Evolutionary Feature Construction for Ultrasound Image Processing and its Application to Automatic Liver Disease Diagnosis

Abstract

In this paper, the self organization properties of genetic algorithms are employed to tackle the problem of feature selection and extraction in ultrasound images, which can facilitate early disease detection and diagnosis. Accurately identifying the aberrant features at a particular location of clinical ultrasound images is important to find the possibly damaged tissues. Unfortunately, it is difficult to exactly detect the regions of interest (ROIs) from relatively low quality of clinical ultrasound images. The presented evolutionary optimization algorithm presents a novel approach to building features for automatic liver cirrhosis diagnosis using a genetic algorithm. The extracted features provide several advantages over other feature extraction techniques which include: automatically construct feature set and tune their parameters, ability to integrate multiple feature sets to improve the diagnosis accuracy, and ability to find local ROIs and integrate their local features into effective global features. As compared with past approaches, we span a new way to unify the processing steps in a clinical application using the evolutionary optimization algorithms for ultrasound images. Experimental results show the effectiveness of the proposed method.

Computer-aided Characterization and Diagnosis of Diffuse Liver Diseases Based on Ultrasound Imaging

“…The basic iterative model of GA is an evaluation-selection-reproduction loop. 83 Chi-square feature selection evaluates each attribute with respect to class by providing ranks to attributes regardless of their correlation with each other. For both feature selection and feature extraction, overfitting should be avoided.…”

“…81 Mostly, textural features are calculated from the entire image or ROI using the gray level values. 82,83 The selected ROI should not include artifacts such as blood vessels, costal shadows, and spikes. 84 In Table 3, distinctive and effectively proved textural features are summarized.…”

Particle swarm optimization‐based liver disorder ultrasound image classification using multi‐level and multi‐domain features

“…Most of the classification works in the literature focus on a small subset of disorders and have their ROIs selected manually by an experienced radiologist as a square or a rectangular region. [9][10][11][12][13][14][15] Various approaches to feature extraction have been attempted in the literature to improve the classification accuracy. Texture features and their subsets such as Gray Level Co-occurrence Matrix (GLCM) features, owing to their sensitivity to the textured nature of the ultrasound images, have been extensively analyzed for their class separability capabilities.…”

“…46 Optimization-based approaches, exploiting the selforganizing properties of Genetic Algorithms (GA) founded on evolutionary principles, have found their application in feature selection. Wu et al 14 extract GLCM features to build feature sets indexed by Scale Invariant Feature Transform (SIFT). The dimensionality of the features is balanced, normalized and cascaded and the chromosomes are constructed using intervals of minimum and maximum gradient, and are combined using genetic operators and used in training an SVM.…”

“…techniques only for feature selection. 14,15 The prime focus of this work is to experiment an optimization-based approach for the entire classification phase, which involves the selection of features, classifier, and its related parameters. Selection of the right combination of features, classifier, and its parameters greatly influences the classification accuracy.…”