Segmentation of speckled ultrasound images based on a statistical model

Abstract

In this work, we devise a segmentation scheme aimed at extracting edge information from speckled images using a maximum likelihood edge detector. The scheme was based on finding a threshold for the probability density function of the summing average field over a neighborhood set. A rigorous stochastic analysis was used to derive an exact expression for the cumulative density function of the pdf of the averaging sum image. Based on this, an accurate prob. of error was derived and the performance of the scheme was analysed. The segmentation performed reasonably well for both simulated and clinical images. The importance of this work is the development of a stochastic-based segmentation, allowing an accurate quantification of the prob. of false detection. Non visual quantification and misclassification in medical images is relatively new and is of interest to the clinician.