基于非晶硒平板探测器的静态数字乳腺断层图像采集研究

发布时间：2018-07-18 15:09

【摘要】：数字乳腺断层成像(digital breast tomosynthesis,DBT)是一项基于数字平板探测器的高级应用技术,是在传统体层成像基础上结合数字图像处理技术的新型体层成像技术。静态数字乳腺断层成像(stationary digital breast tomosynthesis,sDBT)系统,是一种基于多光束X射线源阵列的数字乳腺断层成像系统,采用静态电子扫描的方式实现系统的静态成像,可以有效消除X射线源转动带来的运动伪影,同时缩短扫描成像时间。基于碳纳米管阴极的X射线源具有高时间分辨、可编程控制等优势,多光束X射线源的控制是研究的重点方向。非晶硒数字平板探测器属于直接转换型数字探测器,直接将X射线转换为电信号,具有极高的成像分辨率,足以满足乳腺成像的要求。本文基于AXS-2430非晶硒平板探测器进行静态数字乳腺断层成像系统的图像采集软件开发和图像预处理研究。在Visual Studio 2010软件平台下,利用MFC进行平板探测器的图像采集软件开发,同时设计了针对阵列式多光束碳纳米X射线源的脉冲参数控制软件及相应的控制电路,进行乳腺模体图像采集实验,研究了图像的预处理方法,及实验结果分析。本文主要研究了以下内容:(1)介绍和分析了数字平板探测器的结构和成像原理、碳纳米管X射线源工作原理及静态数字乳腺CT(s-DBT)系统的成像原理。分析目前商用的平板探测器的性能,选用非晶硒平板探测器作为成像媒介。(2)基于MFC和OpenCV完成了探测器图像采集软件和脉冲参数控制软件的开发,并将二者进行集成,实现了图像的采集、存储和实时显示的功能。采用线对卡对探测器分辨率进行初步测试和分析,验证了探测器的成像性能。(3)基于FPGA最小系统设计了阵列式X射线源脉冲控制电路,并结合脉冲参数控制软件进行控制效果测试,实验结果表明,控制电路达到了设计要求。(4)基于实验成像数据初步对投影图像进行了图像预处理研究,包括图像校正、调窗及DICOM转换等,对比分析了不同校正方法的去噪效果,设计出了较为完善的校正算法。
[Abstract]:Digital mammography (digital breast tomosynthesisDBT) is an advanced application technology based on digital flat panel detector. It is a new imaging technology based on traditional tomography and digital image processing. Static digital mammography (stationary digital breast tomosynthesissDBT) system is a digital mammography system based on multi-beam X-ray source array. It can effectively eliminate the motion artifacts caused by the rotation of the X-ray source and shorten the scanning time. The X-ray source based on carbon nanotube cathode has the advantages of high time resolution and programmable control. The control of multi-beam X-ray source is the focus of research. The amorphous selenium digital flat panel detector belongs to the direct conversion type digital detector which converts the X-ray directly into the electric signal and has the extremely high imaging resolution which can satisfy the mammary gland imaging request. Based on the AXS-2430 amorphous selenium flat panel detector, the image acquisition software and image preprocessing of the static digital mammography system are developed in this paper. Under the software platform of Visual Studio 2010, the image acquisition software of flat panel detector is developed by using MFC, and the pulse parameter control software and the corresponding control circuit are designed for the array multi-beam carbon nanocrystalline X-ray source. The preprocessing method of mammary gland motifs was studied and the experimental results were analyzed. The main contents of this paper are as follows: (1) the structure and imaging principle of digital flat panel detector, the working principle of carbon nanotube X-ray source and the imaging principle of static digital mammary CT (s-DBT) system are introduced and analyzed. The performance of the commercial flat panel detector is analyzed, and the amorphous selenium flat detector is selected as the imaging medium. (2) based on MFC and OpenCV, the image acquisition software and pulse parameter control software are developed and integrated. The functions of image acquisition, storage and real-time display are realized. The resolution of the detector is tested and analyzed with a line-pair card, and the imaging performance of the detector is verified. (3) an array X-ray source pulse control circuit based on FPGA minimum system is designed. The experimental results show that the control circuit meets the design requirements. (4) based on the experimental imaging data, the image preprocessing of the projection image is preliminarily studied, including image correction. The noise removal effect of different correction methods is compared and analyzed, and a more perfect correction algorithm is designed.
【学位授予单位】：重庆理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【参考文献】