| Magnetic resonance imaging(MRI)can provide rich and excellent tissue contrast and is a powerful medical imaging technique without ionizing radiation.Threedimensional(3D)MRI has many advantages compared with two-dimensional(2D)MRI,such as larger coverage,higher slice resolution and no slice interval,higher signal-to-noise ratio,three-dimensional reconstruction and arbitrary orientation visulization.However,3D MRI needs to collect much more k-space data than 2D MRI,leading to long scan time and thus not being widely used in clinical practice.This thesis focus on fast 3D MRI and its clinical applications.Specifically,we focus on a novel 3D parallel imaging technique,wave controlled aliasing in parallel imaging(Wave-CAIPI),and conducts in-depth research at multiple levels.For its principle,the relationship between the acceleration performance of Wave-CAIPI and the parameters is studied,and the relationship is analytically derived;a new imaging model is proposed,and the theoretical framework of the correlation matrix is introduced to study the prior knowledge in the new model.For its techniques,the parameter optimization framework of Wave-CAIPI is proposed;an improved reconstruction algorithm is proposed for the VCC-Wave model;For its applications,Wave-CAIPI is applied to two challenging 3D imaging scenes,3D balanced steady state free precession(b SSFP)sequence and highresolution 3D vessel wall imaging(VWI).The main work and contribution of this thesis are as follows:(1)Wave-CAIPI parameter optimization: developing its theory and improving its performanceThrough theoretical calculations,the relationship between the geometric factors and parameters of Wave is derived.It is also verified by simulation experiments.Based on this relationship,a parameter optimization framework is proposed to improve the performance and stability of Wave-CAIPI though parameter optimization,which is superior to that using empirically selected(2)VCC-Wave: combining virtual conjugate coil and wave encodingA novel imaging model(VCC-Wave)is proposed.VCC-Wave can not only combine the advantages of both virtual conjugate coil and wave encoding,but also explore more priors of wave encoding in the new model,thus alleviating wave encoding's limitation in high-resolution and high-bandwidth cases.In addition,the theoretical framework of correlation matrix is introduced to study the priors in the VCC-Wave model.(3)Wave-b SSFP: applying Wave-CAIPI to accelerating 3D b SSFP acquisition We propose the Wave-b SSFP sequence,which applys Wave-CAIPI to accelerating 3D b SSFP sequence acquisition.To deal with the banding artifact caused by nonzero 0-th moment in the conventional wave gradients,we propose a modified wave gradients with truncation.In addition,the proposed Wave-b SSFP is applied to highly accelerated 3D brain imaging,spine imaging and abdominal imaging,which promotes the clinical applications of 3D b SSFP sequence.(4)CS-Wave VWI: applying CS-CAIPI and CS-Wave to accelerating 3D vessel wall imagingThe concept and tool of transform domain geometry-factor is introduced to quantitatively analyze the influence of sampling pattern to high-resolution 3D imaging.Based on the analysis,we apply the compressed sensing technique which adopts CAIPIRINHA sampling(CS-CAIPI)to high-resolution 3D vessel wall imaging(CS-CAIPI VWI),to promote the definition of reconstructed image and the sharpness of vessel wall depiction.Furthermore,we apply CS-Wave,which combines Wave-CAIPI and compressed sensing,to 3D VWI(CS-Wave VWI),to eliminate the residual aliasing artifacts caused by CAIPIRINHA sampling pattern,keeping the image definition and vessel wall sharpness.At an isotropic resolution of 0.6 mm,CS-CAIPI VWI and CS-Wave VWI achieve 11-fold acceleration,reducing the scan time to 3.5 minutes.This paper first systematically investigates the relationship between the performance of Wave-CAIPI and the prorocol parameters,and improves its performance through parameter optimization.And then,an imaging model combining virtual conjugate coil and wave encoding is proposed to alleviate the limitations of Wave-CAIPI in highresolution and high-bandwidth cases.Finally,Wave-CAIPI is applied to accelerate 3D b SSFP and 3D vessel wall imaging. |
PDF Full Text Request