Comparison of three simulation methods for RF ultrasound signal envelope detection

Abstract

Envelope detection is an important step for B-mode ultrasound (US) image reconstruction, demodulating the received radiofrequency echo signals prior to further processing steps to obtain US images. There are several mathematical methods to perform envelope detection and one of them is through Hilbert transform (HT). In this work, we present the results of computational simulations in MATLAB using Field II software associated with the use of standard Simulink blocks and custom blocks of the Xilinx Model Composer library for field-programmable gate arrays (FPGA) using HT. Qualitative and quantitative means were used to compare US images obtained with the three cited methods. For the quantitative analysis of the proposed models, the following metrics were used: normalized root mean square error (NRMSE), normalized residual sum of squares (NRSS), contrast-to-noise ratio (CNR), and contrast resolution (CR). In conclusion, the proposed methods can satisfactorily simulate echo envelope signals for ultrasound imaging, resulting in very similar images, and are useful tools for high-level project development.