Super-resolution ultrasound imaging implemented by SOFI in radio-frequency domain

Abstract

Ultrasound (US) is one of the major medical imaging models and has been widely applied in clinical practice. However, the spatial resolution of US is limited to approximately half-wavelength of sound wave. To address this problem, super-resolution ultrasound (SR-US) imaging technique based on single molecule localization has been proposed, which can achieve a ten-fold resolution improvement compared with the conventional US imaging techniques. But, the temporal resolution of SR-US is low. Inspired by super-resolution optical microscopy, recently, a super-resolution optical fluctuation imaging (SOFI) method has been applied in SR-US to improve temporal resolution. However, in the previous work, SOFI is used to process B-mode US sequence (i.e., the image domain), which may affect the obtained imaging performance of SR-US. To further improve the spatial resolution of SR-US, in this work, we propose an alternative method, termed as RF-SOFI, where SOFI is used to process data in radio-frequency (RF) domain. Further, to reduce data acquisition time, here, US data are acquired by plane wave (PW) scan. To evaluate the performance of the proposed RF-SOFI method, numerical simulation experiments were performed. The results indicate that compared to the previous reported SOFI method (i.e., SOFI applied in image domain), the imaging performance of SR-US can be improved by using RF-SOFI (i.e., SOFI applied in RF domain). As a result, RF-SOFI provides the potential in fast SR-US imaging.