Expert Ophthalmologists Research Articles

Diagnosis of diabetic retinopathy by employing image processing technique to detect exudates in retinal images

Diabetic retinopathy (DR) is a microvascular complication of long-term diabetes and it is the major cause of visual impairment because of changes in blood vessels of the retina. Major vision loss because of DR is highly preventable with regular screening and timely intervention at the earlier stages. The presence of exudates is one of the primitive signs of DR and the detection of these exudates is the first step in automated screening for DR. Hence, exudates detection becomes a significant diagnostic task, in which digital retinal imaging plays a vital role. In this study, the authors propose an algorithm to detect the presence of exudates automatically and this helps the ophthalmologists in the diagnosis and follow-up of DR. Exudates are normally detected by their high grey-level variations and they have used an artificial neural network to perform this task by applying colour, size, shape and texture as the features. The performance of the authors algorithm has been prospectively tested by using DIARETDB1 database and evaluated by comparing the results with the ground-truth images annotated by expert ophthalmologists. They have obtained illustrative results of mean sensitivity 96.3%, mean specificity 99.8%, using lesion-based evaluation criterion and achieved a classification accuracy of 99.7%.

Assessment of vessel tortuosity in retinal images of preterm infants.

Diagnosis of plus disease is crucial for timely treatment and management of retinopathy of prematurity. An indicator of the presence of plus disease is an increase in the tortuosity of blood vessels in the retina. In this work, we propose a new angle-variation-based measure for quantification of tortuosity in retinal fundus images of preterm infants. The methods include the use of Gabor filters to detect vessels as well as to obtain their orientation at each pixel. Morphological image processing methods are used to obtain a skeleton image of the vessels for measurement of tortuosity. Out of 11 vessel segments, marked by an expert ophthalmologist as showing high levels of tortuosity due to plus disease, all were correctly identified using the proposed methods.

Computer-aided diagnosis software for hypertensive risk determination through fundus image processing.

The goal of the software proposed in this paper is to assist ophthalmologists in diagnosis and disease prevention, helping them to determine cardiovascular risk or other diseases where the vessels can be altered, as well as to monitor the pathology progression and response to different treatments. The performance of the tool has been evaluated by means of a double-blind study where its sensitivity, specificity, and reproducibility to discriminate between health fundus (without cardiovascular risk) and hypertensive patients has been calculated in contrast to an expert ophthalmologist opinion obtained through a visual inspection of the fundus image. An improvement of almost 20% has been achieved comparing the system results with the clinical visual classification.

THU0134 Ocular Toxicity Induced by Antimalarial Medications; Necessity for Regular Screening

BackgroundSince the early 50's, chloroquine (CQ) and hydroxychloroquine (HCQ) have been widely used for the treatment of rheumatoid arthritis and other similar inflammatory diseases. So far these medications are the...

IDENTIFICATION OF EXUDATES USING FUZZY MATHEMATICAL MORPHOLOGY

Diabetic Retinopathy is the damage to the retina ca used by complication and the most common cause of blindness in Thailand. Retinal image is essential f or expert ophthalmologists to diagnose diseases. Several of method can achieve good performance on retinal feature are clearly visible. Unfortunately, the color retinal image in Thailand are low-resolution images. The existing method cannot identified lowresolution image. Therefore, this study is part of a larger effort to develop a new method for identification of exudates in low-resolution retina l image. In this study a fuzzy mathematical morphol ogy based on fuzzy logical operator and mathematical mo rphology method is presented. The color retinal image are segmented by using fuzzy logical operator following key preprocessing step, i.e., color normalization, contrast enhancement, noise removal and color space selection. Afterward, a segmentatio n using mathematical morphology method was applied in this step. This enables its difference in our methods compared to other approach and the methods can achieve good performance even on lowresolution retinal images. Respect to the experimental results, the results ob tained with fuzzy mathematical morphology better than the ones obtained with the fuzzy logical operator only method.

