Color Retinal Images Research Articles

A cloud-based system for automatic glaucoma screening.

In recent years, there has been increasing interest in the use of automatic computer-based systems for the detection of eye diseases including glaucoma. However, these systems are usually standalone software with basic functions only, limiting their usage in a large scale. In this paper, we introduce an online cloud-based system for automatic glaucoma screening through the use of medical image-based pattern classification technologies. It is designed in a hybrid cloud pattern to offer both accessibility and enhanced security. Raw data including patient's medical condition and fundus image, and resultant medical reports are collected and distributed through the public cloud tier. In the private cloud tier, automatic analysis and assessment of colour retinal fundus images are performed. The ubiquitous anywhere access nature of the system through the cloud platform facilitates a more efficient and cost-effective means of glaucoma screening, allowing the disease to be detected earlier and enabling early intervention for more efficient intervention and disease management.

Automatic Detection of Exudates from Digital Color Fundus Images

retinopathy is a widespread disease that may cause blindness. Early diagnosis and treatment will reduce its side effects and protect the eye. In this paper, a new algorithm for exudates detection is proposed. In the preprocessing step, the green channel of the color image is used, and then median filter followed by Contrast Limited Adaptive Histogram Equalization (CLAHE) is applied. The K-means clustering technique is used to select exudates objects. Optic disc is localized using maximum entropy filter and morphological closing. It is demonstrated that combining the K-means with CLAHE of the median filtered image results in 99.39% correct exudates. Experimental results show a reliable and accurate method for segmenting exudates from color retinal images. Performance of the proposed method is evaluated using a set of 52 images from a publicly available dataset STARE. KeywordsRetinopathy, Exudates detection, Entropy filter, K- means clustering

, ,

Macular edema is an advance stage of diabetic retinopathy which affects central vision of diabetes patients. The main cause of edema is the appearance of exudates near or on macular region in human retina. If the exudates are present in the macular region of retina, it will lead to diabetic macular edema. Early detection of macular edema in diabetic patients paves a path for prevention from blindness. The automatic system for early detection of diabetic macular edema should identify all possible exudates present on the surface of retina. In the proposed work, a computer added diagnosis system is design for the identification of the exudates in color retinal fundus images. The system consists of three stages; candidate exudates detection, feature extraction and classification. The system is designed with (i) background estimation, morphological reconstruction, normalization for candidate exudates detection (ii) Gray level co-occurrence matrix for feature extraction and (iii) support vector machine for classification. The classifier classifies in between the region of exudates and non-exudates. The system performance is evaluated in terms of the parameters such as sensitivity, specificity, mathews correlation coefficient, positive predicative value, and accuracy whose values are 88.23%, 100%, 88.23%, 100%, 93.75% respectively.

Drusen Quantification for Early Identification of Age Related Macular Degeneration

Age-related macular degeneration (AMD) is a degenerative disorder in people of age 50 and above, in developed nations, characterized on grading of color fundus images by the presence of pathologies such as drusen in macular area. Currently, there is no treatment which can cure irreversible blindness due to age-related macular degeneration. Therefore, the only feasible option is to prevent the incidence of age-related macular degeneration and avoid this unnecessary vision loss. This paper presents an automated method for early diagnosis of AMD by quantifying drusen on the basis of its size, number and area in macular region from standard color retinal images. Previously used methods, generating unsatisfactory results in some cases, are time consuming, complex and prone to error. Therefore, this paper provides a simple drusen detection and quantification method to detect the exact number of drusen , area and size as well as classify drusen into small, intermediate and soft or large which will further help in initial screening of early stage of age-related macular degeneration and its progression i.e. change in drusen area. The proposed method achieved 93.2% accuracy for drusen detection and 91.8% accuracy in small drusen, 98.66% in intermediate drusen and 92.91% in soft drusen quantification in order to grade the severity of AMD which outwits the other methods.

