Wavelet Techniques in Medical Imaging: Classification of UltraSound Images using the Windowed Scattering Transform

Abstract

In this thesis we will study wavelet techniques for image classification in ultrasound(US) images. The aim is to develop a method for classifying the degree of inflammation in finger-joints.We develop and apply the techniques of the windowed scattering transform. This is a wavelet-based technique which is proven to be very efficient in image classification problems. Both theoretical and numerical sides have been considered. We also discuss other possible techniques for classification of US images, in particular a method based on the area of inflammation.