基于声速分布评估的无创超声测温成像算法

发布时间：2018-04-10 05:03

本文选题：超声　切入点：测温成像　出处：《深圳大学》2017年硕士论文

【摘要】：传统上,肿瘤的治疗有放疗、化疗以及手术切除等方法。这些方法不仅疗效不足,还对病人产生较大伤害。介入消融手术是一种新型的肿瘤治疗方法,对病人的损伤更小、更安全、治疗效率更高且能适应更多症状,有非常好的应用前景。为了更好的监测介入消融手术的治疗过程,无创的测温成像技术十分关键。超声无创测温成像技术已经有很多的研究,基于声速变化的回波时移测温法是一种有效的方法,但是它难以克服热透镜、热膨胀以及病人运动、呼吸带来的干扰,因而在临床上难以应用。所以,本文提出一种可以克服这些问题的新方法。在介入消融手术中,治疗区域的升温会形成一个局部声速区域,使得下方的成像会变模糊,根据这些现象,本论文提出一种应用于介入消融治疗手术监控的,基于测量局部声速空间分布的超声无创测温成像方法。本文通过前期基于靶点的超声测温实验,验证了从一帧超声通道数据中测量治疗区声速分布的可行性,从而克服既往超声测温算法中“热膨胀”、“热透镜”以及病人运动干扰的影响。为了更加接近临床的情况,后续的进一步研究中,用器官的成像轮廓代替靶点作为感兴趣区域。实验中,将自制的不同声速(1400-1800m/s)仿体与猪肝置于水槽中(水声速约为1492m/s)以采集超声通道数据。其中仿体模拟手术中的治疗区域升温,从而模拟临床介入手术中的热疗情形。论文算法实现的步骤是:在Matlab平台下,将超声通道数据用合成孔径方法重建出基线图像后,自动搜索成像区域中的猪肝轮廓(如血管壁)作为感兴趣区域,并且以声速、横坐标、纵坐标、半径四个参数构造出初始的局部声速区域。然后,按照目前成像区域中的声速分布,对感兴趣区域进行波束合成图像重建,并量化评估其对焦质量。迭代算法不断调整迭代参数以优化对焦质量,这样,构造出来的局部声速区域也会自适应地逐渐逼近其真实情况,最终迭代出真实的声速分布,并根据声速与温度之间的相关性进行测温成像。实验使用中国台湾S-sharp公司的Prodig超声科研平台,采集128个收发通道的数据,超声探头中心频率为6MHz,采用频率25MHz。算法成功地评估出局部声速值及其分布,按本论文的误差定义,局部声速区域的位置评估误差在1.52mm内,热疗程度的评估误差在10%内。所以,该方法很有潜力应用于介入消融手术的无创测温成像监控中。
[Abstract]:Traditionally, cancer treatment includes radiotherapy, chemotherapy and surgical resection.These methods are not only ineffective, but also injurious to patients.Interventional ablation is a new method of tumor treatment, which has a better application prospect with less injury, more safety, higher therapeutic efficiency and more symptoms.In order to better monitor the procedure of interventional ablation, non-invasive thermography is very important.There have been many studies on ultrasonic noninvasive thermography. Echo time-shift thermometry based on the change of sound velocity is an effective method, but it is difficult to overcome the interference of thermal lens, thermal expansion and patient's movement and breathing.Therefore, it is difficult to apply in clinic.Therefore, this paper proposes a new method to overcome these problems.During the interventional ablation, the temperature of the treatment area will form a local sound velocity area, which will make the imaging of the lower part blurred. According to these phenomena, this paper proposes a new method for monitoring the operation of interventional ablation.Ultrasonic non-invasive temperature imaging method based on measuring the spatial distribution of local sound velocity.In this paper, the feasibility of measuring the distribution of sound velocity in the treatment area from a frame of ultrasonic channel data is verified by the previous target based ultrasonic temperature measurement experiment, which overcomes the "thermal expansion" in the previous ultrasonic temperature measurement algorithm.Thermal lenses and the effects of patient motion interference.In order to get closer to the clinical situation, the imaging contour of the organ is used to replace the target as the region of interest in the further research.In the experiment, the self-made imitation of different sound velocities (1400-1800 m / s) and the porcine liver were placed in the water tank (about 1492 m / s) to collect the ultrasonic channel data.The temperature of the treatment area in simulated surgery is simulated, thus the hyperthermia in interventional surgery is simulated.The steps of the algorithm are as follows: after the ultrasonic channel data is reconstructed by synthetic aperture method on the Matlab platform, the pig liver contour (such as vascular wall) in the imaging region is automatically searched as the region of interest, and the sound velocity is used.The initial local velocity region is constructed by four parameters, namely, the transverse coordinate, the vertical coordinate and the radius.Then, the beam-synthesizing image of the region of interest is reconstructed according to the sound velocity distribution in the current imaging region, and its focus quality is evaluated quantitatively.The iterative algorithm constantly adjusts the iterative parameters to optimize the focusing mass, so that the constructed local sound velocity region will gradually approach its real situation adaptively, and finally iterate out the true sound velocity distribution.According to the correlation between sound velocity and temperature, temperature imaging was carried out.The data of 128 transceiver channels were collected by using the Prodig ultrasonic research platform of Taiwan S-sharp Company. The center frequency of ultrasonic probe was 6 MHz and the frequency of ultrasonic probe was 25 MHz.The local sound velocity and its distribution are evaluated successfully. According to the definition of error in this paper, the error of local sound velocity region is in 1.52mm, and the error of hyperthermia is less than 10%.Therefore, this method has the potential to be used in non-invasive thermography monitoring of interventional ablation surgery.
【学位授予单位】：深圳大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R730.5;TB559

【相似文献】