Automatic detection of microaneurysms using microstructure and wavelet methods

Retinal microaneurysm is one of the earliest signs in diabetic retinopathy diagnosis. This paper has developed an approach to automate the detection of microaneurysms using wavelet-based Gaussian mixture model and microstructure texture feature extraction. First, the green channel of the colour retinal fundus image is extracted and pre-processed using various enhancement techniques such as bottom-hat filtering and gamma correction. Second, microstructures are extracted as Gaussian profiles in wavelet domain using the three-level generative model. Multi-scale Gaussian kernels are obtained and histogram-based features are extracted from the best kernel. Using the Markov Chain Monte Carlo method, microaneurysms are classified using the optimal feature set. The proposed approach is experimented with DIARETDB0 and DIARETDB1 datasets using a classifier based on multilayer perceptron procedure. For DIARETDB0 dataset, the proposed algorithm obtains the results with a sensitivity of 98.32 and specificity of 97.59. In the case of DIARETDB1 dataset, the sensitivity and specificity of 98.91 and 97.65 have been achieved. The accuracies achieved by the proposed algorithm are 97.86 and 98.33 using DIARETDB0 and DIARETDB1 datasets respectively. Based on ground truth validation, good segmentation results are achieved when compared to existing algorithms such as local relative entropy-based thresholding, inverse adaptive surface thresholding, inverse segmentation method, and dark object segmentation.

Registering the Retinal Vasculature in Gray-scale and Color Digital Fundus Images

Some of the most common blinding conditions are caused by choroidal neovascularization (CNV). The relevant conditions include diabetic retinopathy and age-related macular degeneration. At present, the only proven modality of effective treatment is the application of laser energy to the CNV to cauterize the vessels. The key to effective and lasting treatment is the identification of the full extent of the CNV, cauterization of the CNV completely by accurately aiming an appropriate amount of optical energy while ensuring that healthy tissue is not cauterized. Extraction techniques must be developed to discern the retinal blood vessels tree and determine the positions of laser shots in a reference frame. After extracting the blood vessels tree & determining the locations of the laser shots, those locations will be saved in a database. Due to the eye movement the locations of the laser shots will differ. So, registration of the translation between the reference & the sub-sequent frame in different types of retinal images is required. This paper presents a comparison of two different methods to register blood vessels, which are a prominent retinal structure, in both gray-scale and color retinal images. The blood vessel registration is composed of two algorithms, i.e., registering the blood vessel using Normalized 2-D Correlation algorithm and Minimum Error Function algorithm. Results on various retinal images verify the effectiveness of the proposed methods.

Robust point matching method for multimodal retinal image registration

In this paper, motivated by the problem of multimodal retinal image registration, we introduce and improve the robust registration framework based on partial intensity invariant feature descriptor (PIIFD), then present a registration framework based on speed up robust feature (SURF) detector, PIIFD and robust point matching, called SURF–PIIFD–RPM. Existing retinal image registration algorithms are unadaptable to any case, such as complex multimodal images, poor quality, and nonvascular images. Harris-PIIFD framework usually fails in correctly aligning color retinal images with other modalities when faced large content changes. Our proposed registration framework mainly solves the problem robustly. Firstly, SURF detector is useful to extract more repeatable and scale-invariant interest points than Harris. Secondly, a single Gaussian robust point matching model is based on the kernel method of reproducing kernel Hilbert space to estimate mapping function in the presence of outliers. Most importantly, our improved registration framework performs well even when confronted a large number of outliers in the initial correspondence set. Finally, multiple experiments on our 142 multimodal retinal image pairs demonstrate that our SURF–PIIFD–RPM outperforms existing algorithms, and it is quite robust to outliers.

A comparative study on preprocessing techniques in diabetic retinopathy retinal images: Illumination correction and contrast enhancement

To investigate the effect of preprocessing techniques including contrast enhancement and illumination correction on retinal image quality, a comparative study was carried out. We studied and implemented a few illumination correction and contrast enhancement techniques on color retinal images to find out the best technique for optimum image enhancement. To compare and choose the best illumination correction technique we analyzed the corrected red and green components of color retinal images statistically and visually. The two contrast enhancement techniques were analyzed using a vessel segmentation algorithm by calculating the sensitivity and specificity. The statistical evaluation of the illumination correction techniques were carried out by calculating the coefficients of variation. The dividing method using the median filter to estimate background illumination showed the lowest Coefficients of variations in the red component. The quotient and homomorphic filtering methods after the dividing method presented good results based on their low Coefficients of variations. The contrast limited adaptive histogram equalization increased the sensitivity of the vessel segmentation algorithm up to 5% in the same amount of accuracy. The contrast limited adaptive histogram equalization technique has a higher sensitivity than the polynomial transformation operator as a contrast enhancement technique for vessel segmentation. Three techniques including the dividing method using the median filter to estimate background, quotient based and homomorphic filtering were found as the effective illumination correction techniques based on a statistical evaluation. Applying the local contrast enhancement technique, such as CLAHE, for fundus images presented good potentials in enhancing the vasculature segmentation.

Exudate detection in color retinal images for mass screening of diabetic retinopathy.

