To investigate the cerebral functional changes before and after the repetitive transcranial magnetic stimulation (rTMS) in ischemic stroke patients using the resting-state functional magnetic resonance imaging (rs-fMRI).Thirteen patients with unilateral cerebral subcortex lesion in the middle cerebral artery territory detected by diffusion weighted image (DWI) were enrolled. Eight of them were included in the treatment group and received a ten-day rTMS beginning at about five days after onset. The stimulation involved 50 trains of 20 pulses each day over the ipsilesional M1 at a frenquency of 5HZ, with the stimulus intensity set at 120% of the resting motor threshold of the unaffected extremity. The sham group is consisted of the rest five patients with no rTMS. The treatment group were scanned with rs-fMRI prior to and after the rTMS and the sham received them at the same time point, namely at approximately four days and thirty-one days after onset. Motor functional scores were assessed before every fMRI and included: 1) National Institutes of Health Stroke Scale (NIHSS), 2) Barthel Index(BI), and 3) Fugl-Meyer assessment(FMA). (see Table 1) MRI including rs-fMRI and structural images were performed on a 3.0 Tesla MR imaging System (MAGNETOM Skyra System, Siemens, Erlangen, Germany) by using a twenty-channel phased-array head coil. Preprocessing of the fMRI data was carried out by using Data Processing Assistant for Resting-State fMRI (DPARSF)¹. Pearson correlation analysis between the time course of the ipsilesional M1 and that of every voxel within the whole brain was performed for maps of correlation coefficients, which were Fisher’s z-transformed and called as z-functional connectivity (z-FC) maps. The contrast between the z-FC maps post- and pre- the rTMS treatment (for the sham group: at about thirty-one days and four days after onset) was calculated for each patient. The achieved differential contrast images (∆z-FC maps) were then used for a two-sample t test to compare the FC changes between the rTMS treatment group and the sham group. And it was the same with the statistical analysis of motor functional scores of the two groups.The NIHSS scores of the treatment group were decreased (P<0.05) after the rTMS, while the BI and FMA scores increased (P<0.05). And it was the same with the sham group, which indicated motor improvement. However, there is no significant difference between the treatment and the sham group. Compared with the sham group, the rTMS treatment group showed significantly increased functional connectivity between the ipsilesional M1 and the bilateral thalamus, the contralesional postcentral gyrus, the contralesional supplementary motor area (SMA), the contralesional middle occipital gyrus, the bilateral superior temporal gyrus, the ipsilesional pallidum and the contralesional Heschl gyrus. And decreased functional connectivity of the ipsilesional M1 was demonstrated in the ipsilesional inferior and middle temporal gyrus, the ipsilesional inferior and middle frontal gyrus, the contralesional superior temporal gyrus and the contralesional middle and superior frontal gyrus.(see Figure 1-2)To the best of our knowledge, this is the first study to evaluate the cerebral functional changes after rTMS in ischemic stroke patients using the rs-fMRI. Our research is a longitudinal study instead of a cross-sectional one. As we all know, stroke develops over a certain time course and the motor recovery is a dynamic process^2-4, from which aspect the longitudinal study is better. What’s more, our research includes stroke patients in the acute stage only, which can minimize the bias caused by the course of the disease. In addition, former study demonstrated that high-frequency rTMS over ipsilesional M1 improved movement kinematics in most of the patients with subcortical stroke, but not in patients with cortical stroke⁵. Only subcortical stroke patients are included in our research, which on one side improved the effectiveness of rTMS and on the other hand excluded the probable influence of the cerebral cortex to motor function. Motor function of the rTMS group approved after the treatment, assessed by the behavioral testing. However, there is no significant difference of motor recovery between the treatment and the sham group. We thought that it may be due to the small sample size of our research to some degree. The previous study of Mitra Ameli⁵ showed that facilitary rTMS over ipsilesional M1 caused a significant motor improvement of the affected hand in patients with subcortical stroke. They conceived that the high-frequency (10HZ) rTMS reduced the neural activity in the contralesional hemisphere for movements of the affected hand in subcortical stroke. Changes of functional connectivity pre- and post- rTMS provide us with a new approach for revealing its mechanism in improving motor function in stroke patients.