A method is presented for automated segmentation of vessels in two-dimensional color images of the retina. This method can be used in computer analyses of retinal images, e.g., in automated screening for diabetic retinopathy. The system is based on extraction of image ridges, which coincide approximately with vessel centerlines. The ridges are used to compose primitives in the form of line elements. With the line elements an image is partitioned into patches by assigning each image pixel to the closest line element. Every line element constitutes a local coordinate frame for its corresponding patch. For every pixel, feature vectors are computed that make use of properties of the patches and the line elements. The feature vectors are classified using a kNN-classifier and sequential forward feature selection. The algorithm was tested on a database consisting of 40 manually labeled images. The method achieves an area under the receiver operating characteristic curve of 0.952. The method is compared with two recently published rule-based methods of Hoover et al. and Jiang et al. . The results show that our method is significantly better than the two rule-based methods (p<0.01). The accuracy of our method is 0.944 versus 0.947 for a second observer.

/pdf/ridge-based-vessel-segmentation-in-color-images-of-the-1i9un3yqbx.pdf

Ridge-based vessel segmentation in color images of the retina

Describes an automated method to locate and outline blood vessels in images of the ocular fundus. Such a tool should prove useful to eye care specialists for purposes of patient screening, treatment evaluation, and clinical study. The authors' method differs from previously known methods in that it uses local and global vessel features cooperatively to segment the vessel network. The authors evaluate their method using hand-labeled ground truth segmentations of 20 images. A plot of the operating characteristic shows that the authors' method reduces false positives by as much as 15 times over basic thresholding of a matched filter response (MFR), at up to a 75% true positive rate. For a baseline, they also compared the ground truth against a second hand-labeling, yielding a 90% true positive and a 4% false positive detection rate, on average. These numbers suggest there is still room for a 15% true positive rate improvement, with the same false positive rate, over the authors' method. They are making all their images and hand labelings publicly available for interested researchers to use in evaluating related methods.

/pdf/locating-blood-vessels-in-retinal-images-by-piecewise-4lgaualy1f.pdf

Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response

Hexagonal ferrites: A review of the synthesis, properties and applications of hexaferrite ceramics

We present a method for automated segmentation of the vasculature in retinal images. The method produces segmentations by classifying each image pixel as vessel or nonvessel, based on the pixel's feature vector. Feature vectors are composed of the pixel's intensity and two-dimensional Gabor wavelet transform responses taken at multiple scales. The Gabor wavelet is capable of tuning to specific frequencies, thus allowing noise filtering and vessel enhancement in a single step. We use a Bayesian classifier with class-conditional probability density functions (likelihoods) described as Gaussian mixtures, yielding a fast classification, while being able to model complex decision surfaces. The probability distributions are estimated based on a training set of labeled pixels obtained from manual segmentations. The method's performance is evaluated on publicly available DRIVE (Staal et al.,2004) and STARE (Hoover et al.,2000) databases of manually labeled images. On the DRIVE database, it achieves an area under the receiver operating characteristic curve of 0.9614, being slightly superior than that presented by state-of-the-art approaches. We are making our implementation available as open source MATLAB scripts for researchers interested in implementation details, evaluation, or development of methods

/pdf/retinal-vessel-segmentation-using-the-2-d-gabor-wavelet-and-3rb2432h14.pdf

Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification

Fast Fourier Transform

In this paper the authors present a new unsupervised fuzzy algorithm for vessel tracking that is applied to the detection of the ocular fundus vessels. The proposed method overcomes the problems of initialization and vessel profile modeling that are encountered in the literature and automatically tracks fundus vessels using linguistic descriptions like "vessel" and "nonvessel." The main tool for determining vessel and nonvessel regions along a vessel profile is the fuzzy C-means clustering algorithm that is fed with properly preprocessed data, Additional procedures for checking the validity of the detected vessels and handling junctions and forks are also presented. The application of the proposed algorithm to fundus images and simulated vessels resulted in very good overall performance and consistent estimation of vessel parameters.

A fuzzy vessel tracking algorithm for retinal images based on fuzzy clustering

We present an adaptive fuzzy clustering scheme for image segmentation, the adaptive fuzzy clustering/segmentation (AFCS) algorithm. In AFCS, the nonstationary nature of images is taken into account by modifying the prototype vectors as functions of the sample location in the image. The inherent high interpixel correlation is modeled using neighborhood information. A multiresolution model is utilized for estimating the spatially varying prototype vectors for different window sizes. The fuzzy segmentations at different resolutions are combined using a data fusion process in order to compute the final fuzzy partition matrix. The results provide segmentations, having lower fuzzy entropy when compared to the possibilistic C-means algorithm, while maintaining the image's main characteristics. In addition, due to the neighborhood model, the effects of noise in the form of single pixel regions are minimized.

Image segmentation by a fuzzy clustering algorithm using adaptive spatially constrained membership functions

The automatic and accurate P phase arrival identification is a fundamental problem for seismologists worldwide. Several approaches have been reported in the literature, but most of them only selectively deal with the problem and are severely affected by noise presence. In this paper, a new approach based on higher-order statistics (HOS) is introduced that overcomes the subjectivity of human intervention and eliminates the noise factor. By using skewness and kurtosis, two algorithms have been formed, namely, Phase Arrival Identification-Skewness/Kurtosis (PAI-S/K), and some advantages have been gained over the usual approaches, resulting in the automatic identification of the transition from Gaussianity to non-Gaussianity that coincides with the onset of the seismic event, despite noise presence. Experimental results on real seismic data, gathered by the Seismological Network of the Department of Geophysics of Aristotle University, demonstrate an excellent performance of the PAI-S/K scheme, regarding both accuracy and noise robustness. The simplicity of the proposed method makes it an attractive candidate for huge seismic data assessment in a real-time context.

PAI-S/K: A robust automatic seismic P phase arrival identification scheme

The theory and applications of eddy currents induced in conducting materials by time-varying magnetic fields are reviewed. The mathematical methods employed in solving the relevant problems are presented. Both analytical and numerical methods are described. Applications based on effects arising from eddy currents are discussed in detail. These applications are to magnetic levitation, electromagnetic launching, hyperthermia treatment of cancer, and nondestructive testing. >

Eddy currents: theory and applications

A novel approach for enhancing the results of fuzzy clustering by imposing spatial constraints for solving image segmentation problems is presented. We have developed a Sugeno (185) type rule-based system with three inputs and 11 rules that interacts with the clustering results obtained by the well-known fuzzy c-means (FCM) and/or possibilistic c-means (PCM) algorithms. It provides good image segmentations in terms of region smoothness and elimination of the effects of noise. The results of the proposed rule-based neighborhood enhancement (RB-NE) system are compared to well-known segmentation algorithms using stochastic field modeling. They are found to be of comparable quality, while being of lower computational complexity.

Stavros M. Panas

Papers

A fuzzy vessel tracking algorithm for retinal images based on fuzzy clustering

Image segmentation by a fuzzy clustering algorithm using adaptive spatially constrained membership functions

PAI-S/K: A robust automatic seismic P phase arrival identification scheme

Eddy currents: theory and applications

On applying spatial constraints in fuzzy image clustering using a fuzzy rule-based system