Jihene Malek

Computational analysis of blood flow in the retinal arteries and veins using fundus image

The retina is the only tissue in which blood vessels can be visualized non-invasively in vivo. Thus, the study of the retinal hemodynamic has special interest for both physiological and pathological conditions. The aim of this study has been to develop a detailed computational model for a quantitative analysis of the blood flow in physiologically realistic retinal arterial and venous networks. The geometrical outlines of both retinal artery and vein have been extracted from the retinal image acquired from a healthy young adult by a retinal camera Topcon TRC-50EX. The microvascular diameter effect (i.e., Fahraeus-Lindqvist effect) and the hematocrit have been considered in determining the viscosity of the blood in the retinal vessel segments. The blood moves at a velocity that is 2 times less in the veins (maximum 5.4 cm/s) than the velocity at which it moves in the arteries (maximum 11 cm/s) which are in good agreement with in vivo measurements reported in the literature. The pressure drop has been in the range of 11-14 mmHg between the inlet and outlets for the arterial network, and 13-14 mmHg for the vein network. The developed method can be used as a tool for continuous monitoring of the retinal circulation for clinical assessments as well as experimental studies.

Blood vessels extraction and classification into arteries and veins in retinal images

Many retinal diseases are characterized by changes in retinal vessels. The retina vascular structure consists of two kinds of vessels: arteries and veins. An important symptom for Diabetic Retinopathy DR is irregularly wide veins, leading to an unusually low ratio of the average diameter of arteries to veins (AVR). In this paper, we present an approach to separate arteries and veins based on a segmentation and neural classification method. Blood vessels are segmented using two-dimensional matched filters, which derived from Gaussian functions. We used feature vectors based on vessel profile extraction for each segment. The obtained features will be introduced as the input vector of a Multi-Layer Perceptron (MLP); to classify the vessel into arteries and veins ones. Our approach achieves 95.32% correctly classified vessel pixels classification.

Automatic extraction of blood vessels in the retinal vascular tree using multiscale medialness

We propose an algorithm for vessel extraction in retinal images. The first step consists of applying anisotropic diffusion filtering in the initial vessel network in order to restore disconnected vessel lines and eliminate noisy lines. In the second step, a multiscale line-tracking procedure allows detecting all vessels having similar dimensions at a chosen scale. Computing the individual image maps requires different steps. First, a number of points are preselected using the eigenvalues of the Hessian matrix. These points are expected to be near to a vessel axis. Then, for each preselected point, the response map is computed from gradient information of the image at the current scale. Finally, the multiscale image map is derived after combining the individual image maps at different scales (sizes). Two publicly available datasets have been used to test the performance of the suggested method. The main dataset is the STARE projects dataset and the second one is the DRIVE dataset. The experimental results, applied on the STARE dataset, show a maximum accuracy average of around 94.02%. Also, when performed on the DRIVE database, the maximum accuracy average reaches 91.55%.

Automated optic disc detection in retinal images by applying region-based active aontour model in a variational level set formulation

An efficient optic disk localization and segmentation are important tasks in an automated retinal image analysis system. General-purpose edge detection algorithms often fail to segment the optic disk due to fuzzy boundaries, inconsistent image contrast or missing edge features. This paper presents a method to automatically locate and boundary detect of the optic disk. The detection procedure comprises two independent methodologies. On one hand, a location methodology obtains a pixel that belongs to the OD using iterative thresholding method followed by Principal Component Analysis techniques (PCA) and, on the other hand, a boundary segmentation methodology estimates the OD boundary by applying region-based active contour model in a variational level set formulation (RSF). The method uses an improved geometric active contour model which can not only solve the boundary leakage problem but also is less sensitive to intensity inhomogeneity The results from the RSF method were compared with conventional optic disk detection using a geometric active contour models (ACM) and later verified with hand-drawn ground truth. Results indicate 89% accuracy for identification and 95.05% average accuracy in localizing the optic disc boundary.

Problems in pattern classification in high domensional spaces: behavior of a class of combined neuro-fuzzy classifiers

The designer of a pattern classification system is often faced with the following situation: finite sets of samples, from the various classes, are available along with a large set of measurements, or features, to be computed from the patterns. In pattern recognition literature, several investigations have demonstrated the relationship between dimensionality, sample size and recognition accuracy. The ultimate goal of this paper is to study the generalization error of statistical (1_NN, 4_NN), fuzzy (FCM, FKCN), neural (MLPNN, RBFNN) and neuro-fuzzy classifiers in high dimensional spaces. Computational complexity of classification algorithms in high dimensional spaces is also discussed. Our experimental results show the robustness of fuzzy, neural, and neuro-fuzzy classifiers to the curse of dimensionality.

