New probabilistic tissue characterization for ultrasound images implementation published

A new probabilistic tissue characterization for ultrasound images implementation has been published in the Insight Journal: "Probabilistic Tissue Characterization for Ultrasound Images".

Purpose: This document describes the derivation of the mixture models commonly used in the literature to describe the probabilistic nature of speckle: The Gaussian Mixture Model, the Rayleigh Mixture Model, the Gamma Mixture Model and the Generalized Gamma Mixture Model. New algorithms were implemented using the Insight Toolkit
ITK for tissue characterization by means of a mixture model. The source code is composed of a set of reusable ITK filters and classes. In addition to an overview of our implementation, we  provide the source code, input data, parameters and output data that the authors used for validating the different probabilistic tissue characterization variants described in this paper. This adheres to the fundamental principle that scientific publications must facilitate reproducibility of the reported results.
 
 
Conclusions: We have proposed several ITK implementations of mixture models (Rayleigh, Gamma and Generalized Gamma), and several filters used for initialization of the expectation maximization strategy and for tissue characterization. To the best of our knowledge, this is the first open-source implementation of these mixture models and filters within the Insight Toolkit. The design of our implementation tries to follow the design of ITK and thus provides templated N-dimensional filters. The code should be easily integrated to ITK and provide reusable blocks.

ITK source code available at http://hdl.handle.net/10380/3517