Translator Disclaimer
20 March 2014 Minimum variance image blending for robust ultrasound image deconvolution
Author Affiliations +
When the ultrasound wave propagates the human body, its velocity and attenuation change at each region, which make the PSF shape different. To solve the PSF estimation problem is ill-posed case and rarely error free which produces the PSF estimation errors and make the image overblurred by the sidelobe artifacts. For the commercialization of the ultrasound deconvolution method, the robustness of the image deconvolution without artifacts is essential. There exist many minimum variance beamformer algorithms. It is robust to noise and shows high resolution efficiently. We consider the channel data as image pixel and we present a new spatial varying MV (minimum variance) blending scheme with the deconvolved imageges in the image processing domain. With a stochastic image blending of the deconvolution images, we obtain high resolution results which suppress the blur artifacts enough although the input deconvolution images have restoration errors. We verify our algorithm on the real data. In all the case, we can observe that the artifacts are suppressed and show the highest resolution among the deconvolution methods.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Sungchan Park, Jooyoung Kang, Yun-Tae Kim, Kyuhong Kim, Jung-Ho Kim, and Jong Keun Song "Minimum variance image blending for robust ultrasound image deconvolution", Proc. SPIE 9040, Medical Imaging 2014: Ultrasonic Imaging and Tomography, 90400E (20 March 2014);

Back to Top