DIXON water-fat separation method in magnetic resonance imaging

Abstract

The invention discloses a DIXON water-fat separation method in magnetic resonance imaging. The method comprises the following steps: a) in the magnetic resonance scanning process, three different echo signals including the first echo signal S1, the second echo signal S2 and the third echo signal S3 are acquired, wherein the water-fat signal procession phase difference of the first echo signal S1 and the third echo signal S3 is 2npi, and n is a natural number; b) the acquired k space signals are converted into image signals through Fourier transform; c) effective image signal pixels are extracted in a image domain, the phase diagram of the effective image signal pixel composite signal is extracted, and phase unwrapping is performed on the phase diagram to obtain a static magnetic field distribution diagram delta B0; d) and the water image and the fat image in the echo signals are separated by utilizing the static magnetic field distribution diagram delta B0. According to the DIXON water-fat separation method provided in the invention, the echo time can be set flexibly according to a magnetic resonance imaging sequence and application requirements, so not only data can be processed by applying the conventional DIXON algorithm, and but also the limitation on imaging sequence parameter setting can be reduced at the same time.

Description

technical field [0001] The invention relates to a water-fat separation method in magnetic resonance imaging, in particular to an improved DIXON water-fat separation method. Background technique [0002] Human MRI (magnetic resonance imaging) signals are mainly derived from two components: water and fat. The chemical bond of hydrogen protons in water molecules is O-H bond, while the chemical bond of hydrogen protons in fat molecules is C-H bond. Due to the difference in the distribution of electron clouds around the hydrogen protons in these two structures, the magnetic field strength felt by the hydrogen protons in the water molecules is slightly higher, and finally the precession frequency of the hydrogen protons in the water molecules is slightly faster than that of the hydrogen protons in the fat molecules Some, the difference is 3.5ppm, equivalent to 150Hz / T, this precession frequency difference increases with the increase of field strength. For example, the field stre...

AI Technical Summary

Problems solved by technology

However, because the binary system involved in water-fat imaging is nonlinear, the least-squares optimization algorithm adopted by IDEAL is prone to errors caused by local minimum interference, resulting in the failure of water-fat separation

Although many improved methods have emerged in recent years to reduce the possibility of errors, they still cannot completely solve the fundamental problem of water-fat separation failure caused by local valley bottoms.

Images