Mikio Suga1,2, Hisayuki Miura1, Hiroshi Fujiwara3, Takao Yamamoto4, Takashi Tanaka1, Qiuming YU5, Ken Arai1, Jian Ping GONG5, Gen Nakamura5, Hiroo Ikehira2, Takayuki Obata2
1Graduate School of Technology, Chiba University, Chiba, Japan; 2Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan; 3Graduate School of Informatics, Kyoto University; 4Faculty of Engineering, Gunma University; 5Graduate School of Science, Hokkaido University
Magnetic resonance elastography (MRE) can noninvasively visualize shear waves in tissue. One of the general approaches to estimating stiffness from MRE data is a local frequency estimation (LFE) algorithm, but the LFE estimate is blurred at sharp boundaries. Other methods are based on equations of motion, but they tend to be very sensitive to noise. In this study, we propose a new inversion algorithm that is applicable for noisy long shear wave images and the estimation of the shear modulus quantitatively with high spatial resolution. To confirm the reliability of this method, computer simulations and a phantom study were performed.