Quantifying Brain Iron Deposition In Patients With Parkinson’s Disease Using QuantitativeSusceptibility Mapping

Iron in the brain, which closely related to synthesis and metabolism of the dopamine, is mostly distributed in some deep gray matter structures. Excessive iron deposition in specific regions of the brain has been observed in many neurodegenerative diseases, such as Parkinson’s disease (PD). Quantitative susceptibility mapping (QSM) is based on the relationship between magnetic medium, magnetic field and phase, reflecting susceptibility contrast in tissue composition, especially for iron-rich gray matter, tissue contrast more obvious. In this dissertation, we present a tentative automatic measurement of susceptibility within regions of interest (ROI) of nuclei, and improved the multi-echo phase fitting method, finally applied the improved method to investigate iron overload in patient with PD.1. Auto quantification of susceptibility in the deep gray matter nuclei of brainRegions of interest (ROI) of the nuclei were drawn manually in previous studies, which were subjective and nagging. In the first part of this paper, we attempted to obtain ROI of the nuclei using auto image registration (AIR) method. This method is a tentative research for auto obtain of ROI, and need further study in the future.2. Phase fitting algorithm on multi-echo quantitative susceptibility mappingGenerally, three dimension multi-echo gradient echo sequence (GRE) is used for raw data acquisition, flowing the weighted least square (WLS) method for fitting phase. Unfortunately, QSM achieved by this algorithm has low signal to noise ratio (SNR) in some regions, especially in the bottom of the brain. We present the truncated-WLS, an improved multi-echo phase fitting method, in the second part of this paper, which adopt appropriate echo number for different tissue and truncate the low SNR signal before fitting phase by WLS. Our study showed that the truncated-WLS method is an effective way to improve the QSM quality, especially in the bottom of the brain.3. Quantifying brain iron deposition in patients with Parkinson’s disease using quantitative susceptibility mappingPD is a progressive neurodegenerative disease which is associated with the degeneration of dopaminergic neurons in the substantia nigra (SN). QSM, in principle, can reflect the iron content more accurately due to the direct calculation of the magnetic susceptibility distribution, whereas, R2* imaging can only reflect the brain iron content indirectly by measuring the change of the magnetic field. In the third part of this paper, we quantified brain iron deposition in patients with PD using QSM and R2* mapping, and evaluated the sensitivities of both mapping methods in clinical diagnoses of PD. Twenty-nine patients with PD and twenty-five healthy controls (HC) were recruited. Two-tailed Pearson correlation, two-tailed t-test and the receiver operating characteristic curves (ROC) was used to data analysis. Neither significant difference between PD patients and healthy controls in the deep gray matter nuclei but SN was found on QSM or R2* mapping. A significant increase on susceptibility of the SN in PD patients than healthy controls (154.80±43.36ppb vs.127.50±21.05ppb, P=0.006) but not on R2* mapping. The ROC results showed that QSM was more sensitive than R2* mapping to classify PD patients from healthy controls, with the area under the curve (AUC) of 0.68 and 0.51, respectively. The UPDRS-Ⅲ motor scores did not correlate with mean susceptibility or R2*values in PD group. Therefore, we demonstrated QSM is a more accurate and sensitive method than R2* mapping to detect the pathologic changes in the SN of patients with PD.