (P/PDR). After collection of the general medical history and analysis of medical data, patients underwent vascular and muscle perfusion examination of the lower limbs.

Results

In the P/PDR group, disturbances of lower-limb perfusion were more frequent than in the NPDR group. Analysis of the blood flow and results of lower-limb muscle perfusion for the two groups showed a significant relationship with the severity of microvascular complications observed in examination of the fundus.

Conclusions

Ophthalmological assessment of the progression of diabetic retinopathy in patients with type 2 diabetes is a reliable indicator of the changes in peripheral vessel systems and perfusion defects in the lower limbs.     

Segmentation of the optic disk in color eye fundus images using an adaptive morphological approach

The identification of some important retinal anatomical regions is a prerequisite for the computer aided diagnosis of several retinal diseases. In this paper, we propose a new adaptive method for the automatic segmentation of the optic disk in digital color fundus images, using mathematical morphology. The proposed method has been designed to be robust under varying illumination and image acquisition conditions, common in eye fundus imaging. Our experimental results based on two publicly available eye fundus image databases are encouraging, and indicate that our approach potentially can achieve a better performance than other known methods proposed in the literature. Using the DRIVE database (which consists of 40 retinal images), our method achieves a success rate of 100% in the correct location of the optic disk, with 41.47% of mean overlap. In the DIARETDB1 database (which consists of 89 retinal images), the optic disk is correctly located in 97.75% of the images, with a mean overlap of 43.65%.     

Nitric oxide-cGMP and prostacyclin-cAMP pathways in patients with type II diabetes and different types of retinopathy

The aim of this study was to investigate two factors of endothelial dysfunction and their platelet second messengers in patients with type II diabetes and different types of retinopathy. We compared 20 healthy volunteers and 117 patients with type II diabetes (34 with no signs of diabetic retinopathy, 26 with background diabetic retinopathy, 29 with ischemic-proliferative diabetic retinopathy and 28 with edematous diabetic retinopathy). The following parameters were recorded: platelet aggregometry, nitrites, 6-keto-prostaglandin-F(1alpha) and intraplatelet cAMP and cGMP. Platelet aggregation was greater in patients with diabetic retinopathy. The concentration of ADP that produced 50% maximum intensity of aggregation was 1.81 microM in patients without diabetic retinopathy, 0.92 microM in patients with background diabetic retinopathy, 0.85 microM in patients with ischemic-proliferative diabetic retinopathy and 0.44 microM in patients with edematous diabetic retinopathy. The platelets in these patients were more resistant to inhibition by SIN-1 (concentrations of SIN-1 that produced 50% inhibition of maximum intensity of collagen-induced aggregation in the four patient groups: 18.1, 13.6, 16.2 and 33.2 microM, respectively). Nitrite concentration in patients with ischemic-proliferative diabetic retinopathy was one sixth of the value in healthy controls, but there was no significant difference between the control group and patients with edematous diabetic retinopathy. In the latter group, neutrophils increased nitrite production by 68.7 +/- 3%, whereas in patients with ischemic-proliferative diabetic retinopathy, this increase was 18.7 +/- 2.0%. We conclude that nitric oxide production is higher in patients with type II diabetes and edematous retinopathy than in those with ischemic-proliferative retinopathy. This finding, together with the possibly greater production of free radicals, may explain the greater impairment of platelet function in the former patients.     

Identification of the optic nerve head with genetic algorithms

OBJECTIVE: This work proposes creating an automatic system to locate and segment the optic nerve head (ONH) in eye fundus photographic images using genetic algorithms. METHODS AND MATERIAL: Domain knowledge is used to create a set of heuristics that guide the various steps involved in the process. Initially, using an eye fundus colour image as input, a set of hypothesis points was obtained that exhibited geometric properties and intensity levels similar to the ONH contour pixels. Next, a genetic algorithm was used to find an ellipse containing the maximum number of hypothesis points in an offset of its perimeter, considering some constraints. The ellipse thus obtained is the approximation to the ONH. The segmentation method is tested in a sample of 110 eye fundus images, belonging to 55 patients with glaucoma (23.1%) and eye hypertension (76.9%) and random selected from an eye fundus image base belonging to the Ophthalmology Service at Miguel Servet Hospital, Saragossa (Spain). RESULTS AND CONCLUSIONS: The results obtained are competitive with those in the literature. The method's generalization capability is reinforced when it is applied to a different image base from the one used in our study and a discrepancy curve is obtained very similar to the one obtained in our image base. In addition, the robustness of the method proposed can be seen in the high percentage of images obtained with a discrepancy delta<5 (96% and 99% in our and a different image base, respectively). The results also confirm the hypothesis that the ONH contour can be properly approached with a non-deformable ellipse. Another important aspect of the method is that it directly provides the parameters characterising the shape of the papilla: lengths of its major and minor axes, its centre of location and its orientation with regard to the horizontal position.     

MRI retinovascular studies in humans: research in patients with diabetes

PURPOSE: To review existing applications of MRI for detecting blood-retinal barrier (BRB) damage and retinal oxygenation response abnormalities in patients with diabetes and highlight new information available from such applications. METHODS: BRB studies were accomplished using dynamic contrast-enhanced MRI, and the retinal oxygenation response studies were accomplished by monitoring changes in the MRI signal associated with hyperoxic provocation. Participants were patients with diabetes and macular edema, with either no detectable or mild to moderate background retinopathy, as well as non-diabetic individuals of similar age (controls). Single-slice FLASH images were obtained using a Siemens Sonata, 1.5 T together with a Siemens 'Loop Small' surface coil fixed in place over the eye. Time-dependent changes in image contrast in the pre-retinal vitreous were quantified, and differences between patients and controls were assessed statistically. RESULTS: The BRB breakdown studies showed a significant difference in the temporal evolution of the MRI signal enhancement post-contrast injection between the controls and the patients with diabetic macular edema. The retinal oxygenation studies revealed a supernormal oxygenation response in the pre-retinal vitreous in patients with diabetes who had no evidence of retinopathy, as well as in patients with background diabetic retinopathy. A nasal-temporal asymmetry in the evolution of retinal oxygenation response was found in patients with diabetes that was not present in healthy subjects. CONCLUSIONS: These studies show that subtle differences in retinovascular function between patients with diabetes and non-diabetic individuals, including changes that occur in advance of the clinical appearance of diabetic retinopathy, can be detected with MRI. These results, together with previous extensive preclinical data, establish MRI as a powerful non-invasive method for measuring spatial and temporal changes in the same key retinovascular metrics in both animals and humans. Wide application of these techniques for diagnosis and evaluation of treatment efficacy in a variety of human retinopathies, including diabetic retinopathy, is expected.