Automatic Detection of Optic Disc in Digital Retinal Images

On the research work leading to automatic detection of optic disc from retinal images is very essential and crucial for expert ophthalmologists to diagnose diseases. Many of techniques can achieve good performance on retinal feature that is clearly visible. Unfortunately, it is a normal situation that the color retinal images in Thailand are poor-quality images. The existing algorithm cannot detected poor-quality images. Therefore, this study is a part of larger efforts to develop a novel method for detection of optic disc in poorquality retinal images. A novel method is presented towards the development for detection of optic disc in poor-quality retinal images. The digital retinal images are detected by using morphological method and Otsu’s algorithm after the key preprocessing steps, i.e., color normalization, contrast enhancement and noise removal. This enables the difference in the proposed method compared to other approaches and the algorithm can achieve good performance even on poor-quality retinal images. The proposed method was evaluated using the local dataset and the publicly available of the STARE project’s dataset. The optic disc was detected correctly in 91.35% using the STARE dataset and 97.61% using the local dataset. This system intends to help expert ophthalmologists in screening process to detect of optic disc faster and more easily.

Automatic Evaluation of Plus Disease in Retinopathy of Prematurity

Abnormal dilation and tortuosity of retinal blood vessels are the primary signs of plus disease in retinopathy of prematurity. Timely prognosis could help reduce the delay in treatment and the risk of retinal detachment. Our objectives is to determine whether tortuosity and dilation sufficient for plus disease could be assessed most accurately by considering only arterioles, venules, or both. Tortuosity estimation and width measurement is done using previously proposed methods. Image preprocessing is applied before the two features namely, tortuosity and width of blood vessels are estimated to supply as input parameters for classification using K-means clustering technique. The results are validated by comparing with expert ophthalmologists ground truths. Performance is evaluated based on measures as sensitivity, specificity, predictive values and accuracy. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy values obtained when considering both the arteriolar tortuosity and venous dilation are 85.86%, 90.74%, 88.76%, 88.28% and 88.50% respectively.

Automatic Optic Disc Boundary Extraction from Color Fundus Images

Efficient optic disc segmentation is an important task in automated retinal screening. For the same reason optic disc detection is fundamental for medical references and is important for the retinal image analysis application. The most difficult problem of optic disc extraction is to locate the region of interest. More over it is a time consuming task. This paper tries to overcome this barrier by presenting an automated method for optic disc boundary extraction using Fuzzy C Means combined with thresholding. The discs determined by the new method agree relatively well with those determined by the experts. The present method has been validated on a data set of 110 colour fundus images from DRION database, and has obtained promising results. The performance of the system is evaluated using the difference in horizontal and vertical diameters of the obtained disc boundary and that of the ground truth obtained from two expert ophthalmologists. For the 25 test images selected from the 110 colour fundus images, the Pearson correlation of the ground truth diameters with the detected diameters by the new method are 0.946 and 0.958 and, 0.94 and 0.974 respectively. From the scatter plot, it is shown that the ground truth and detected diameters have a high positive correlation. This computerized analysis of optic disc is very useful for the diagnosis of retinal diseases.

Comparison and supervised learning of segmentation methods dedicated to specular microscope images of corneal endothelium.

The cornea is the front of the eye. Its inner cell layer, called the endothelium, is important because it is closely related to the light transparency of the cornea. An in vivo observation of this layer is performed by using specular microscopy to evaluate the health of the cells: a high spatial density will result in a good transparency. Thus, the main criterion required by ophthalmologists is the cell density of the cornea endothelium, mainly obtained by an image segmentation process. Different methods can perform the image segmentation of these cells, and the three most performing methods are studied here. The question for the ophthalmologists is how to choose the best algorithm and to obtain the best possible results with it. This paper presents a methodology to compare these algorithms together. Moreover, by the way of geometric dissimilarity criteria, the algorithms are tuned up, and the best parameter values are thus proposed to the expert ophthalmologists.

