Remote Measurement of the Ultrasound Contrast Agent Concentration

Abstract

Over the past decade, ultrasound contrast agents (UCA) in the form of tiny gas bubbles were introduced to improve the echographic image quality. The recent research is devoted to exploit the potentiality of new UCA, known as targeted UCA. This contrast agent is able to selectively adhere to cancer cells, as a consequence, the number of attached microbubbles that composes the UCA should be correlated with the status of the cancer, providing in this way useful medical information. In this chapter, a method to derive the microbubble volumetric concentration, working on the signals remotely acquired by means of an ultrasound scanner, is proposed. A preliminary step is devoted to separate the echoes related to microbubbles from that of surrounding tissue, identifying a suitable signal processing technique. A non-linear regression approach, based on the support vector machine, is then considered to estimate the concentration in a region of interest. The training phase is obtained extracting several significant features from simulated signals, while the signals acquired by an echographic scanner represent the test set. The estimation accuracy, obtained from in vitro experiments, seems to be sufficient in order to provide, when coupled with the binding mechanism of the targeted UCA, useful diagnostic information about the vascularization degree, and consequently, about the staging and grading of the pathology.