An efficient quantitative ultrasonic elastography based on variational mesh refinement

Abstract

Palpation is often utilised in medicine to detect pathological changes in soft tissue such as those associated with breast cancer or prostate tumor. Ultrasound elastography can be used to measure the stiffness in regions deeper than those reachable by palpation. Conceptually, ultrasound echo signals are measured before and after a defined compression. A displacement field can be calculated with these signals and subsequently, the stiffness distribution can be determined using quantitative elastography. In this regard, the difference between measured und estimated displacement fields is minimised (inverse problem). The region of interest is discretised using the finite element method. In order to reduce the resulting numerical costs and to improve the accuracy of the method, a variational mesh refinement combined with a clustering technique known from digital image compression is elaborated and presented.