A teleconsultation network improves the efficacy of anti-VEGF therapy in retinal diseases

We investigated the care of patients with age-related macular degeneration (AMD) managed via a physician-to-physician teleconsultation network for ophthalmology. Eleven groups of ophthalmologists took part in the study. The groups were located in 10 cities across Italy. Each group was based on a Retina Centre located at a university or hospital, with one or two expert ophthalmologists (20 expert ophthalmologists in total). In each region containing a Retina Centre, 6-10 general ophthalmologists (94 ophthalmologists in total) referred patients via the network for a period of three months between June 2011 and December 2012. An automatic grading system quantified the risk of disease progression, and a remote booking system allowed the referring ophthalmologist to make appointments directly with the appropriate Retina Centre. There were 360 network patients and 318 control patients (consecutive patients undergoing usual care during the previous three months). The time delay before therapy was significantly shorter in the network patients (mean 5.5 days) compared with the usual care patients (mean 28.7 days; P < 0.0001). There was a significant improvement in visual acuity in the network patients after treatment (first visit = 0.29 logMAR; after treatment = 0.22 logMAR; P < 0.05). In contrast, there was no improvement in the usual care patients (first visit = 0.29 logMAR; after treatment = 0.27 logMAR; P > 0.05). The telemedicine network allows regional ophthalmologists to quantify the risk of disease progression, and to send patients to a Retina Centre quickly and easily, when required.

Computerized Exudate Detection in Fundus Images Using Statistical Feature based Fuzzy C-mean Clustering

Diabetic retinopathy(DR) is considered as the root cause of vision loss for diabetic patients .One of the greatest concern and immediate challenges to the current health care is the severe progression of diabetes. Diabetic retinopathy is an eye disease and appearance of hard exudates is one of its earliest signs. The accuracy of the automated disease identification techniques should be high .Besides being accurate; the techniques need to possess a quick convergence rate enabling them to be suitable for real-time applications. In order to lessen the cost of these screenings, modern image processing techniques are used to voluntarily detect the existence of abnormalities in the retinal images acquired during the screenings. Hard and Soft exudates (Cotton wool spots) are a major indicator of diabetic retinopathy that can possibly be quantified automatically. Automatic computerized screening should facilitate screening process, reduce inspection time and increase accuracy. In this paper an automated method to detect exudates from low-contrast digital images of retinopathy patients with non-dilated pupils using feature based Fuzzy C-Means (FCM) clustering technique with a combination of morphology techniques and pre-processing to improve the robustness of blood vessel and optic disk detection. Pre-processing involve colour, normalization ,contrast enhancement and brightness preserving dynamic fuzzy histogram equalization whereas addressable features are converted colour spaces, intensity, standard deviation, edge strength, size, colour, texture and entropy. The detection accuracy is calculated with comparison to expert ophthalmologists’ hand-drawn ground-truths and the results are comparatively analysed.

Fluorescein Angiography–Based Diagnosis for Retinopathy of Prematurity: Expert-non Expert Comparison

To evaluate accuracy and inter-rater reliability of RetCam fundus images and digital camera fluorangioscopic images in acute retinopathy of prematurity (ROP) by comparing diagnoses given by trainee ophthalmologists with those provided by expert ophthalmologists. This is a multicenter retrospective observational study of diagnostic data from 48 eyes of 24 premature infants with classical ROP, stage II, as evaluated by RetCam 3 and fluorescein angiography (FA). Average gestational age was 25.4 weeks, average weight 804.7 g. A staging grid (with ocular fundus divided into 3 concentric zones) and 24 15° sectors centered around the optic papilla were superimposed on 360° retina photomontages (Photoshop) made from RetCam and FA images. Non expert vs expert diagnosis agreement was measured for each sector by means of Cohen kappa (Fleiss, 1981). A high degree of concordance was found. Inter-rater agreement between expert and non expert interpretations of retinal photomontages was greater for fluorangiographic images than for RetCam images, with κ = 0.61-1 for 120/152 (78.9%) sectors examined on the RetCam images and κ = 0.61-1 for 168/198 (84.8%) sectors examined on the FA images. The FA images appear to be easier to interpret than RetCam images, both by expert and non expert ophthalmologists. The results confirm that FA is a good examination technique with a high degree of reliability, even where trainee practitioners are involved. This suggests that retinopathy management can be improved by entrusting diagnostic responsibilities to trainee ophthalmologists, in order to extend access to correct diagnosis, recognition of threshold lesions, and prompt treatment.

