Abstract—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. Index Terms—Fundus, Gabor, pattern classification, retina, vessel segmentation, wavelet.

Motivated by the goals of improving detection of low-contrast and narrow vessels and eliminating false detections at non-vascular structures, a new technique is presented for extracting vessels in retinal images. The core of the technique is a new likelihood ratio test that combines matchedfilter responses, confidence measures and vessel boundary measures. Matched filter responses are derived in scale-space to extract vessels of widely varying widths. A vessel confidence measure is defined as a projection of a vector formed from a normalized pixel neighborhood onto a normalized ideal vessel profile. Vessel boundary measures and associated confidences are computed at potential vessel boundaries. Combined, these responses form a 6-dimensional measurement vector at each pixel. A training technique is used to develop a mapping of this vector to a likelihood ratio that measures the "vesselness" at each pixel. Results comparing this vesselness measure to matched filters alone and to measures based on the Hessian of intensities show substantial improvements both qualitatively and quantitatively. The Hessian can be used in place of the matched filter to obtain similar but less-substantial improvements or to steer the matched filter by preselecting kernel orientations. Finally, the new vesselness likelihood ratio is embedded into a vessel tracing framework, resulting in an e#cient and e#ective vessel centerline extraction algorithm.

The segmentation and analysis of blood vessels has received much attention in the research community. The results aid numerous applications for diagnosis and treatment of vascular diseases. Here we use level set propagation with local phase information to capture the boundaries of vessels. The basic notion is that local phase, extracted using quadrature filters, allows us to distinguish between lines and edges in an image. Noting that vessels appear either as lines or edge pairs, we integrate multiple scales and capture information about vessels of varying width. The outcome is a “global” phase which can be used to drive a contour robustly towards the vessel edges. We show promising results in 2D and 3D. Comparison with a related method gives similar or even better results and at a computational cost several orders of magnitude less. Even with very sparse initializations, our method captures a large portion of the vessel tree. 1.

Motivated by the goals of improving detection of low-contrast and narrow vessels and eliminating false detections at non-vascular structures, a new technique is presented for extracting vessels in retinal images. The core of the technique is a new likelihood ratio test that combines matched filter responses, confidence measures and vessel boundary measures. Matched filter responses are derived in scale-space to extract vessels of widely varying widths. A vessel confidence measure is defined as a projection of a vector formed from a normalized pixel neighborhood onto a normalized ideal vessel profile. Vessel boundary measures and associated confidences are computed at potential vessel boundaries. Combined, these responses form a 6-dimensional measurement vector at each pixel. A learning technique is applied to map this vector to a likelihood ratio that measures the "vesselness" at each pixel. Results comparing this vesselness measure to matched filters alone and to measures based on the intensity Hessian show substantial improvements both qualitatively and quantitatively. When the Hessian is used in place of the matched filter, similar but less-substantial improvements are obtained. Finally, the new vesselness likelihood ratio is embedded into a vessel tracing framework, resulting in an e#cient and e#ective vessel extraction algorithm.

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 continuous twodimensional Morlet wavelet transform responses taken at multiple scales. The Morlet 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 [34] and STARE [16] databases of manually labeled non-mydriatic images. On the DRIVE database, it achieves an area under the receiver operating characteristic (ROC) curve of 0.9598 and an accuracy of 0.9467 versus 0.9473 for a second human observer.

Abstract. We present a novel approach to 3D delineation of dendritic networks in noisy image stacks. We achieve a level of automation beyond that of stateof-the-art systems, which model dendrites as continuous tubular structures and postulate simple appearance models. Instead, we learn models from the data itself, which make them better suited to handle noise and deviations from expected appearance. From very little expert-labeled ground truth, we train both a classifier to recognize individual dendrite voxels and a density model to classify segments connecting pairs of points as dendrite-like or not. Given these models, we can then trace the dendritic trees of neurons automatically by enforcing the tree structure of the resulting graph. We will show that our approach performs better than traditional techniques on brighfield image stacks. 1

This paper presents an improved version of our specific methodology to detect the vessel tree in retinal angiographies. The automatic analysis of retinal vessel tree facilitates the computation of the arteriovenous index, which is essential for the diagnosis several eye diseases. The developed system is inspired in the classical snake but incorporating domain specific knowledge, such as blood vessels topological properties. It profits from the automatic localization of the optic disc, the vessel creases extraction and, as a recent innovation, the morphological vessel segmentation, all developed in our research group. After researching and testing our system, the parameter configuration has been enhanced. Significantly better results in the detection of arteriovenous structures are obtained, keeping a high efficiency, as shown by the systems performance evaluation on the publicly available DRIVE database. 1.

This paper presents a new method to extract a network of vessels centerlines from a medical image. The network is composed of local geodesics over a four-dimensional space that includes local orientation and scale. These shortest paths follow closely the center of tubular structures and can deal robustly with crossings over the image plane. The vessel network is grown by an iterative algorithm that distributes seed points according to a geodesic saliency field. Numerical experiments on a database of synthetic and medical images show the superiority of our approach with respect to several methods based on shortest paths extractions. With a minimum of user interaction, it allows to compute a complex network of vessels over noisy medical. 1

Accurate extraction of retinal blood vessels is an important task in computer aided diagnosis of retinopathy. The Matched Filter (MF) is a simple yet effective method for vessel extraction. However, a MF will respond not only to vessels but also to non-vessel edges. This will lead to frequent false vessel detection. In this paper we propose a novel extension of the MF approach, namely the MF-FDOG, to detect retinal blood vessels. The proposed MF-FDOG is composed of the original MF, which is a zero-mean Gaussian function, and the first-order derivative of Gaussian (FDOG). The vessels are detected by thresholding the retinal image’s response to the MF, while the threshold is adjusted by the image’s response to the FDOG. The proposed MF-FDOG method is very simple; however, it reduces significantly the false detections produced by the original MF and detects many fine vessels that are missed by the MF. It achieves competitive vessel detection results as compared with those state-of-the-art schemes but with much lower complexity. In addition, it performs well at extracting vessels from pathological retinal images.

for level set segmentation