Optic Disc Detection Method Research Articles

Optic disc detection based on fully convolutional network and weighted matrix recovery model.

Eye diseases often affect human health. Accurate detection of the optic disc contour is one of the important steps in diagnosing and treating eye diseases. However, the structure of fundus images is complex, and the optic disc region is often disturbed by blood vessels. Considering that the optic disc is usually a saliency region in fundus images, we propose a weakly-supervised optic disc detection method based on the fully convolution neural network (FCN) combined with the weighted low-rank matrix recovery model (WLRR). Firstly, we extract the low-level features of the fundus image and cluster the pixels using the Simple Linear Iterative Clustering (SLIC) algorithm to generate the feature matrix. Secondly, the top-down semantic prior information provided by FCN and bottom-up background prior information of the optic disc region are used to jointly construct the prior information weighting matrix, which more accurately guides the decomposition of the feature matrix into a sparse matrix representing the optic disc and a low-rank matrix representing the background. Experimental results on the DRISHTI-GS dataset and IDRiD dataset show that our method can segment the optic disc region accurately, and its performance is better than existing weakly-supervised optic disc segmentation methods. Graphical abstract of optic disc segmentation.

Fully Convolutional Network and Visual Saliency-Based Automatic Optic Disc Detection in Retinal Fundus Images.

We present in this paper a novel optic disc detection method based on a fully convolutional network and visual saliency in retinal fundus images. Firstly, we employ the morphological reconstruction-based object detection method to locate the optic disc region roughly. According to the location result, a 400 × 400 image patch that covers the whole optic disc is obtained by cropping the original retinal fundus image. Secondly, the Simple Linear Iterative Cluster approach is utilized to segment such an image patch into many smaller superpixels. Thirdly, each superpixel is assigned a uniform initial saliency value according to the background prior information based on the assumption that the superpixels located on the boundary of the image belong to the background. Meanwhile, we use a pretrained fully convolutional network to extract the deep features from different layers of the network and design the strategy to represent each superpixel by the deep features. Finally, both the background prior information and the deep features are integrated into the single-layer cellular automata framework to gain the accurate optic disc detection result. We utilize the DRISHTI-GS dataset and RIM-ONE r3 dataset to evaluate the performance of our method. The experimental results demonstrate that the proposed method can overcome the influence of intensity inhomogeneity, weak contrast, and the complex surroundings of the optic disc effectively and has superior performance in terms of accuracy and robustness.

A novel method for optic disc detection in retinal images using the cuckoo search algorithm and structural similarity index

Accurate and reliable optic disk (OD) localization is vital for eye disease monitoring and fundus image analysis. This paper describes a novel technique to localize the OD in retinal images using the cuckoo search algorithm and the structural similarity index measure (SSIM). SSIM uses the average OD value to compare with candidate OD. Hence, the average OD values were calculated from randomly selected images. The average OD values and the colored retina fundus images were given as input to the proposed algorithm. The adaptive histogram equalization method was applied to ensure that the brightness and contrast values in all images were within a similar range. Next, candidate OD centers were calculated using the search algorithm and the similarity value between each candidate OD and the average OD was determined. Finally, the computed similarity was maximized by the search algorithm and the true OD center was found. The performance of the OD detection algorithm was evaluated on three public datasets. The experimental results showed that proposed method achieved comparable performance, without employing complex image pre-processing, compared with the state-of-the-art techniques. Specifically, the accuracy of 100%, 100%, and 97.5% were obtained for ONHSD, DRIONS and DRIVE datasets, respectively.

