Application of synchrotron source based DEI method in guinea pig cochleae imaging

Hard X-ray diffraction enhanced imaging (DEI), which is based on a synchrotron source and monochromator-analyzer-crystal system, is an effective method for imaging X-ray phase shift. Utilizing an analyzer crystal with high angular sensitivity of micro-radian, DEI can measure the transmitted, refracted and scattered X-rays when projecting onto a sample. It dramatically improves the contrast and spatial resolution of the resultant images. At the topography station of Beijing Synchrotron Radiation Facilities (BSRF), we implemented DEI method in guinea pig cochleae imaging and acquired a series of DEI images. Based on these images, the apparent absorption and refraction images were calculated. The DEI images revealed the holistic spiral structures and inner details of guinea pig cochleae clearly, even including the structures at the cellular level, such as the static cilia of hairy cells and the limbus of Hansen cell. Due to the advantages of high contrast, high spatial resolution and distinct edge-enhanced effect, DEI method promises extensive applications in biology, medicine and clinic in the near future.     

Application of synchrotron source based DEI method in guinea pig cochleae imaging

Hard X-ray diffraction enhanced imaging (DEI), which is based on a synchrotron source and monochromator-analyzer-crystal system, is an effective method for imaging X-ray phase shift. Utilizing an analyzer crystal with high angular sensitivity of micro-radian, DEI can measure the transmitted, refracted and scattered X-rays when projecting onto a sample. It dramatically improves the contrast and spatial resolution of the resultant images. At the topography station of Beijing Synchrotron Radiation Facilities (BSRF), we implemented DEI method in guinea pig cochleae imaging and acquired a series of DEI images. Based on these images, the apparent absorption and refraction images were calculated. The DEI images revealed the holistic spiral structures and inner details of guinea pig cochleae clearly, even including the structures at the cellular level, such as the static cilia of hairy cells and the limbus of Hansen cell. Due to the advantages of high contrast, high spatial resolution and distinct edge-enhanced effect, DEI method promises extensive applications in biology, medicine and clinic in the near future.     

Hard X-ray diffraction enhanced imaging only using two crystals

Different configurations for the monochromator crystals and the analyzer crystals have been used in hard X-ray diffraction enhanced imaging (DEI) methods to overcome the complex task to adjust each of them to the ideal position. Here we present a very compact DEI configuration, and preliminary results of experiments performed at the Beijing Synchrotron Radiation Facility (BSRF) using only two crystals: the first one acting as monochromator and the second one as analyzer in the Bragg geometry. Refraction contrast images characterized by high contrast and spatial resolution are obtained and compared with absorption images. Differences among these images will be outlined and discussed emphasizing the potential capabilities of this very simple layout that guarantees a high transmission efficiency.     

基于灰度共生矩阵的新疆地方性肝包虫CT图像特征提取方法

 特征提取是图像理解与分析的关键。为提取表征新疆地方性肝包虫病的CT影像特征,提出一种基于灰度共生矩阵对肝脏和包虫病灶进行特征提取的方法。首先,对肝脏CT切片图像进行归一化,利用中值滤波和直方图均衡化对肝脏及病灶区同时进行去噪和增强,从而得到更清晰的灰度图像;然后进行灰度级压缩,利用基于灰度共生矩阵的纹理特征提取方法分别提取新疆地方性单囊型、多囊型肝包虫和正常肝脏CT图像的角二阶矩、熵、惯性矩、逆差分矩及相关性的均值和标准差作为纹理特征。统计分析发现,单囊型和多囊型肝包虫CT图像在角二阶矩、熵和逆差分矩等方面存在显著差异,具有统计学意义。最后,采用Bayes判别分类,分类正确率达到93.33%。结果表明,研究采用的纹理提取方法对描述肝包虫CT图像特征具有较理想的效果,一定程度上有助于对肝包虫CT图像进行分类和检索。     