Specialist Nurse Screening for Retinopathy of Prematurity-A Pilot Study.

To compare the accuracy of retinopathy of prematurity (ROP) screening between nurse specialists and an expert pediatric ophthalmologist. A comparative case series of ROP screening in a level 2 neonatal intensive care unit setting where there is a higher threshold for accepting very premature or unwell neonates. Trained specialist ROP nurses acquired wide-field digital images, graded ROP, and proposed a follow-up plan. This was compared with the findings by an ROP expert ophthalmologist. Outcomes include sensitivity, specificity, positive predictive value, and negative predictive value of ROP grading by trained ROP nurses. Mean gestational age was 28.6 weeks, and mean birth weight was 1184 g of the 64 consecutive neonates included. A total 316 eye screens was performed. Image acquisition, grading, and a management plan by the ROP nurse were possible in all screens. In right eyes, the presence of any ROP (stage >0) was 15%. Sensitivity, specificity, positive predictive value, and negative predictive value of ROP grading were 91.7% (95% CI, 73%-99%), 80.6% (95% CI, 72.9%-86.9%), 45.8% (95% CI, 31.4%-60.80%), and 98.2% (95% CI, 93.6%-99.8%), respectively. Agreement on the management plan occurred in 84.8% of cases. In virtually all circumstances of disagreement, ROP nurses exaggerated the ROP grading present and/or recommended a repeat screen when discharge from service was more appropriate. Our preliminary findings demonstrated good agreement between ROP nurses and the ROP expert ophthalmologist. Further research in expanding the role of utilizing nonphysician health workers in ROP screening is suggested.

Quantification and classification of retinal vessel tortuosity

The clinical recognition of abnormal retinal tortuosity enables the diagnosis of many diseases. Tortuosity is often interpreted as points of high curvature of the blood vessel along certain segments. Quantitative measures proposed so far depend on or are functions of the curvature of the vessel axis. In this paper, we propose a parallel algorithm to quantify retinal vessel tortuosity using a robust metric based on the curvature calculated from an improved chain code algorithm. We suggest that the tortuosity evaluation depends not only on the accuracy of curvature determination, but primarily on the precise determination of the region of support. The region of support, and hence the corresponding scale, was optimally selected from a quantitative experiment where it was varied from a vessel contour of two to ten pixels, before computing the curvature for each proposed metric. Scale factor optimization was based on the classification accuracy of the classifiers used, which was calculated by comparing the estimated results with ground truths from expert ophthalmologists for the integrated proposed index. We demonstrate the authenticity of the proposed metric as an indicator of changes in morphology using both simulated curves and actual vessels. The performance of each classifier is evaluated based on sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and positive likelihood ratio. Our method is effective at evaluating the range of clinically relevant patterns of abnormality such as those in retinopathy of prematurity. While all the proposed metrics are sensitive to curved or kinked vessels, the integrated proposed index achieves the best sensitivity and classification rate of 97.8% and 93.6%, respectively, on 45 infant retinal images.