Efficient and robust optic disc detection and fovea localization using region proposal network and cascaded network

The optic disc (OD) and fovea are two important anatomical landmarks of the retina. Localization of OD and fovea plays an important role in retinal image analysis. This paper proposes a novel, efficient and robust method for OD detection and fovea localization. The proposed method consists of two steps. First, we use region proposal network to generate multiple OD region proposals, and then OD’s bounding box is identified as the proposal with the highest probability. Second, we create a region of interest containing the fovea based on the geometrical relationship between OD and fovea, and then employ a three-level cascaded convolutional neural network to locate the fovea, which is treated as a regression problem. The proposed method is trained and evaluated on three sets of fundus images (the first dataset contains 1955 fundus images and the other two are publicly available datasets respectively containing 516 and 1200 fundus images). According to 5-fold cross-validation experiments, our OD detection accuracy is respective 99.59%, 100%, and 100% for the three datasets and the corresponding fovea localization accuracy is respective 97.8%, 99.03%, and 99.25%. The proposed method is found to be robust and effective in OD detection and fovea localization, and its accuracy is comparable, or even superior to representative state-of-the-art algorithms.

Automated Method for Optic Disc Detection and Elimination in Digital Fundus Images

Localizing, segmenting and eliminating the optic disc region of a fundus image is a prerequisite task in the automatic investigation of a number of retinal diseases such as Diabetic retinopathy, Glaucoma, Macular Edema, etc. Accurate detection of optic disc is a challenging task due to a number of reasons. Optic disc in most fundus images does not exhibit clear disc boundaries and there are number of blood vessels crossing it. An important task in automated retinal image analysis system is the detection and elimination of optic disc because the lesion regions in diabetic retinopathy closely resemble the color and texture of an optic disc. Hence, eliminating the optic disc region can improve the performance of diabetic retinopathy detection. The proposed work presents a novel method for optic disc segmentation which is not restricted by the location of the optic disc on the retina. The proposed algorithm localizes the position of the optic disc that is independent of its location and dynamically finds its center. The proposed method is tested on images from DRISHTI-GS, DIARETDB1, DRIONS-DB and DRIVE databases based on morphological operation and finding the largest connected component. The precision values of segmentation for digital fundus images from DRISHTI-GS, DIARETDB1, DRIONS-DB, and DRIVE databases are 0.98, 0.99, 0.98 and 0.99 respectively using the proposed method. The algorithm has yielded consistent high values of precision and recall indicating its robustness and efficiency.

Automatic Optic Disc Center and Boundary Detection in Color Fundus Images

Accurately detection of retinal landmarks, like optic disc, is an important step in the computer aided diagnosis frameworks. This paper presents an efficient method for automatic detection of the optic disc’s center and estimating its boundary. The center and initial diameter of optic disc are estimated by employing an ANN classifier. The ANN classifier employs visual features of vessels and their background tissue to classify extracted main vessels of retina into two groups: the vessels inside the optic disc and the vessels outside the optic disc. To this end, average intensity values and standard deviation of RGB channels, average width and orientation of the vessels and density of the detected vessels their junction points in a window around each central pixel of main vessels are employed. The center of detected vessels, which are belonging to the inside of the optic disc region, is adopted as the optic disc center and the average length of them in vertical and horizontal directions is selected as initial diameter of the optic disc circle. Then exact boundary of the optic disc is extracted using radial analysis of the initial circle. The performance of the proposed method is measured on the publicly available DRIONS, DRIVE and DIARETDB1 databases and compared with several state-of-the-art methods. The proposed method shows much higher mean overlap (70.6%) in the same range of detection accuracy (97.7%) and center distance (12 pixels). The average sensitivity and predictive values of the proposed optic disc detection method are 80.3% and 84.6% respectively.