MSCT诊断泌尿系统疾病的价值研究

目的：探讨MSCT应用CTU重建成像在泌尿系统结石、占位及畸形中的技术与应用.方法：对泌尿系统多层螺旋CT扫描的病例91例,扫描包括常规薄层平扫、增强扫描（包括动脉期、静脉期、延迟期）.所有数据传输到Vitrea2后处理工作站进行MPR、CPR、VR及MIP等三维重建.结果：（1）重建图像质量与扫描方法、图像重建方法等密切相关（.2）所有病例图像三维重建后,图像能旋转、切割、放大等清晰显示泌尿系统整体和局部细微结构、形状、走行及与周围组织关系（.3）结石患者可显示结石位置、大小及扩张的肾盂和输尿管;（4）肿瘤者局部管腔内造影剂柱变细或突然中断,管壁不光整;单纯狭窄者管腔呈逐渐变细,管壁无增厚（.5）迷走血管压迫狭窄者均位于肾盂输尿管交界处（.6）重复肾并输尿管畸形者显示双肾盂、双输尿管.结论：采用适当的扫描方法和重建方法,MSCT应用CTU重建成像在泌尿系结石、占位、畸形诊断中与静脉泌尿系造影（IVU）相比有明显优越性,更能满足临床应用.     

螺旋CT三维重建及多平面成像对胸腰段脊柱爆裂骨折的诊断价值探讨

目的：探讨螺旋CT三维重建（CT3D）及多平面重建（MPR）对胸腰段脊柱爆裂骨折诊断的优缺点及临床应用价值。方法分析53例胸腰段脊柱爆裂骨折病例。分别行X线平片，CT及螺旋CT三维重建、多平面重建检查及技术处理。结果：CT对椎体中后柱骨折的显示有明显优越性，而三维重建及多平面重建不仅能具体、直观、形象显示骨折情况，而且对脊髓、神经根损伤的显示价值清楚。结论平片诊断爆裂骨折需与单纯压缩骨折鉴别，明确诊断需CT证实，螺旋CT3D和MPR对诊断及了解椎管内损伤有很好的应用价值。     

螺旋CT三维重建及多平面成像对胸腰段脊柱爆裂骨折的诊断价值探讨

目的探讨螺旋CT三维重建(CT3D)及多平面重建(MPR)对胸腰段脊柱爆裂骨折诊断的优缺点及临床应用价值。方法分析53例胸腰段脊柱爆裂骨折病例。分别行X线平片,CT及螺旋CT三维重建、多平面重建检查及技术处理。结果CT对椎体中后柱骨折的显示有明显优越性,而三维重建及多平面重建不仅能具体、直观、形象显示骨折情况,而且对脊髓、神经根损伤的显示价值清楚。结论平片诊断爆裂骨折需与单纯压缩骨折鉴别,明确诊断需CT证实,螺旋CT3D和MPR对诊断及了解椎管内损伤有很好的应用价值。     

基于变形模型的肝CT序列图像分割

研究了一种基于变形模型的肝轮廓提取方法。该方法以人的经验知识作为先验信息，利用灰度特征、肝CT序列图像特点、肝区轮廓的整体几何信息作为区域聚合依据进行肝区图像分割。实验表明该方法能够较好地克服噪声和轮廓初始位置的影响，对肝CT序列图像的分割效果较好。     

简化的图像三维重建算法

提出了一种简化的图像三维重建算法,该算法利用灰度信息的相似性测度自动寻找两幅图像对应点,再根据双目成像几何模型计算图像的深度信息.给定了相似性测度,两幅图像的匹配可以描述为一个最小化的问题.在实际应用中利用区域对应、唯一性控制等机制以增强算法的精确度和稳定性.     

Longitudinal non-vascular transport pathways originating from acupuncture points in extremities visualised in human body

A subcutaneous transport pathway in humans has been previously visualised by injection of a paramag- netic tracer into the specific region of skin. In the current study, we further investigated the behaviours of imaging tracers in the transport pathways using imaging techniques. Three types of imaging tracers, including Gd-DTPA, iodised oil, and fluorescein sodium, were hypodermically injected into different acupuncture points of healthy volunteers to visualise the transport pathways by magnetic resonance imaging （MRI）, X-ray, CT, and a digital fluorescein camera. Several longitudinal transport pathways in the arms and feet, which represent venous vessels and subcutaneous transport pathways （STPs） without well-defined boundaries, were visualised by MRI. Among the visualised veins, the partial walls of a few veins were enhanced by Gd-DTPA. The fluorescent images further indicated that STPs likely repre- sent boundless pathways. The findings in this study sug- gested that specific anatomical structures within the subcutaneous tissues or the partial walls of certain veins were not vascular conduits but rather represented unique transport pathways in humans that warrant extensive future studies.     