The automatic detection of exudates in color eye fundus images is an important task in applications such as diabetic retinopathy screening. The presented work has been undertaken in the framework of the TeleOphta project, whose main objective is to automatically detect normal exams in a tele-ophthalmology network, thus reducing the burden on the readers. A new clinical database, e-ophtha EX, containing precisely manually contoured exudates, is introduced. As opposed to previously available databases, e-ophtha EX is very heterogeneous. It contains images gathered within the OPHDIAT telemedicine network for diabetic retinopathy screening. Image definition, quality, as well as patients condition or the retinograph used for the acquisition, for example, are subject to important changes between different examinations. The proposed exudate detection method has been designed for this complex situation. We propose new preprocessing methods, which perform not only normalization and denoising tasks, but also detect reflections and artifacts in the image. A new candidates segmentation method, based on mathematical morphology, is proposed. These candidates are characterized using classical features, but also novel contextual features. Finally, a random forest algorithm is used to detect the exudates among the candidates. The method has been validated on the e-ophtha EX database, obtaining an AUC of 0.95. It has been also validated on other databases, obtaining an AUC between 0.93 and 0.95, outperforming state-of-the-art methods.

An effective automated system for grading severity of retinal arteriovenous nicking in colour retinal images.

Retinal arteriovenous (AV) nicking is a precursor for hypertension, stroke and other cardiovascular diseases. In this paper, an effective method is proposed for the analysis of retinal venular widths to automatically classify the severity level of AV nicking. We use combination of intensity and edge information of the vein to compute its widths. The widths at various sections of the vein near the crossover point are then utilized to train a random forest classifier to classify the severity of AV nicking. We analyzed 47 color retinal images obtained from two population based studies for quantitative evaluation of the proposed method. We compare the detection accuracy of our method with a recently published four class AV nicking classification method. Our proposed method shows 64.51% classification accuracy in-contrast to the reported classification accuracy of 49.46% by the state of the art method.

Segmenting the Optic Disc in Retinal Images using Thresholding

Segmentation of an optic disk location is a very essential task for retinal image analysis. Most of the pervious method uses the principal component analysis method, which are computationally complex. This paper presents a simple and semi automatic approach for segmentation of the optic disc in retinal images using thresholding and boundary extraction. First, the true colour Retinal images are converted to gray image and then image is enhanced using histogram equalization. This improves the efficiency of the conventional thresholding method. The proposed method of segmentation utilizes gray level thresholding for segmentation and then morphological operators are used for extracting the optical disk. The method is tested on the different optical disc images. The method is less complex and efficiently recognizes the boundary of the optic disc and also evaluates the optical canter. The optical canters are tabulated for original and corrupted images.

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.

Computationally efficient optic nerve head detection in retinal fundus images

This paper presents a computationally efficient method for detection of optic nerve head in both color and fluorescein angiography retinal fundus images. It involves Radon transformation of multi-overlapping windows within an optimization framework in order to achieve computational efficiency as well as high detection rates in the presence of various structural, color, and intensity variations in such images. Three databases of STARE, DRIVE, and a local database have been examined. It is shown that this method provides high detection rates while achieving faster processing speeds than the existing methods that have reported comparable detection rates. For example, the detection rate for the STARE database which is the most widely used database is found to be 96.3% with a processing time of about 3s per image.

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.

DETECTION OF DIABETIC MACULOPATHY IN HUMAN RETINAL IMAGES USING MORPHOLOGICAL OPERATIONS

Diabetic Retinopathy (DR) is a common retinal complication associated with diabetes. Adverse changes in the retinal blood vessels lead to loss of vision without any symptoms. Diabetic retinopathy is frequent, blinding complication of the patient with high blood sugar levels that characterize diabetes. Automatic Recognition of DR lesions like Exudates, in digital fund us images can contribute to the diagnosis and screening of this disease. The Exudates present in the macula which is the center portion of the retina is called maculopathy or macular edema. In this approach, an automatic and efficient method to detect diabetic maculopathy which is a severe stage of Diabetic Retinopathy, is proposed. The real time retinal images were obtained from a nearby hospital. The retinal images were pre-processed via. Contrast Limited Adaptive Histogram Equalization (CLAHE). The preprocessed color retinal images are subjected to Top Hat Transform and Bottom Hat Transform. The Macula which is the darkest region was obtained. To classify the preprocessed image into Exudates and Non-Exudates, a set of features based on color and texture were extracted. Classification was done using support Vector Machine This method appears promising as it can detect the severity of the disease.

