Study On Removal Of Lipid Artifacts In Chemical Exchange Saturation Transfer MRI

Chemical exchange saturation transfer?CEST?MRI has shown great potential in the diagnosis of the brain disease,such as the assessment of ischemic strokes,the detection of tumors,the monitoring of tumor treatment responses,and the evaluation of neurodegenerative diseases.Recently,CEST MRI is increasingly evolving from brain to body applications.Many applications of CEST imaging on the body have also been reported,such as the detection of prostate cancer,the measurement of hepatic glycogen,the detection of breast cancer.Due to the existence of lipid in the body organs,CEST imaging is prone to lipid artifacts,which limits its application on the body organs.In this thesis,three studies were focused on the removal of lipid artifacts in the body CEST MRI.In the first study,the fat-suppression effectiveness of the conventional water-selective binomial-pulse excitation was investigated.The results showed that B₀ inhomogeneity can lead to an insufficient fat suppression.Hence,we further developed a self-adapting multipeak water-fat reconstruction to remove lipid artifacts in CEST imaging in the presence of B₀ inhomogeneity.Furthermore,the first two studies showed that the lipid artifacts were originated from magnetization transfer asymmetry,which is used to extract the CEST effects in the conventional CEST imaging.Hence,in the third study,a new chemical exchange MRI method was proposed to avoid utilizing the magnetization transfer asymmetry to extract the CEST effects.1.Effectiveness of fat suppression using a water-selective binomial-pulse excitation in CEST MRIThe binomial RF pulse has been designed for the water excitation?WE?with an emphasis on the suppression of the main fat peak.The inferior peaks in the fat signals are ignored,which may introduce lipid artifacts into the CEST contrasts.The purpose of this study was to characterize the individual contributions of multiple fat peaks to the CEST signals of exchange sites when using water-selective binomial-pulse excitation in the presence of B₀ inhomogeneity.The excitation profiles of multiple binomial pulses were simulated.A CEST sequence with binomial-pulse excitation and MRS acquisition was then applied to the in vivo lumbar spinal vertebrae in order to determine the signal contributions of the lipid peaks.Numerical simulations and the in vivo experiments showed that water excitation using a 1-3-3-1 pulse?WE-4?provided a broad signal suppression,which produced a partial robustness against B₀ inhomogeneity.In WE-4-based CEST MRI,B₀inhomogeneity is a limiting factor of fat suppression.Generally,artifact-free imaging of amides is viable only when B₀ inhomogeneity is limited in a narraw frequency range,but imaging of hydroxyl and amine sites is significantly more challenging.2.Self-adapting multi-peak water-fat reconstruction for the removal of lipid artifacts in CEST imagingThis study aimed to develop an accurate water-fat reconstruction method based on the multi-echo Dixon technique to remove lipid artifacts in CEST imaging.Considering that RF pulses in CEST preparation saturate each fat peak at different levels and change relative amplitudes of the fat peaks,fat modeling in the water-fat reconstruction is very complicated.Therefore,a self-adapting multi-peak model?SMPM?was proposed to update the relative amplitudes of fat peaks using a Bloch-equation-based numerical calculation.With SMPM-based updating,nonlinear least-squares fitting combined with IDEAL algorithms were used for water-fat reconstruction and B₀ mapping.The proposed method was validated in the phantom study and in the in vivo experiments on human breast.The results show that the SMPM-based method successfully removes lipid artifacts and significantly improves the accuracy of CEST contrast.3.A study on lipid artifacts in stimulated-echo-based chemical-exchange MRIIn the conventional CEST imaging,the net result of the chemical exchange effect is a reduction in the signal intensity of the bulk water.Magnetization transfer asymmetry is a general metric to extract chemical exchange effect.The fat signal in the Z-spectrum is asymmetrical,which leads to lipid artifacts in this metric.In this study,a new chemical exchange MRI method was proposed to avoid utilizing the magnetization transfer asymmetry to extract the chemical exchange effect.In the propsed method,the acquired signal completely arises from the chemical exchange effect,ensuring that the fat can not lead to lipid artifact.In this study,the proposed method was validated by the phantom experiments with salycilic acid as contrast agent.The results from the salycilic-acid phantom demonstate that the image contrast by the proposed method only arised from the chemical exchange effect.Furthermore,the results from a fat phantom showed that fat could not contribute to the MRI signals measured in this method,ensuring a fat-free CEST imaging.