基于正则化的SPACE-RIP并行磁共振成像重建算法

采用并行磁共振成像可以提高成像速度和图像分辨率,但这是以牺牲重建图像的信噪比为代价的。为此,该文提出了一种基于正则化的并行磁共振成像重建算法,降低由于阵列线圈的几何相关而造成的信噪比损失。以预扫描图像作为最后重建图像的先验信息,用L曲线方法求解最佳正则化参数。实验表明,该方法能较大限度地减弱噪声对重建结果的影响,当SPACE-RIP并行成像技术的加速因子达到4时,仍可得到高质量的高分辨率重建影像。重建后的图像质量良好,对临床诊断具有较高实用价值。     

超分辨率数字图像特征提取及重构方法研究

当前超分辨率数字图像特征提取及重构方法容易受到外界环境的干扰,导致重构结果不可靠,重构图像质量较低。为此,提出一种新的超分辨率数字图像特征提取方法,通过BRISK描述子对超分辨率数字图像特征进行提取,以提高重构图像质量。详细分析了重构约束的构建过程;在此基础上,通过低分辨率数字图像与平滑性求解获取高分辨率数字图像,从而实现超分辨率数字图像的重构。实验结果表明,采用所提的新的超分辨率数字图像特征提取及重构方法对图像进行重构,不仅匹配性能高,而且重构图像质量优、效果佳。     

基于正则化的SPACE-RIP并行磁共振成像重建算法

采用并行磁共振成像可以提高成像速度和图像分辨率,但这是以牺牲重建图像的信噪比为代价的。为此,该文提出了一种基于正则化的并行磁共振成像重建算法,降低由于阵列线圈的几何相关而造成的信噪比损失。以预扫描图像作为最后重建图像的先验信息,用L曲线方法求解最佳正则化参数。实验表明,该方法能较大限度地减弱噪声对重建结果的影响,当SPACE-RIP并行成像技术的加速因子达到4时,仍可得到高质量的高分辨率重建影像。重建后的图像质量良好,对临床诊断具有较高实用价值。     

Nuclear magnetic resonance imaging of a single cell

Nuclear magnetic resonance (NMR) imaging is now an established tool in clinical imaging and competes favourably with conventional X-ray computerized tomography (CT) scanning. The drive behind NMR imaging has primarily been in the area of whole-body imaging, which has been limited clinically to fields of up to 1.5 T (60 MHz). It is recognized that there may be substantial advantages in obtaining images with sub-millimetre spatial resolution. Also, there may be benefits to imaging at higher fields, since the signal increases as the square of the magnetic field. Using a modified 9.5 T 89-mm-bore high-resolution NMR spectrometer, we have now obtained the first NMR images of a single cell, demonstrating the advent of the NMR imaging microscope. The NMR microscope is expected to have considerable impact in the areas of biology, medicine and materials science, and may serve as a precursor to obtaining such resolutions on human subjects.     

基于增广拉格朗日的全变分正则化CT迭代重建算法

采用一种基于增广拉格朗日方法(augmented Lagrangian method)求解全变分正则化(total variation regularization)算法(ALMTVR)来进行CT图像重建.将ALMTVR算法与经典的代数重建算法(algebraic reconstruction technique,ART)进行比较,并采用仿真数据与实际数据进行实验.在实验中,使用ALMTVR算法与ART算法分别进行图像重建,并对重建图像进行对比分析.实验结果表明:所提算法与ART算法相比,显著提高了图像重建的质量与速度,显示了其对图像重建的有效性及在CT成像系统中潜在的应用价值.     

基于全场光学相干层析术的猪肝组织亚细胞成像

 建立全场光学相干层析系统(OCT),可对生物细胞进行高分辨率层析成像.该系统基于Linnik 干涉结构,采用低相干白光光源卤钨灯照明和大数值孔径的显微物镜,理论分辨率可达0.7 μm × 0.7 μm(横向× 轴向).系统采用电荷耦合元件(CCD)采集数张样品和参考镜的干涉图,通过三步移相算法获取层析图,最后合成三维图像.与荧光标记细胞等其他生物组织成像的方法相比,全场OCT 成本低廉、对细胞无损伤、分辨率高.该系统能够完成对洋葱表皮细胞和猪肝组织切片细胞等生物组织实现层析成像,且分辨率高、成本低、结构简单便于调节,为实现高分辨率光学相干层析成像提供了简单易行的方法.     