Automated Method for Identification and Artery-Venous Classification of Vessel Trees in Retinal Vessel Networks

The separation of the retinal vessel network into distinct arterial and venous vessel trees is of high interest. We propose an automated method for identification and separation of retinal vessel trees in a retinal color image by converting a vessel segmentation image into a vessel segment map and identifying the individual vessel trees by graph search. Orientation, width, and intensity of each vessel segment are utilized to find the optimal graph of vessel segments. The separated vessel trees are labeled as primary vessel or branches. We utilize the separated vessel trees for arterial-venous (AV) classification, based on the color properties of the vessels in each tree graph. We applied our approach to a dataset of 50 fundus images from 50 subjects. The proposed method resulted in an accuracy of 91.44 correctly classified vessel pixels as either artery or vein. The accuracy of correctly classified major vessel segments was 96.42.

Color Retinal Image Coding Based on Structural Information

Color Retinal Image Coding Based on Structural Information

Bio-medical imaging: Localization of main structures in retinal fundus images

Retinal fundus images have three main structures, the optic disk, fovea and blood vessels. By examining fundus images, an ophthalmologist can diagnose various clinical disorders of the eye and the body, typically indicated by changes in the diameter, area, branching angles and tortuosity of the three ma in retinal structures. Knowledge of the optic disk position is an important diagnostic index fo r many diseases related to the retina. In this paper, localization of optic disc is discussed. Optic disk detection is based on morphological operationsand smoothing filters. Blood vessels are extracted using the green component of a colour retinal image with the help of a median filter. Maximum intensity values are validated with blood vessels to localize the optic disk location. The proposed method has shown significant improvements in results.

Performance Evaluation of Color Retinal Image Quality Assessment in Asymmetric Channel VQ Coding

The RGB color retinal image has an interesting characteristic, i.e. the G channel contains more important information than the other ones. One of the most important features in a retinal image is the retinal blood vessel structure. Many diseases can be diagnosed based on in the retinal blood vessel, such as micro aneurysms that can lead to blindness. In the G channel, the contrast between retinal blood vessel and its background is significantly high. The authors explore this retinal image characteristic to construct a more suitable image coding system. The coding processes are conduct in three schemes: weighted R channel, weighted G channel, and weighted B channel coding. Their hypothesis is that allocating more bits in the G channel will improve the coding performance. The authors seek for image quality assessment (IQA) metrics that can be used to measure the distortion in retinal image coding. Three different metrics, namely Peak Signal to Noise Ratio (PSNR), Structure Similarity (SSIM), and Visual Information Fidelity (VIF) are compared as objective assessment in image coding and to show quantitatively that G channel has more important role compared to the other ones. The authors use Vector Quantization (VQ) as image coding method due to its simplicity and low-complexity than the other methods. Experiments with actual retinal image shows that the minimum value of SSIM and VIF required in this coding scheme is 0.9940 and 0.8637.

An efficient approach for robust multimodal retinal image registration based on UR-SIFT features and PIIFD descriptors

Existing algorithms based on scale invariant feature transform (SIFT) and Harris corners such as edge-driven dual-bootstrap iterative closest point and Harris-partial intensity invariant feature descriptor (PIIFD) respectivley have been shown to be robust in registering multimodal retinal images. However, they fail to register color retinal images with other modalities in the presence of large content or scale changes. Moreover, the approaches need preprocessing operations such as image resizing to do well. This restricts the application of image registration for further analysis such as change detection and image fusion. Motivated by the need for efficient registration of multimodal retinal image pairs, this paper introduces a novel integrated approach which exploits features of uniform robust scale invariant feature transform (UR-SIFT) and PIIFD. The approach is robust against low content contrast of color images and large content, appearance, and scale changes between color and other retinal image modalities like the fluorescein angiography. Due to low efficiency of standard SIFT detector for multimodal images, the UR-SIFT algorithm extracts high stable and distinctive features in the full distribution of location and scale in images. Then, feature points are adequate and repeatable. Moreover, the PIIFD descriptor is symmetric to contrast, which makes it suitable for robust multimodal image registration. After the UR-SIFT feature extraction and the PIIFD descriptor generation in images, an initial cross-matching process is performed and followed by a mismatch elimination algorithm. Our dataset consists of 120 pairs of multimodal retinal images. Experiment results show the outperformance of the UR-SIFT-PIIFD over the Harris-PIIFD and similar algorithms in terms of efficiency and positional accuracy.