Functional Magnetic Resonance Imaging In Patients With Ischemic Stroke

Part I Probing issues of magnetic resonance imaging related to post-processing in ischemic stroke patientsObjective:To explore the impact of lesion occurrence on spatial normalization in the processing of MRI data among acute stroke patients, and to investigate the effect of lesions drawn manually on T2WI/DWI as Lesion_Mask and the segmentation methods for spatial normalization, providing theoretical basis for future functional data analysis.Methods:Among5patients having unilateral ischemic acute stroke within a week, given the situations whether to do the removal of lesions, whether to do segmentations, and lesions drawn on T2WI/DWI respectively, we investigate their effects on spatial normalization through the lesions’ volume, overlapping rate and volume error in the MNI space.Results:(1) By comparing the volume, overlapping rate and volume error of Lesion_Mask in the MNI space (MNI_Lesion_Auto) under different conditions (Lesion_Mask in T2WI/DWI high resolution image, segmentation/non-segmentation), there was no statistically significant (P>0.05) difference on the procedure of spatial normalization with or without brain lesion mask.(2) With or without Lesion_Mask, the volume of MNI_Lesion_Auto by the normalization of non-segmentation model was larger than that of the unified segmentation model. Meanwhile the overlapping rate of MNI_Lesion_Auto by two kinds of normalization models was merely0.25±0.10.(3) With or without segmentation method, the overlapping rate from Lesion_Mask on high resolution DWI was higher than that of high resolution T2WI with statistical significance (P<0.05), which implied that using DWI lesion mask was better.Conclusion:1) There was no obvious difference on normalization results with or without lesion mask in our acute ischemic patient group.2) Lesion_Mask on high resolution DWI was better than high resolution T2WI.3) Two normalization methods, the unified segmentation model and non-segmentation methods, had a significant difference on spatial normalization. We found that the unified model produced significantly better results in terms of both brain lesion anatomical positions and volume. Part II Alterations of brain motor functional connectivity of ischemic stroke patients in the resting state after rTMS treatmentsObjective:To investigate the difference of brain motor functional connectivity of the ipsilesional primary motor cortex(M1) in acute ischemic stroke patients after repetitive Transcranial Magnetic Stimulation (rTMS) treatments with resting-state functional magnetic resonance imaging (rs-fMRI).Methods:Five patients with unilateral acute ischemic stroke were involved in our study after they signed the informed consent form. They were obtained with both functional scoring and resting state magnetic resonance imaging (rs-fMRI) examinations before and after the treatment of high frequency rTMS. Meanwhile, ten healthy volunteers recruited as control group were performed same magnetic resonance imaging (MRI) sequences including rs-fMRI and structure imaging. The time course of ipsilesional M1of stroke patients and the right M1in control group was extracted. Pearson correlation analysis between the time course of M1and the time course of every voxel in the whole brain was performed for acquisition of a map of correlation coefficients, which were transformed by Fisher’s z-transformation and called as z-functional connectivity (z-FC) maps. Two sample t-tests were then conducted to compare z-FC maps between the control group and pre-or post-treatment group. Paired t test were conducted for pre-or post-treatment group.Results:(1) Compared with z values of pre-treatments in patients, the z values of bilateral M1increased in the post-treatment group, while the z values of bilateral M1in pre-treatment patients were significantly lower than the healthy subject’s bilateral M1. However, there was no significant difference in z values between post-treatments’ patients and healthy subjects.(2) Compared with the control group, patients demonstrated higher functional connectivity with bilateral cerebellum between pre-and post-rTMS treatments. In addition, the pre-treatment group showed higher functional connectivity of ipsilesional M1with ipsilesional inferior temporal gyrus, but decreased functional connectivity with contralateral M1and contralesional supplementary motor area. However, the post-treatment group showed higher functional connectivity of ipsilesional M1with ipsilesional middle temporal gyrus, contralesional inferior temporal gyrus, contralesional middle frontal gyrus and contralesional precuneus, but decreased functional connectivity with ipsilesional premotor cortex, ipsilesional M1, contralesional paracenteral lobe and contralesional M1. Compared with the pre-treatment group, higher functional connectivity was found in contralesional frontal medial gyrus. All the functional regions were projected into the3D ICBM templates.(3) Compared with the scores of the pre-treatment group, the NIHSS of the post-treatment group were significantly decreased (p<0.05), while the FMA of upper and lower extremities were significantly increased (p<0.05), and BI values of the post-treatments were also significantly increased (p<0.05), suggesting the functional improvement after treatments.Conclusion:The alterations of functional connectivities between pre-and post-rTMS treatments using rs-fMRI may provide a new approach for revealing the rTMS mechanism in ameliorating the movement function.