Drusen Detection and Quantification for Early Identification of Age Related Macular Degeneration using Color Fundus Imaging

Objective: To develop a method for detecting drusen and quantifying drusen size in macular region from standard color retinal images for early diagnosis of Age related Macular Degeneration (AMD). Materials and methods: Color retinal images were used which were captured by Canon D60 non-mydriatic camera for genetic and epidemiology study. Local intensity distribution, adaptive intensity thresholding and edge information were used to detect potential drusen areas. For validation, we considered 50 images with various types of drusen. For the drusen area segmentation accuracy (DAA), 12 images were selected, and an expert grader marked the drusen regions in pixel level. We then quantified the areas and compute the sensitivity and specificity by comparing the drusen detected output images with the hand-labeled ground truth (GT) images. Results: The proposed method detected the presence of any drusen with 100% accuracy (50/50 images). For drusen detection accuracy (pixel level), mean sensitivity and specificity values of 74.94% and 81.17%, respectively. For drusen subtypes we achieved 79.59% accuracy in intermediate drusen and 82.14% in soft drusen which is a highly significant result for early and intermediate AMD detection. Conclusion: In this study, we applied a novel automated method for drusen detection and quantification which is ready to be used in initial screening of early stage of AMD and drusen area changes i.e., AMD progression. The method will also be highly suitable for telemedicine platforms in ophthalmology for selecting patient from rural areas using fundus imaging - for refereeing to an expert ophthalmologist.

Development of community based curriculum on ophthalmology for under graduate medical course in Bangladesh

The curriculum represents the expression of educational ideas in practice. Ophthalmic education is the corner stone to improve eye care globally. Curriculum needs continuous modification varying in different geographic locations. Though 90% of common conditions are either preventable or curable but emphasis on the common conditions is inadequate. This is a stepwise descriptive study aiming to develop a community based ophthalmology curriculum for undergraduate medical course in Bangladesh conducted during March 2007 to February 2008 at UniSA School of Public Health and Life Sciences, University of South Asia, Banani, Dhaka. Delphi technique, a modified qualitative method was used to accumulate data and reaching a consensus opinion for developing the curriculum. Study approach includes two iterative rounds and finally a workshop. Iteration of round-I was "What are the eye diseases with overall knowledge of their management one MBBS physician should acquire"; followed by a list of eye diseases and topics for expert opinion. The response was collated. Iteration round-II was "How much a MBBS student should have percentage of knowledge, attitude and skills on each topic while being taught". The response was collated and presented to panel of expert ophthalmologists for discussion and validation. In the round-I Delphi, 400 (62%) out to total 641 ophthalmologist were randomly selected dividing in categories (62% in each) of Professor-22, Associate Professor-12, Assistant Professor-26, Consultant-27, ophthalmologists working in NGO-56 and ophthalmologists in private sector-257. Sixty (15%) responded with opinion. In the round-II, 200 (31%) including 60 of round-I, selected randomly but proportionately as before. Forty five (22.5%) responded with opinion. Result collated. The results and opinion of respondents were presented at a workshop attended by 24 (80%), out of 30 invited expert ophthalmic specialists for discussion, criticism, opinion, addition, modification and finally for validation. On the basis of the opinion of the respondents, reviewing literature, analyzing the ocular disease pattern in Bangladesh and also analyzing the present ophthalmology curriculum, a community and need based ophthalmology curriculum for undergraduate medical course in Bangladesh was developed. This research would help developing community and need based ophthalmology curriculum for undergraduate medical course in Bangladesh.

Semi-Automated Computer Analysis of Retinal Vessels in Premature Infants

Almost all ocular and systemic diseases affect blood vessel attributes (tortuosity, length, width, and curvature). Quantitative measurements of these attributes could thus provide useful tool for diagnosing the severity of several diseases. However, it is still unclear how best to represent the attribute values of multiple vessels in a single image. Graphical user interface (GUI) is a promising step towards the development of a semi-automated computer assisted tool. The objective of this study is to develop a GUI for effective observation and robust retinal blood vessels analysis by ophthalmologists and to comprehend the distribution of vessels attributes. Blood vessels from 45 digital fundus images of infant retina are extracted, its centerline is delineated and tortuosity is analyzed from different putative and proposed techniques to provide reliable and comprehensive information for the retinal vasculature. K means clustering technique is used for classification analysis of different tortuosity metrics and its performance is evaluated based on sensitivity, specificity, and accuracy. The results are validated by comparing with expert ophthalmologists’ ground truths. Among the various proposed tortuosity metrics, one of our tortuosity indexes attains the highest classification accuracy of 91.42% with sensitivity and specificity of 86.36% and 97.82% respectively.

