基于超声射频信号处理的高分辨率心肌弹性成像方法研究

Three million people die of cardiovascular disease (CVD) every year in China, with the second most deaths from coronary heart disease (CHD) in the world. Noninvasive and quantitative measurement of regional myocardial function is of important significance for early diagnosis of myocardial ischeia caused by CHD and assessment of cardiac function. Myocardial elastography is a novel method based on high-frame-rate, ultrasound radio-frequency signal processing, with the advantages of high precision, high spatial and temporal resolutions. In order to solve the current critical issues of myocardial elastography such as low stability, long acuqisiton time, insufficient accuracy and low compuational speed, a new method for high-frame-rate ultrasound radio-frequency data acquisition will be developed to decrease the influence of respiration and arrhythmia and to decease the acquisition time from 15 seconds to 2-3 seconds; the theory and algorithms of myocardial elastography with the use of a phased array probe will be investigated to increase the computational accuracy; fast algorithms will be investigated to increase the compuational speed. In addition, two-dimensional flow imaging will be developed to achieve multi-parametric fusion imaging of B-mode, myocardial elastography and flow imaging for more comprehensive assessment of cardiac function. This project will provide non-invasive and quantitative measurments of regional myocardial function, and signifcantly increase of efficiency of early dection and localization of myocardial ischamia and assemessment of cardiac function, so as to treat the CVD patients in time and decrease the mortality.

我国每年有300万人死于心血管病，其中，冠心病死亡人数已列世界第二位。局部心肌功能的无创、定量测量对冠心病引起的心肌缺血的早期诊断和心功能的评价具有重要意义。心肌弹性成像是一种基于高帧频超声射频信号处理的局部心肌功能测量新方法，具有高精度、高空间和时间分辨率等优点，为解决其目前存在的稳定性差、采集时间长、准确性不足、计算速度低等关键问题，本项目将通过研究新的高帧频超声射频数据采集方法，减小呼吸、心律不齐等影响，实现采集时间从15秒缩短至2-3秒；通过对相控阵下的心肌弹性成像的理论与算法研究，提高计算准确性，通过快速算法的开发，提高计算速度；并进一步开发二维血流成像，实现B超、心肌弹性成像、血流成像的多参数融合，更全面反映心脏功能状况。本项目将实现局部心肌功能的无创、定量测量，显著提高心肌缺血早期检测、定位及心功能评价的效率，从而对心血管疾病患者及时治疗，降低死亡率。

局部心肌功能的无创、定量测量对冠心病引起的心肌缺血的早期诊断和心功能的评价具有重要意义。本项目针对超声心肌弹性成像与血流成像展开研究。首先，建立了基于平面波、球面波、多扫描线并行发射/接收、编码激励、压缩感知的超声波束合成方法，在保证成像质量的同时，获得了高帧频，并实现波束合成的并行计算加速。接着，提出了多种弹性成像的算法，包括基于仿射模型光流法的组织位移估计算法、基于两步光流法的组织应变估计算法、无需正则化的弹性模量重建算法，提高了组织位移估计、应变估计、弹性模量重建的质量，并对弹性成像算法实现了并行计算加速。然后，基于横向振荡方法，建立了超声相控阵矢量血流成像方法；基于多平面二维成像，实现了心脏三维矢量血流成像方法，并利用磁共振血流成像进行了结果验证。最后，通过初步的临床在体实验，验证了研发的波束合成方法、弹性成像算法、血流成像算法的可行性、可靠性和优势。本项目研发的心肌弹性成像与血流成像方法，可望用于局部心肌功能的无创、定量测量，提高心肌缺血早期检测、定位及心功能评价的效率，从而对心血管疾病患者及时治疗，降低死亡率。项目合计发表SCI论文16篇，包括医学影像领域的顶级期刊《IEEE Transactions on Medical Imaging》2篇，医学超声领域的主流期刊《IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control》、《Ultrasound in Medicine & Biology》与《Ultrasonics》9篇、高血压领域的主流期刊《Journal of Hypertension》1篇（影响因子=5）。3名研究生获得IEEE国际超声年会的Student Travel Support Award，1名研究生获得中国生物医学工程联合学术年会 “蒋大宗青年论文竞赛”三等奖。获中国发明专利授权1项，申请中国发明专利2项，申请国际PCT专利1项。