Abstract
In this work, a super-resolution approach based on generative adversary network (GAN) was used to interpolate (up-sample) ultrasound radio-frequency (RF) echo data along the lateral (perpendicular to the acoustic beam direction) direction before motion estimation. Our primary objective was to investigate the feasibility of using a GAN-based super-solution approach to improve lateral resolution in the RF data as a means of improving strain image quality in quasi-static ultrasound strain elastography (QUSE). Unlike natural scene photographs, axial (parallel to the acoustic beam direction) resolution is significantly higher than that of lateral resolution in ultrasound RF data. To better handle RF data, we first modified a super-resolution generative adversary network (SRGAN) model developed by the computer vision community. We named the modified SRGAN model as super-resolution radio-frequency neural network (SRRFNN) model. Our preliminary experiments showed that, compared with axial strain elastograms obtained using the original ultrasound RF data, axial strain elastograms using ultrasound RF data up-sampled by the proposed SRRFNN model were improved. Based on the Wilcoxon rank-sum tests, such improvements were statistically significant ( $p ) for large deformation (3-5%). Also, the proposed SRRFNN model outperformed a commonly-used method ( i.e. bi-cubic interpolation used in MATLAB [Mathworks Inc., MA, USA]) in terms of improving axial strain elastograms. We concluded that applying the proposed (SRRFNN) model was feasible and good-quality strain elastography data could be obtained in in vivo tumor-bearing breast ultrasound data.
Highlights
Qusai-static ultrasound strain elastography (QUSE) [1], [2], which mimics manual palpation, has emerged as an useful imaging tool for non-invasive differentiation of breast and thyroid tumors [3]–[5], assessment of vascular plaques [6] and monitoring thermal ablation therapies [7], [8]
The improvements were greater by using the super-resolution radio-frequency neural network (SRRFNN) model than those obtained by the bi-cubic interpolation approach except at the small deformation (i.e. < 0.5%)
contrast-to-noise ratio (CNR) values obtained from axial strain images were improved when the proposed SRRFNN model was used to up-sample RF data
Summary
Qusai-static ultrasound strain elastography (QUSE) [1], [2], which mimics manual palpation, has emerged as an useful imaging tool for non-invasive differentiation of breast and thyroid tumors [3]–[5], assessment of vascular plaques [6] and monitoring thermal ablation therapies [7], [8]. Motion tracking is performed between the pair of pre- and post-compression echo frames to obtain ultrasonically-tracked tissue displacements. A strain elastogram representing spatial gradients of tissue displacements is formed to infer the hardness of tissues being imaged. These ultrasonically-estimated displacements contain fundamental information and relentless efforts have been devoted to the improvement of displacement tracking accuracy
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
