Orthogonal moments such as Zernike moments and Legendre moments have been proven to have superior feature representation capability and low information redundancy. The number of orthogonal moments to be used as features or numerical attributes to perform any application is minimal due to the orthogonal nature. However, the information possessed by each moment order needs to be analysed to identify the appropriate moment orders for the undertaken task. In this work, a statistical significance test has been performed to select the best moment orders to discriminate normal and abnormal tissues in liver images. The experimental results reveal the efficacy of the proposed features.

Orthogonal Moments Based Texture Analysis of CT Liver Images

http://aragorn.pb.bialystok.pl/~mkret/docs/bbe2006.pdf

Texture characterization for hepatic tumor recognition in multiphase CT

Texture constitutes one of the fundamental properties of objects besides color and shape. In several image analysis applications, it is often the only exploitable quality of objects. As such, it has been studied, described, segmented, synthesized, or in short, analyzed extensively. Among the plethora of texture description methods, mathematical morphology deserves special attention as it excels at the exploitation of spatial relationships among pixels, rendering it inherently suitable for texture description. In this chapter, we focus on morphological texture description methods for grey-scale and color images in an effort to spread the advantages of this framework in the context of texture analysis. We review several descriptors, ranging from the basic granulometries and pattern spectra to the most advanced multivariate and multidimensional size, shape, orientation, and distance distributions, providing application examples as well as experimental results.

Morphological Texture Description of Grey-Scale and Color Images

Recent advances in computer science provide the intelligent computation tools needed to design and develop Diagnostic Support Systems (DSSs) that promise to increase the efficiency of physicians during their clinical practice. This chapter provides a brief overview of the use of computational intelligence methods in the design and development of DSSs aimed at the differential diagnosis of hepatic lesions from Computed Tomography (CT) images. Furthermore, examples of DSSs developed by our research team for supporting the diagnosis of focal liver lesions from non-enhanced CT images are presented.

Diagnostic Support Systems and Computational Intelligence: Differential Diagnosis of Hepatic Lesions from Computed Tomography Images

Identification of Liver and liver tumors from CT images is of great interest to physicians and image processing researchers. In this paper a simple and clinically useful system has been developed for segmenting the liver tumor from CT images. Level set methods have been widely used in image processing for segmenting the biomedical images such as liver images. Various methods of segmentation were explored, and a few were chosen for implementation and further development. Liver Images were collected and the region of interest was selected. Segmentation has been performed by using Fuzzy C means algorithm followed by fine delineation using level sets. The method could clearly segment the tumor regions and their boundaries are well defined. Index Terms—FCM, Level Set method, Liver tumors

Medical Image Segmentation Using CT Scans-A Level Set Approach

A quantitative study for the discrimination of different hepatic lesions is presented in this paper. The study is based on the fractal analysis of CT liver images in order to estimate their fractal dimension and to differentiate normal liver parenchyma from hepatocellular carcinoma. Four fractal dimension estimators have been implemented throughout this work; three well-established methods and a novel implementation of a method. Analytically, these methods correspond to the power spectrum method, the box counting method, the morphological fractal estimator and the novel modification of the kth-nearest neighbour method. The Fuzzy C-Means algorithm is finally applied revealing that the k-th nearest neighbour method outperforms the other methods; thus discriminating up to 93% of the normal parenchyma and up to 82% of the hepatocellular carcinoma, correctly.

C.-P.A. Sariyanni

Papers

A fractal analysis of CT liver images for the discrimination of hepatic lesions: a comparative study