Lensless imaging of magnetic nanostructures by X-ray spectro-holography

Our knowledge of the structure of matter is largely based on X-ray diffraction studies of periodic structures and the successful transformation (inversion) of the diffraction patterns into real-space atomic maps. But the determination of non-periodic nanoscale structures by X-rays is much more difficult. Inversion of the measured diffuse X-ray intensity patterns suffers from the intrinsic loss of phase information, and direct imaging methods are limited in resolution by the available X-ray optics. Here we demonstrate a versatile technique for imaging nanostructures, based on the use of resonantly tuned soft X-rays for scattering contrast and the direct Fourier inversion of a holographically formed interference pattern. Our implementation places the sample behind a lithographically manufactured mask with a micrometre-sized sample aperture and a nanometre-sized hole that defines a reference beam. As an example, we have used the resonant X-ray magnetic circular dichroism effect to image the random magnetic domain structure in a Co/Pt multilayer film with a spatial resolution of 50 nm. Our technique, which is a form of Fourier transform holography, is transferable to a wide variety of specimens, appears scalable to diffraction-limited resolution, and is well suited for ultrafast single-shot imaging with coherent X-ray free-electron laser sources.     

不适定逆成像问题的多准则正则化求解方法

不适定性（ｉｌｌ－ｐｏｓｅｄｎｅｓｓ）是图像重建（Ｘ射线ＣＴ的投影重建、心电ＥＣＧ及电ＥＥＧ信号的逆成像重建）逆问题求解中非常普遍的情况,由于对观测数据及正问题数字离散化过程中所产生的误差极为敏感,它在银行大程度上将影响到重建图像的质量、稳定性和求解精度。通过对由投影重建图像逆问题求解过程中的不适宜性分析,给出了求解不适定逆成像问题的多准则正则化方法及其理论基础,实验结果表明该方法比传统的单准则正     

基于模拟退火的多尺度岩心三维图像融合重建

在岩心图像分析中,低分辨率岩心图像能够显示较大视域,具有较好的全局代表性,而对于低尺度信息无法准确表征;高分辨率岩心图像能够准确地表征岩心的低尺度信息,但通常仅能显示较小的视域.为了能综合分析高低分辨率下的不同岩心图像,本研究提出了一种基于模拟退火,将低分辨率三维岩心图像和高分辨率二维岩心图像融合重建为高分辨率三维岩心的算法.具体地,对于给定的高分辨率二维岩心图像,首先,将低分辨率三维岩心图像进行插值放大以统一两者的点长度,统计二者在二维中的孔隙分布情况,只保留高分辨率二维岩心图像中的小尺寸孔隙,以作为训练图像;然后,在融合重建过程中将低分辨率三维岩心中的大尺寸孔隙相设置为硬数据,以两点相关函数为目标函数重建其中的小尺寸孔隙.实验结果表明,本研究提出的融合重建算法可以很好的将低分辨率岩心重建为高分辨率岩心结构,且融合重建结果有效,准确.     

基于光学相干断层成像含支架的冠脉血管三维重建

光学相干断层成像(OCT)以其具有高分辨率,实时成像等特征,已经成为冠状动脉疾病检测的一个新方法.对含支架的OCT冠脉血管图像进行三维重建研究,以期直观地显示三维空间结构图.通过比较选择合适的降噪算法对图像预处理,根据OCT图像的特性,利用改进的最大类间方差算法(Otsu)对冠脉血管进行分割,并利用一个新的自动算法对冠脉支架进行检测.由于图像存在大量可视支架,在对极坐标图像序列预处理之后,用全局的强度轮廓检测候选的像素,去除伪影,确定支架上下边界和位置,并合并相邻的候选像素为一个支架点,成功标记出检测到的支架点,可有效地排除伪影造成的干扰,提升聚类算法的精确度.在此基础上,利用Amira软件实现了对冠脉血管和支架进行三维重建和融合.