|Alternative Title||Microwave-induced Thermo-Acoustic Tomography and its Application in Early Breast Cancer Detection|
|Keyword||微波热声成像 乳腺癌检测 后向投影 Capon波束形成 导向矢量校正|
|Place of Conferral||沈阳|
各种临床统计表明，乳腺癌居女性肿瘤之首，在全球范围内女性乳腺癌的发病例和死亡例每年都保持增长趋势。当今人类对于癌症的治疗效果有限，只能延缓癌细胞恶化时间，唯一提高癌症患者生存率的途径就是早发现并治疗，乳腺癌也是这样。然而，目前用于早期乳腺癌检测的手段都因存在这样或那样的弊端，而致使检测结果不理想。 微波热致超声成像（MITAT, Microwave-Induced Thermo-Acoustic Tomography,本文均简称微波热声成像）技术是基于微波成像和超声成像方法的一种新兴的早期乳腺癌检测手段。该技术使用微波脉冲照射人体乳腺组织，乳腺组织因吸收了微波能而产生超声波，超声波经周围组织向外传播，直至被超声传感器捕捉，根据传感器收集到的数据通过一定的信号处理方法，即可反演乳腺组织内部结构图像。癌变组织与乳腺正常组织因电磁特性的不同而对微波能的吸收存在明显强弱差异，因而产生的超声信号也会有显著区别，于是MITAT技术具有很高的对比度。MITAT最终利用超声信号重建图像，而超声信号的波长较短，故该技术也具有较高的成像分辨率。再加上其无电离辐射、费用便宜、无痛苦等优势，使它被提出时就被认为是一种非常有前景的早期乳腺癌检测技术。 本文首先对MITAT技术的机理进行了深入研究，包括生物组织对微波能的吸收特性、生物组织将微波能转化为超声压力波过程以及超声波在生物组织中的传播过程，从理论上说明了MITAT技术的可行性，并为后续对MITAT在早期乳腺癌检测中的应用提供了理论基石。 在对MITAT理论研究的基础上，本文接着研究MITAT的硬件系统设计问题。重点分析了微波脉冲源重要参数的选择依据，确定了本文成像系统微波源参数：3.0GHz微波频率、脉冲宽度0.4μs、峰值功率60kW、重复频率1Hz。按照MITAT对信号采集的要求，设计了50MHz采样率的信号采集系统，并结合系统控制软件简单介绍了硬件系统各模块同步控制时序。最后，对采集到的超声数据中可能包含的噪声来源进行了深入分析，并逐一给出了处理方法。特别是，将小波分析去噪方法应用于本实验系统中，取得了明显的去噪效果。 接着本文对MITAT技术可用成像算法中后向投影和鲁棒Capon波束形成方法原理进行了研究和推导，分析了它们各自的优势和劣势，总结出它们在早期乳腺癌检测中各自的适用场景。 最后，使用后向投影和鲁棒Capon波束形成方法对实际采集到的超声信号进行了成像实验，图像重建结果表明：1）将MITAT技术作为早期乳腺癌临床检测的手段是合理可行的；2）本文所设计的MITAT成像系统能够有效获取和采集微弱的超声信号；3）在成像质量方面，鲁棒Capon波束形成方法好于后向投影方法，但后向投影算法的实时性好于鲁棒Capon波束形成。
Various clinical statistics show that breast cancer ranks first in the women, and in the global range of breast cancer cases and deaths every year maintain the growth trend. Today, human treatment of cancer is limited, can only delay the deterioration of cancer cells, and the only way to improve the survival rate of cancer patients is early detection and treatment, including breast cancer. However, the current means for early detection of breast cancer are due to the existence of this or that drawbacks, resulting in the test results are not ideal. Microwave-induced Thermo-Acoustic Tomography (MITAT) is a new early detection method of early breast cancer based on microwave imaging and ultrasound imaging. The technology uses microwave pulse to irradiate of human breast tissue, breast tissue produces ultrasound due to its absorption of microwave energy, ultrasound spreads out through the surrounding tissue until it is captured by the ultrasound sensor, we can collect the data according to the sensor, and through a certain signal processing method, we can make an inversion of breast tissue internal structure image. There is a significant difference in the absorption of microwave energy between the cancerous tissue and the normal tissue of mammary gland due to the electromagnetic properties, resulting in that ultrasound signal will be significantly different, and so MITAT technology has a high contrast. MITAT eventually uses the ultrasonic signal to reconstruct the image, while the ultrasonic signal wavelength is short, so the technology also has a high imaging resolution. Coupled with its non-ionizing radiation, cheap cost, no pain and other advantages, it is considered a very promising early breast cancer detection technology since it was put forward. In this thesis, the mechanism of MITAT technology is studied in depth, including the absorption characteristics of biological tissue to microwave energy, the process of biological tissue conversion of microwave energy into ultrasonic pressure wave and the propagation process of ultrasonic wave in biological tissue. Theoretically, MITAT technology is feasible and provides a theoretical basis for the subsequent application of MITAT in early breast cancer detection. Based on the MITAT theory, this thesis then studies the system design of MITAT. After analyze the important parameters selection of the microwave pulse source, the microwave source parameters of 3.0GHz, the pulse width of 0.4us, the peak power of 60kW and the repetition frequency of 1Hz are determined for our imaging system. According to the requirements of MITAT for signal acquisition, a 50MHz sampling rate signal acquisition system is designed, and the synchronization control timing of each module of hardware system is briefly introduced with the system control software. Finally, the sources of noise that may be included in the collected ultrasonic data are analyzed in depth and the processing methods are given one by one. In particular, the wavelet analysis denoising method is applied to the experimental system, and the obvious denoising effect is achieved. Then, this paper studies and deduces the principle of backward projection and robust Capon beamforming in MITAT technology, analyzes their advantages and disadvantages, and summarizes their respective application scenarios in early breast cancer detection. Finally, imaging experiments were carried out using the backward projection and the robust Capon beamforming method. The reconstructed results show that: (1) it is reasonable to use MITAT as a means of clinical detection of early breast cancer; (2) the MITAT imaging system in this paper can effectively acquire and collect the weak ultrasonic signal; (3) The robust Capon beamforming method is better than the backward projection method in terms of image quality, but the real-time performance of the backward projection algorithm is better than that of the robust Capon beamforming.
|王卫锋. MITAT及其在早期乳腺癌检测中的应用研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2017.|
|Files in This Item:|
|MITAT及其在早期乳腺癌检测中的应用研（3140KB）||学位论文||开放获取||CC BY-NC-SA||Application Full Text|
|Recommend this item|
|Export to Endnote|
|Similar articles in Google Scholar|
|Similar articles in Baidu academic|
|Similar articles in Bing Scholar|
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.