A shortest path approach to optic disc detection in retinal fundus images

Aim: To evaluate a new algorithm to detect the optic disc in retinal fundus images on a number of publicly available datasets. Optic Disc detection is an important first step in many automated algorithms, either to be masked out of future processing or for the use in optic disc related disease such as glaucoma and papilledema.Methods: We propose a new method for optic disc detection that converts the retinal image into a graph and exploits vessel enhancement methods to calculate edge weights in finding the shortest path between pairs of points on the periphery of the image. The line segment with the maximum number of shortest paths is considered the optic disc location, with refinement from a combination template matching approach in the found region. The method was tested on three publicly available datasets: DRIVE, DIARETDB1, and Messidor consisting of 40, 89, and 1200 images. All images were acquired at a 45°-50° field of view.Results: The method achieves an accuracy of 100, 98.88, and 99.42% on the DRIVE, DIARETDB1, and Messidor databases respectively.Conclusions: The method performs as well or better than state-of-the-art methods on these datasets. Processing takes an average of 32 seconds (+-1.2) to detect the optic disc, with 26 of those seconds used for the vessel enhancement process. The accuracy over a wide variety of images shows that the method is robust and would be optimal for retinal analysis systems that perform vessel enhancement as part of their processing.

Effective optic disc detection method based on swarm intelligence techniques and novel pre-processing steps

Diabetic retinopathy affects the vision of a significant fraction of the population worldwide. Retinal fundus images are used to detect the condition before vision loss develops to enable medical interventions. Optic disc detection is an essential step for the automatic detection of the disease. Several techniques have been introduced in the literature to detect the optic disc with different performance characteristics such as speed, accuracy and consistency. For optic disc detection, a nature-inspired algorithm called swarm intelligence has been shown to have clear superiority in terms of speed and accuracy compared to traditional detection algorithms. We therefore further investigated and compared several swarm intelligence techniques. Our study focused on five popular swarm intelligence algorithms: artificial bee colony, particle swarm optimization, bat algorithm, cuckoo search and firefly algorithm. This work also featured a novel pre-processing scheme that enhances the detection accuracy of the swarm techniques by making the optic disc region the highest grayscale value in the image. The pre-processing involves multiple stages of background subtraction, median filtering and mean filtering and is named Background Subtraction-based Optic Disc Detection (BSODD). The best result was obtained by combining our pre-processing technique, firefly algorithm and the parameters used for the algorithm. The obtained accuracy was superior to the other tested algorithms and published results in the literature. The accuracy of the firefly algorithm was 100%, 100%, 98.82% and 95% when using the DRIVE, DiaRetDB1, DMED and STARE databases, respectively.

Optic disc detection and boundary extraction in retinal images.

With the development of digital image processing, analysis and modeling techniques, automatic retinal image analysis is emerging as an important screening tool for early detection of ophthalmologic disorders such as diabetic retinopathy and glaucoma. In this paper, a robust method for optic disc detection and extraction of the optic disc boundary is proposed to help in the development of computer-assisted diagnosis and treatment of such ophthalmic disease. The proposed method is based on morphological operations, smoothing filters, and the marker controlled watershed transform. Internal and external markers are used to first modify the gradient magnitude image and then the watershed transformation is applied on this modified gradient magnitude image for boundary extraction. This method has shown significant improvement over existing methods in terms of detection and boundary extraction of the optic disc. The proposed method has optic disc detection success rate of 100%, 100%, 100% and 98.9% for the DRIVE, Shifa, CHASE_DB1, and DIARETDB1 databases, respectively. The optic disc boundary detection achieved an average spatial overlap of 61.88%, 70.96%, 45.61%, and 54.69% for these databases, respectively, which are higher than currents methods.

A Comparison and Evaluation of Computerized Methods for OD Localization and Detection in Retinal Images

Automated localization and detection of optic disc (OD) is essential step in the analysis of digital diabetic retinopathy systems. Retinal images give unique diagnostic about eye diseases. The detection of OD helps in early detection of proliferative diabetic retinopathy. In this paper two computerized optic disc localization and detection techniques are compared. First technique is uses edge detection method to localize the OD and Hough transform method to detect the OD. Second technique uses brightest region method for OD localization and LOG filtering method for OD detection. These techniques can be used to develop automated diagnostic system for OD localization and detection. We have used publicly available DRIVE and STARE databases for the testing of the two techniques. Experimental results showed that Hough transform technique gives better results.