Automatic Tortuosity Classification Using Machine Learning Approach

Retinopathy of Prematurity (ROP) is a vital cause of vision loss in premature infants, but early detection of its symptoms enables timely treatment and prevents blindness. Tortuosity is the major indicator of ROP that can potentially be automatically quantified. In this paper, which focuses on automatic tortuosity quantification and classification in images from infants at risk of ROP, we present a series of experiments on preprocessing, feature extraction, image feature selection and classification using nearest neighbor classifier. Fisher linear Discriminant analysis is used as a feature selection algorithm. We observe that the best feature set is a combination of two features: tortuosity as estimated based on combination of curvature of improved chain code and number of inflections and tortuosity as measured by inflection count metric. Accuracy, sensitivity and specificity are used as performance measures for the classifier. The results are validated against the judgments of expert ophthalmologists. The overall accuracy, sensitivity and specificity achieved on the best feature set are 95%, 95.65% and 96.74% respectively.

Evolutionary algorithm based classifier parameter tuning for automatic diabetic retinopathy grading: A hybrid feature extraction approach

Human eye is one of the most sophisticated organ, with retina, pupil, iris cornea, lens and optic nerve. Automatic retinal image analysis is emerging as an important screening tool for early detection of eye diseases. Uncontrolled diabetes retinopathy (DR) and glaucoma may lead to blindness. DR is caused by damage to the small blood vessels of the retina in the posterior part of the eye of the diabetic patient. The main stages of DR are non-proliferate diabetes retinopathy (NPDR) and proliferate diabetes retinopathy (PDR). The retinal fundus photographs are widely used in the diagnosis and treatment of various eye diseases in clinics. It is also one of the main resources used for mass screening of DR. We present an automatic screening system for the detection of normal and DR stages (NPDR and PDR). The proposed systems involves processing of fundus images for extraction of abnormal signs, such as area of hard exudates, area of blood vessels, bifurcation points, texture and entropies. Our protocol uses total of 156 subjects consisting of two stages of DR and normal. In this work, we have fed thirteen statistically significant (p<0.0001) features for Probabilistic Neural Network (PNN), Decision Tree (DT) C4.5, and Support Vector Machine (SVM) to select the best classifier. The best model parameter (σ) for which the PNN classifier performed best was identified using global optimization techniques such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). We demonstrated an average classification accuracy of 96.15%, sensitivity of 96.27% and specificity of 96.08% for σ=0.0104 using threefold cross validation using PNN classifier. The computer-aided diagnosis (CAD) results were validated by comparing with expert ophthalmologists. The proposed automated system can aid clinicians to make a faster DR diagnosis during the mass screening of normal/DR images.

Parabolic Modeling of the Major Temporal Arcade in Retinal Fundus Images

Monitoring measurements of the openness of the major temporal arcade (MTA) and how they change over time could facilitate improved diagnosis and optimized treatment of retinopathy. We propose methods for the detection, modeling, and measurement of the openness of the MTA, including Gabor filters to detect retinal vessels and the Hough transform to detect and parameterize parabolic forms. Results obtained with 40 images of the Digital Retinal Images for Vessel Extraction database, compared with traces of the MTA drawn by an expert ophthalmologist and a retinal specialist, indicate a low mean distance to the closest point of about 12 pixels (0.24 mm). The proposed methods should facilitate quantitative analysis of the MTA and overcome limitations associated with subjective manual analysis.