频域光学相干层析成像像质增强技术研究
文献类型：学位论文
| 作者 | 黄炳杰 |
| 学位类别 | 博士 |
| 答辩日期 | 2013 |
| 授予单位 | 中国科学院上海光学精密机械研究所 |
| 导师 | 王向朝 |
| 关键词 | 光学相干层析成像 色散补偿 复共轭镜像 散斑 |
| 其他题名 | Research on Image Quality Enhancement Techniques for Fourier Domain Optical Coherence Tomography |
| 中文摘要 | 光学相干层析成像(Optical Coherence Tomography, OCT)是一种非侵入、高分辨率、可在体检测生物组织内部微结构的生物医学光学成像技术。与时域OCT技术相比，频域OCT技术具有高成像速度、高灵敏度的优点，更适合生物组织的在体成像，代表着OCT技术的发展趋势。然而，频域OCT存在色散、固有寄生像、散斑噪声等问题，使其难以满足生物医学对高像质光学成像技术的需求。色散会导致频域OCT系统点扩展函数的展宽和扭曲，降低系统的纵向分辨率。固有寄生像会导致频域OCT有效成像深度范围减半。散斑噪声会降低频域OCT系统的信噪比和灵敏度。本论文针对以上影响频域OCT成像质量的问题，提出几种改进的频域OCT技术，进行理论分析和实验研究。主要工作包括以下几个方面： 1．提出一种基于空间滤波算法的深度分辨色散补偿方法。通过对样品不同深度的干涉信号采用相应的色散系数进行相位补偿，从而精确地补偿样品中不同深度处的色散，提高频域OCT系统的纵向分辨率。该方法和已有色散补偿方法相比，解决了采用统一色散系数进行色散补偿时存在的过补偿和欠补偿问题，色散补偿效果更好。仿真模拟与实验结果表明，采用该方法频域OCT系统的纵向分辨率得到较大幅度提高。 2．提出一种基于空间载波的复频域OCT技术。将空间载波外差干涉术引入频域OCT系统，利用倾斜的参考波在二维频域干涉条纹的横向方向引入空间载波。采用傅立叶变换法提取干涉条纹的相位和幅度信息，进而重建复频域干涉条纹，消除频域OCT的复共轭镜像、直流背景和自相干噪声，最终将有效成像深度范围扩大到原来的两倍，实现全深度探测。与已有的外差干涉复频域OCT技术相比，该技术无需横向和深度方向的机械扫描即可重建一幅全深度的二维层析图像，简化了系统结构，提高了系统稳定性。并且，该技术对物体的运动模糊不敏感，更适合对运动物体成像。对鱼眼和虾尾的成像实验表明复频域OCT系统的复共轭镜像抑制比达到38 dB。 3．将并行频域OCT与角度合成散斑抑制技术相结合，提出一种基于角度合成的并行频域OCT散斑抑制技术。通过面阵CCD一次性采集一幅含有不同横向探测位置和不同背向散射角度信息的二维频域干涉信号，利用空间角度合成方法在抑制散斑噪声的同时获得一幅二维层析图像。与已有的物理散斑抑制方法相比，该技术在不降低纵向分辨率和成像速度的同时，可以提高频域OCT系统的信噪比和灵敏度。对虾尾和珍珠的成像实验表明并行频域OCT系统的信噪比增强因子达到1.76。 |
| 英文摘要 | Optical coherence tomography (OCT) is a promising interferometric imaging technique that can provide micron-scale cross-sectional images of biological tissue in a noninvasive and noncontact way. Fourier domain OCT (FDOCT) has attracted great attention in the past decade due to its high speed and high sensitivity compared with conventional time domain OCT (TDOCT). However, FDOCT suffers from the problems of optical dispersion, complex conjugate artifact, and speckle noise, which deteriorate the image quality. The optical dispersion expands and distorts the point spread function of the FDOCT system, and reduces the axial resolution. The complex conjugate artifact reduces the available imaging depth range to half of the original area. The speckle noise reduces the system’s sensitivity. In view of these problems, several improved FDOCT techniques for enhacing the image quality are presented in this dissertation. These techniques are validated both theoretically and experimentally. The main work of this dissertation is as follows. 1．A depth-resolved dispersion compensation method is proposed for improving the axial resolution in FDOCT. The dispersion of light over the whole imaging depth is effectively compensated through different dispersion coefficients. The proposed method can simultaneously compensate the broadening and distortion of the point spread function in different imaging depth. Thus, the problems of over and under dispersion compensation which exist in the traditional dispersion compension method are soloved. The results of numerical simulations and experiments indicate that the dispersion can be compensated effectively and the axial resolution is improved obviously. 2．A parallel complex FDOCT based on spatial carrier frequency is proposed. A spatial carrier frequency is instantaneously introduced into the two dimensional (2D) spectral interferogram by a grating-generated line reference beam. This spatial-carrier-contained 2D spectral interferogram is Fourier-transformed to obtain a complex 2D spectral interferogram. And then a full range B-scan tomogram is reconstructed from the obtained complex 2D spectral interferogram. Owing to parallel detection, the proposed method is free of motion artifact. Hence, it is more suitable for in vivo imaging. The method is validated by imaging fish anterior chamber and shrimp telson in vivo. The suppression ratio of the complex conjugate artifact can reach to 38 dB. 3．Combining angular compounding technique with parallel FDOCT, a speckle reduction method is proposed. Through a 2D-CCD camera, multiple angle-resolved images are acquired simultaneously with adjacent pixels in parallel detection plane. By compounding these angle-resolved images, a speckle-suppressed B-scan tomogram is obtained without sacrificing axial resolution and imaging speed. The method is validated by imaging pearl and shrimp telson in vivo. A signal-to-noise ratio improvement factor of 1.76 is achieved by compounding four angle-resolved FDOCT images. |
| 语种 | 中文 |
| 源URL | [http://ir.siom.ac.cn/handle/181231/15728]  |
| 专题 | 上海光学精密机械研究所_学位论文 |
| 推荐引用方式 GB/T 7714 | 黄炳杰. 频域光学相干层析成像像质增强技术研究[D]. 中国科学院上海光学精密机械研究所. 2013. |
入库方式： OAI收割
来源：上海光学精密机械研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。