脑深部电刺激（Deep Brain Stimulation，DBS）是一种对多种神经系统疾病有良好疗效的治疗方法，也是目前唯一能够直接干预脑深部神经活动的技术手段。脑深部电刺激器也称为脑起搏器，是一种有前景的研究脑科学的工具。磁共振成像（MRI）是一种广泛应用于临床检查的技术，成像过程无创且无电离辐射，同时也是常用的脑科学研究工具。将脑起搏器与MRI结合起来对于研究DBS的作用机制以及脑科学研究具有重要意义。保证植入脑起搏器的患者在MRI扫查情况下安全以及MRI图像质量，是两者联合进行脑科学研究的前提。随着对脑起搏器磁共振兼容问题认识的深入，当前的标准和规范已无法涵盖所有问题。本研究对MRI与脑起搏器之间的相互作用进行分析，提出磁共振兼容脑起搏器的三个关键问题：（1）脑起搏器在MRI下射频感生电场的电效应，（2）MRI射频致热评估，（3）脑起搏器对MRI图像质量的影响。本研究首先对射频感生电场的电效应问题进行理论分析。通过研究刺激、记录电极的几何特征，分析了电极几何参数对射频感生电场强度的影响规律。直径更小的锥形电极周围的射频感生电场能够达到引起细胞膜穿孔的限度，这种风险应该引起重视。基于这一发现建立了电极设计准则。通过测量脑起搏器电极在3.0 T 磁共振扫查情况下的射频感生电场，发现典型脑起搏器电极的射频感生电场不足以引起细胞膜穿孔效应，并通过动物实验对这一结论进行了验证。射频致热风险是植入脑起搏器的患者进行MRI扫查面对的最大风险。本研究提出了采用电极路径周围局部射频场（B1）均方值预测温升的方法，通过理论推导发现了局部B1均方值与稳态温升之间的线性关系，并通过数值计算以及体模实验验证了该关系。脑起搏器开机状态进行MRI扫查，刺激信号成为新的干扰源。本文研究了脑起搏器刺激电流对功能磁共振（fMRI）的影响，发现该电流不足以引起fMRI图像质量的显著变化。但是发现脑起搏器开机与关机的切换会导致噪声。原因是不同电路连接方式下，MRI射频场在脑起搏器电极中感生电流的幅值不同，导致电极周围B1差异。通过体模实验进行B1测量分析验证了这一理论。修改脑起搏器刺激程序，使两种工作模式下脑起搏器电路连接方式保持不变，该噪声被去除。该方法应用于磁共振兼容脑起搏器导线的临床试验中，为获取高质量fMRI、进行DBS与fMRI联合研究奠定了基础。上述研究为研制磁共振兼容的脑起搏器提供良好的设计、测试理论以及实验基础，为后续工作铺平了道路。

Deep brain stimulation (DBS) is an effective clinical treatment for a variety of neurological diseases. It remains one of the only neurosurgical methods that directly interferes with the activity of deep brain nuclei and holds the potential for propelling brain research. Magnetic resonance imaging (MRI) is a widely used technique for clinical examination. With the advantages in soft tissue imaging, non-invasion and non-ionization, it is commonly used in neuroscience research. The combination of DBS with MRI is of great significance for studying the mechanisms of DBS therapy. This technical combination requires personal safety, equipment safety and the maintained MRI quality in patients with DBS. This study analyzes the use of MRI in DBS, and proposes three key points for the development of MRI compatible DBS. These include (1) the electrical effect of the radio frequency (RF) induced electric field under MRI, (2) the individual risk assessment of RF heating, and (3) the impact of DBS on the quality of MRI.Firstly, we summarized the nerve stimulation and acquisition electrodes with the extraction of basic geometric features. We then studied the influences of electrode geometry parameters on the magnitude of the RF induced electric field. We found that as the electrodes were developed towards higher spatial resolutions, the RF induced electric field of the tapered electrode with a smaller diameter reached the limit of causing cell electroporation. This concluded that the risk should be considered. The RF induced electric field of the current conventional DBS lead in 3.0 T MRI was measured. It was found that it was insufficient to cause cell electroporation, which was validated in animal experimentations.The different placements of DBS has a large impact on RF heating, meaning that individual risk assessment is important to ensure safety. This study proposed a method for predicting temperature rise using the mean square value of local B1 along the lead track. Firstly, based on the dipole antenna model with double-end bare under the lossy medium, it was found that there was a consistent linear relationship between the averaged local B12 and the steady temperature rise with different placements. The data were verified by numerical calculations and phantom experimentations. Based on this theory, methods for pre-obtaining B1 images in a short-time, with low SAR sequence in MRI pre-scan, and predicting the RF heating risk was established.As MRI compatible DBS enters the clinic, it is necessary to evaluate the impact of DBS on the quality of MRI. In this study, the effects of DBS stimulation on functional MRI (fMRI) were studied by theoretical analysis, numerical calculation, phantom experiments and in vivo. It was found that the DBS stimulation current was insufficient to cause significant changes in the image quality. However, DBS in the cycle mode caused fMRI noise. Using theoretical analysis and phantom experiments, the principle of the noise was determined. According to the data, the DBS stimulation program was modified to reduce the noise by decreasing the differences of the RF-induced magnetic field in the cycle mode of DBS. The noise was removed and verified in the phantom experiment and in vivo. This method was applied in a clinical trial of 3.0 T MRI compatible DBS, which lays a foundation for obtaining high quality MRI images and establishing MRI compatible DBS.The research provides a good design, test theory and experimental basis for the development of MRI compatible DBS, paving the way for follow-up work.