Abstract
Texture analysis is an important topic in Ultrasound (US) image analysis for structure segmentation and tissue classification. In this work a novel approach for US image texture feature extraction is presented. It is mainly based on parametrical modelling of a signal version of the US image in order to process it as data resulting from a dynamical process. Because of the predictive characteristics of such a model representation, good estimations of texture features can be obtained with less data than generally used methods require, allowing higher robustness to low Signal-to-Noise ratio and a more localized US image analysis. The usability of the proposed approach was demonstrated by extracting texture features for segmenting the thyroid in US images. The obtained results showed that features corresponding to energy ratios between different modelled texture frequency bands allowed to clearly distinguish between thyroid and non-thyroid texture. A simple k-means clustering algorithm has been used for separating US image patches as belonging to thyroid or not. Segmentation of thyroid was performed in two different datasets obtaining Dice coefficients over 85%.
Highlights
Texture analysis is the term used for methods developed to quantify image texture through description of image properties by textural features
We show the usability of the proposed approach for US feature extraction. 2D US data from thyroid is used in order to analyze the capabilities of the 30 extracted energy ratios (ER) features to differentiate between thyroid and non-thyroid tissue in order to use them for segmenting thyroid
The image was transformed into a signal, which was decomposed in several dynamics representing different aspects of the texture
Summary
Texture analysis is the term used for methods developed to quantify image texture through description of image properties by textural features. These features aim to measure smoothness, coarseness, and regularity of pixels, which form an image [1, 2]. Used approaches for computing texture features are based on statistical and frequency domain techniques. Statistical approaches compute histograms, entropy, homogeneity, mean and variance values for estimating features from the texture. Frequency domain techniques or spectral techniques collect a distribution of filter responses to extract different aspects from the texture [1]. Gabor filters and Wavelet decomposition are examples of this type of approach
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
