Development of an Anatomically Realistic Ultrasound Elastography Radiofrequency Database

This work presents the development of an anatomically realistic ultrasound elastography (USE) database, developed by modifying pre-existing B-mode images from open-source databases. Cooper's ligaments, a curvilinear network of ligaments running through the breast volume, are known to alter the stiffness distribution of the tissue due to their mechanical properties, affecting the tissue displacement field resulting from mechanical stimulation. In this paper, a method is presented for integrating these ligaments in B-mode images, in order to generate anatomically accurate radiofrequency (RF) data. This RF data serves as the foundation for training a convolutional neural network (CNN) designed to automatically detect breast cancer using USE, advancing the potential of this technology in breast cancer diagnosis.