【作者】 王震；

【导师】 刘守斌；

【作者基本信息】 哈尔滨工业大学 ， 机械工程（专业学位）， 2018， 硕士

【摘要】 肿瘤已经成为危害人类健康的主要原因之一,研究肿瘤治疗方法成为了当务之急。基于高强度聚焦超声的热疗以其无毒无创的优越性成为了治疗肿瘤的最重要手段。热疗的关键是实时掌握病灶区的位置和加热温度,避免损伤正常组织细胞和碳化肿瘤细胞。磁共振不仅可以对组织结构进行成像,还可以从图像中提取出隐含的温度信息,实现绿色测温。因此,磁共振引导高强度聚焦超声热疗成为了当今研究的热点之一。本文在分析现有测温原理的基础上,从提高测温的速度和保证测温效果出发进行深入研究。针对现有测温原理的缺点和不足,提出了基于相位差补偿的磁共振测温模型。采用基于水质子共振频率偏移(Proton Resonance Frequency,PRF)的测温原理,将热疗环境、测温序列、磁场漂移以及不均匀性等外界因素纳入到新的测温模型。通过动态相位差处理,极大的降低了运动引起的相位突变。根据实际温度与重建温度的偏差,使用反馈传播算法计算相位差补偿量,并与参考相位相结合计算出相位差修正量。改进测温序列,使用双极性多回波GRE序列进一步提高测温的时间利用率,推导GRAPPA欠采样重建算法,在保证图像图就较高峰值信噪比(Peak Signal to Noise Ratio,PSNR)的前提下,实现数据的加速采集和重建。深入研究了重建图像所存在的噪声特性,采用非局部平均算法进行降噪处理计算处理后的图像数据所隐含的相位差,使用之前获得的相位差修正量对相位差进行修正最终计算出温度,实现磁共振测温。制作实验仿体、设计试验方案,对本文提出的测温模型进行验证。实验结果表明,基于GRAPPA算法的欠采样重建和非局部平均算法降噪处理可以获取较高信噪比的图像。本文测温模型使用双极性多回波GRE序列和常规梯度回波序列,在仿体和离体猪肉上进行测温实验,其最大温度偏差绝对值分别为1.07℃和1.15℃,在保证测温结果的前提上,加快测温的速度,对磁共振引导高强度聚焦超声热疗具有重要进步意义。更多还原

【Abstract】 Tumors have become one of the major causes of endangering human health,and research on the treatment of tumors has become a top priority.Hyperthermia focused ultrasound-based hyperthermia has become the most important treatment for cancer with its non-toxic and non-invasive advantages.The key to hyperthermia is to grasp the location of the lesion area and the heating temperature in real time to avoid damage to normal tissue cells and carbonized tumor cells.Magnetic resonance imaging can not only tissue structure,but also extracted from the image of the hidden temperature information to achieve green temperature measurement.Therefore,magnetic resonance guided highintensity focused ultrasound hyperthermia has become one of the hot topics in the research.In this paper,based on the analysis of the existing temperature measurement principle,this paper makes an in-depth study from the aspects of improving the temperature measurement speed and ensuring the temperature measurement effect.Aiming at the shortcomings and deficiencies of existing temperature measurement principle,a temperature measurement model of magnetic resonance based on curve fitting phase difference compensation is proposed.Based on the principle of temperature measurement based on Proton Resonance Frequency(PRF),the external factors such as hyperthermia environment,temperature measurement sequence,magnetic field drift and inhomogeneity are incorporated into the new temperature measurement model.Through the dynamic phase difference processing,the phase mutation caused by the motion is greatly reduced,and the phase difference compensation calculated and calculated by the curve fitting is deduced and programmed,and the phase difference correction amount is calculated by combining with the reference phase.Improve the temperature measurement sequence,use the bipolar multi-echo GRE sequence to further improve the time utilization of temperature measurement.The GRAPPA under-sampling reconstruction algorithm is deduced.Under the precondition of ensuring peak-to-noise ratio(PSNR)of the image,the data acquisition and reconstruction are accelerated.In-depth study of the noise characteristics of reconstructed images,using non-local average algorithm for noise reduction processing of image data implied by the phase difference.Using the phase difference correction obtained before the phase difference correction to calculate the final temperature,magnetic resonance temperature measurement.Produce the experimental body,design the test program,and verify the temperature measurement model proposed in this paper.Experimental results show that undersampling reconstruction and non-local averaging algorithm based on GRAPPA algorithm can obtain high SNR images.The temperature measurement model in this paper uses the bipolar multi-echo GRE sequence and the conventional gradient echo sequence to perform temperature measurement experiments on the flesh and ex vivo pork.The absolute maximum temperature deviations are 1.07 ℃ and 1.15 ℃ respectively,that speed up the temperature measurement of magnetic resonance guided high intensity focused ultrasound hyperthermia has important significance,under ensuring the temperature measurement results premise.更多还原