Study On Non-Cartesian Spatiotemporally Encoded Magnetic Resonance Imaging

In the field of magnetic resonance imaging,echo planar imaging(EPI)is one of the most commonly used ultrafast imaging methods,but it is very sensitive to inhomogeneous field and chemical shifts.As an ultrafast imaging method,spatiotemporally encoded(SPEN)magnetic resonance imaging method which inherits the advantages of EPI in sampling time and has the strong robustness to inhomogeneous field and chemical shifts.Due to the linear frequency sweep pulse(Chirp pulse),SPEN utilizes quadratic phase encoding in the encoding dimension.Therefore,the SPEN image cannot be reconstructed by conventional Fourier transform.The proposed super-resolution reconstruction greatly improves the spatial resolution of SPEN MRI.Compared with Cartesian sampling,non-Cartesian sampling has natural advantages in sampling efficiency and sampling form.In this article,multi-slice spiral spatiotemporally encoded MRI(MSS-SPEN)is proposed.Combined with the form of storage pulses,the multi-slice image signals are collected by non-Cartesian sampling and reconstructed by the deconvolution super-resolution algorithm.The results show that,compared with the Cartesian multi-slice SPEN MRI,this method greatly shortens the acquisition time and the image has a higher spatial resolution and signal-to-noise ratio.At the same time,we propose the flexible ROI(regions of interest)imaging method,and demonstrate the feasibility of flexible ROI imaging of MS S-SPEN theoretically and experimentally.We can focus the sampling points on the regions of interest,and less or even not on uninteresting area.Combined with the flexibility of non-Cartesian sampling,we can complete the reconstruction of the image by collecting fewer points.Finally,we will discuss the optimization problem of sampling templates.