Mahmoud R. El-Sakka

Carotid artery ultrasound image segmentation using fuzzy region growing

In this paper, we propose a new scheme for extracting the contour of the carotid artery using ultrasound images. Starting from a user defined seed point within the artery, the scheme uses the fuzzy region growing algorithm to create a fuzzy connectedness map for the image. Then, the fuzzy connectedness map is thresholded using a threshold selection mechanism to segment the area inside the artery. Experimental results demonstrated the efficiency of the proposed scheme in segmenting carotid artery ultrasound images, and it is insensitive to the seed point location, as long as it is located inside the artery.

Watershed segmentation for carotid artery ultrasound images

Summary form only given. This paper introduces a novel segmentation scheme for carotid artery ultrasound images. The proposed scheme is based on watershed segmentation algorithm. It consists of four major stages. These stages are preprocessing, watershed segmentation, region merging and finally boundary extraction. The proposed scheme is tested using a set of carotid artery ultrasound images. The experimental results show that the proposed scheme can produce accurate contours.

Fuzzy C-means clustering for segmenting carotid artery ultrasound images

This paper introduces a fully automated segmentation scheme for carotid artery ultrasound images. The proposed scheme is based on fuzzy cmeans clustering. It consists of four major stages. These stages are preprocessing, feature extraction, fuzzy c-means clustering, and finally boundary extraction. Experimental results demonstrated the efficiency of the proposed scheme in segmenting carotid artery ultrasound images.

Carotid ultrasound segmentation using DP active contours

Ultrasound provides a non-invasive means for visualizing various tissues within the human body. However, these visualizations tend to be filled with speckle noise and other artifacts, due to the sporadic nature of high frequency sound waves. Many techniques for segmenting ultrasound images have been introduced in order to deal with these problems. One such technique is the active contouring. In this paper, two proposed alterations to the dynamic programming parametric active contour model (or snake) are introduced. The first alteration allows the snake to converge to the one-response result of a modified Canny edge detector. The second provides a function that allows a user to preset apriori knowledge about a given object being detected, by means of curve fitting and energy modification. The results yield accurate segmentations of crosssectional transverse carotid artery ultrasound images that are validated by an independent clinical radiologist. Utilizing the proposed alterations leads to a reduction of clinician interaction time while maintaining an acceptable level of accuracy for varying measures such as percent stenosis.

Localization and security enhancement of block-based image authentication

Most block-based image authentication techniques that are presented in the literature sacrifice localization accuracy in order to resist vector quantization (VQ) counterfeiting attacks. In this paper, we show that strong cryptography schemes, which produce a long signature, can be used to sign image blocks without regard to the size of these blocks. In addition, a new approach to generate overlapped watermark segments for image blocks is presented. These watermarks are generated using one-way function based on an NP-complete problem. Moreover, a block-based image authentication technique is proposed. This technique provides strong protection against the VQ attack, as well as a great enhancement in localization accuracy and system security.

Optical flow active contours with primitive shape priors for echocardiography

Accurate delineation of object borders is highly desirable in echocardiography, especially at the left ventricle. Among other model-based techniques, active contours (or snakes) provide a unique and powerful approach to image analysis. In this work, we propose the use of a new external energy for a gradient vector flow (GVF) snake, being the optical flow of a moving sequence (modeling the mechanical movement of the heart). This external energy can provide additional information to the active contour model garnering adequate results for moving sequences. An automatic iterative primitive shape prior was also applied in order to further improve the results of a GVF snake, when dealing with especially noisy echocardiographic images. Results were compared with expert-defined segmentations yielding acceptable sensitivity, precision rate and overlap ratio performance.

Grayscale true two-dimensional dictionary-based image compression

Dictionary-based encoding methods are popular forms of data compression. These methods were initially implemented to reduce the one-dimensional correlation in data, since they are designed to compress text. Therefore, they do not take advantage of the fact that adjacent pixels in images are correlated in two dimensions. Previous attempts have been made to adapt dictionary-based compression schemes to consider the two-dimensional nature of images, but mostly for binary images. In this paper, a two-dimensional dictionary-based lossless image compression scheme for grayscale images is introduced. The proposed scheme reduces correlation in image data by finding two-dimensional blocks of pixels that are approximately matched throughout the data and replacing them with short codewords. Test results show that the compression performance of the proposed scheme outperforms and surpasses any other existing dictionary-based lossless compression scheme. The results also show that it slightly outperforms JPEG-2000s compression performance, when it operates in its lossless mode, and it is comparable to JPEG-LSs and CALICs compression performance, where JPEG-2000 and JPEG-LS are the current image compression standards, and CALIC is a Context-based Adaptive Lossless Image Coding scheme.

Fuzzy Entropy Based Detection of Suspicious Masses in Digital Mammogram Images

Mammography is the standard method for screening and detecting breast abnormalities. In this paper, we propose a novel scheme for suspicious lesion detection in digital mammograms. The proposed scheme is based on image thresholding. The optimal threshold is determined by minimizing the fuzzy entropy of the image. Moreover, the paper introduces a new block-based performance criterion to compare between the computer generated and the radiologist segmented images. Experimental results over a set of sample images showed that the proposed scheme produces accurate segmentation results when compared with the manual results produced by radiologists. Hence the proposed scheme can be used as an effective tool in monitoring and detecting suspicious lesions on digital mammogram images

Active contours initialization for ultrasound carotid artery images

One of the major issues of active contouring methods is their sensitivity to the initial contour that is provided by the user. Unless it is drawn close enough to the actual contour, it may lead to unsatisfactory results. Thus, most active contour algorithms require considerable user interaction to provide a good initial contour. In this paper we present an efficient and fast rule- based algorithm for generating a good carotid artery initial contour from ultrasound images. Our algorithm reduces user interaction, as a user is only required to place a seed point inside the region of interest. Sensitivity, precision rate, and overlap ratio have been used to assess the amount of correlation between manually segmented lumens by an experienced clinician and by our algorithm, where the obtained 95% confidence interval of the mean value of these measures over all test cases are: [78.32% plusmn 1.98], [91.77% plusmn 1.96], and [72.68% plusmn 1.90], respectively. Furthermore, the output of our proposed scheme can be used as an input to any active contour algorithm to produce even better results.

An information theoretic image-quality measure

Lossy image compression techniques aim at encoding images with a minimal representation. During this process, some visually useful information may be lost. Assessing the information loss in decompressed images is not an easy task. In this paper, a new quantitative image-quality measure is introduced. This new measure incorporates information theory into the most commonly used objective criterion (the mean square error). The new measure has been tested by experiments performed on a wide variety of images. The results show an increase in the correlation between subjective rating by human observers and the normalized mean square error after applying the new measure.