Adaptive noise-reducing anisotropic diffusion filter

AbstractIn image processing and computer vision, the denoising process is an important step before several processing tasks. This paper presents a new adaptive noise-reducing anisotropic diffusion (ANRAD) method to improve the image quality, which can be considered as a modified version of a speckle-reducing anisotropic diffusion (SRAD) filter. The SRAD works very well for monochrome images with speckle noise. However, in the case of images corrupted with other types of noise, it cannot provide optimal image quality due to the inaccurate noise model. The ANRAD method introduces an automatic RGB noise model estimator in a partial differential equation system similar to the SRAD diffusion, which estimates at each iteration an upper bound of the real noise level function by fitting a lower envelope to the standard deviations of pre-segment image variances. Compared to the conventional SRAD filter, the proposed filter has the advantage of being adapted to the color noise produced by today’s CCD digital camera. The simulation results show that the ANRAD filter can reduce the noise while preserving image edges and fine details very well. Also, it is favorably compared to the fast non-local means filter, showing an improvement in the quality of the restored image. A quantitative comparison measure is given by the parameters like the mean structural similarity index and the peak signal-to-noise ratio.

An automated vessel segmentation of retinal images using multiscale vesselness

The ocular fundus image can provide information on pathological changes caused by local ocular diseases and early signs of certain systemic diseases, such as diabetes and hypertension. Automated analysis and interpretation of fundus images has become a necessary and important diagnostic procedure in ophthalmology. The extraction of blood vessels from retinal images is an important and challenging task in medical analysis and diagnosis. In this paper, we introduce an implementation of the anisotropic diffusion which allows reducing the noise and better preserving small structures like vessels in 2D images. A vessel detection filter, based on a multi-scale vesselness function, is then applied to enhance vascular structures.

Fundus image denoising using FPGA hardware architecture

Image processing algorithms, implemented in hardware, have recently emerged as the most viable solution for improving the performance of image processing systems. In this paper, a version of an anisotropic diffusion technique is used to reduce noise from retinal images, namely Speckle Reducing Anisotropic Diffusion SRAD. The SRAD filter can improve images corrupted by multiplicative or additive noise, but it has been the most computationally complex and it has not been suitable for software implementation in real-time processing. In this paper, an efficient Field-Programmable Gate Array FPGA-based implementation of the SRAD filter is presented to accelerate the processing time. A comparison of the most used classical suppression filters like Gaussian, Median, Perona and Malik anisotropic diffusion has been carried out. The experimental results reveal a 38× performance improvement over the original MATLAB implementation and a 1.33× performance improvement over the hardware implementation using the Xilinx System Generator tool.

Restoration of retinal images using anisotropic diffusion like algorithms

In image processing by the partial differential equations (PDEs), the first and the simplest models to have and to use are based on linear diffusion. The common difficulty of linear filters is the excessive smoothing which makes track edges difficult. Therefore, we can affirm that any improvement of these linear models must be carried out inside the operator of diffusion, thus sacrificing their linearity. We will see how these difficulties can be overcome by the use of the nonlinear models. The work achieved in this context will make the subject of the following paper. This document treats the automatic preprocessing of retinal vascular network in fundus images in order to improve the interpretation of the images for the doctors diagnosis. We propose to deal with the image restoration using original equation of anisotropic diffusion. Compared to traditional anisotropic diffusion filters, it has interesting capacities of smoothing, like the expected conservation of the details and contours, and especially a more continuous smoothing intra-area, avoiding the pitfall of stairs or of the mosaics.

Fast oriented Anisotropic Diffusion filter

In image and video processing the denoising process is an important step before several processing tasks. This paper presents a Faster Oriented Speckle Reducing Anisotropic Diffusion filter (FOSRAD) method to speed up the processing time and keep a higher quality of image, which can be considered as a modified version of the Oriented Speckle Reducing Anisotropic Diffusion (OSRAD) filter. The OSRAD works very well for denoising images with speckle noise. However, this filter has a powerful computational complexity and is not suitable for real time implementation. In this paper we propose a new scheme for optimizing the processing time based on look ahead decomposition technique. This method leads to dividing the processing time by two. Compared to the conventional OSRAD filter, the proposed filter has the advantage of speeding up the numerical scheme. The simulation result show that the FOSRAD filter improved the execution time by 14× compared to the original OSRAD filter. A comparison measure is given by the metrics like the mean structural similarity index and the peak signal-to-noise ratio.