Using Graphics Processing Units to Parallelize the FDK Algorithm for Tomographic Image Reconstruction

The paper presents the implementation of a parallel version of FDK （Felkamp, David e Kress） algorithm using graphics processing units. Discussion was briefly some elements the computed tomographic scan and FDK algorithm; and some ideas about GPUs （Graphics Processing Units） and its use in general purpose computing were presented. The paper shows a computational implementation of FDK algorithm and the process of parallelization of this implementation. Compare the parallel version of the algorithm with the sequential version, used speedup as a performance metric. To evaluate the performance of parallel version, two GPUs, GeForce 9400GT （16 cores） a low capacity GPU and Quadro 2000 （192 cores） a medium capacity GPU was reached speedup of 3.37.

Compressed sensing and Otsu's method based binary CT image reconstruction technique in non-destructive detection

This paper tries to address the problem of binary CT image reconstruction in non-destructive detection with an algorithm based on compressed sensing(CS) and Otsu's method, which could reconstruct binary CT image of test object from incomplete detection data. According to binary CT image characteristics, we employ Splitbregman method based on L1/2regularization to solve piecewise constant region reconstruction. To improve the reconstructed image quality from incomplete detection data, we utilize a priori knowledge and Otsu's method as the optimization constraint. In our study, we make numerical simulation to investigate our proposed method,and compare reconstructed results from different reconstruction methods. Finally, the experimental results demonstrate that the proposed method could effectively reduce noise and suppress artifacts, and reconstruct high-quality binary image from incomplete detection data.

Three-dimensional finite element simulation and reconstruction of jointed rock models using CT scanning and photogrammetry

The geometry of joints has a significant influence on the mechanical properties of rocks.To simplify the curved joint shapes in rocks,the joint shape is usually treated as straight lines or planes in most laboratory experiments and numerical simulations.In this study,the computerized tomography (CT) scanning and photogrammetry were employed to obtain the internal and surface joint structures of a limestone sample,respectively.To describe the joint geometry,the edge detection algorithms and a three-dimensional (3D) matrix mapping method were applied to reconstruct CT-based and photogrammetry-based jointed rock models.For comparison tests,the numerical uniaxial compression tests were conducted on an intact rock sample and a sample with a joint simplified to a plane using the parallel computing method.The results indicate that the mechanical characteristics and failure process of jointed rocks are significantly affected by the geometry of joints.The presence of joints reduces the uniaxial compressive strength (UCS),elastic modulus,and released acoustic emission (AE) energy of rocks by 37%–67%,21%–24%,and 52%–90%,respectively.Compared to the simplified joint sample,the proposed photogrammetry-based numerical model makes the most of the limited geometry information of joints.The UCS,accumulative released AE energy,and elastic modulus of the photogrammetry-based sample were found to be very close to those of the CT-based sample.The UCS value of the simplified joint sample (i.e.38.5 MPa) is much lower than that of the CT-based sample (i.e.72.3 MPa).Additionally,the accumulative released AE energy observed in the simplified joint sample is 3.899 times lower than that observed in the CT-based sample.CT scanning provides a reliable means to visualize the joints in rocks,which can be used to verify the reliability of photogrammetry techniques.The application of the photogrammetry-based sample enables detailed analysis for estimating the mechanical properties of jointed rocks.

基于宽体探测器CTP提取4D-CTA最佳重建参数的研究

目的:探讨基于320排640层宽体探测器CT全脑容积灌注数据提取动态四维CT血管成像(4D-CTA)的最佳重建参数。方法:回顾性收集30例行颅脑CT灌注成像(CTP)检查的原始数据,提取动脉峰值相,采用滤波反投影(FBP)和混合迭代重建算法(AIDR 3D)重建层厚/层间距为1/1 mm、0.5/0.5mm、0.5/0.25 mm的薄层图像,分别记为F1、F2、F3、A1、A2、A3组。测量大脑中动脉和及周围脑实质的CT值和噪声(SD),计算信噪比(SNR)和对比度噪声比(CNR)。用最大密度投影(MIP)和实景渲染技术(cVRT)重组血管,2名放射科医生采用5分法评估血管显示的清晰度、锐利度以及图像噪声。采用单因素方差分析、Kruskal-Wallis H检验和组内相关系数(ICC)进行统计学分析。结果:6组客观评价指标中仅CT值差异无统计学意义(P>0.05)。SD以A3组最低,A2组次之,都仅与F1组差异有统计学意义;SNR以A3组最高,A2组次之,A3组与F1组和F2组差异有统计学意义,A2组与F1组差异有统计学意义;CNR以A2组最高,A1组次之,A3组第三(三者差异无统计学意义),都与F1组、F2组、F3组差异有统计学意义。MIP评分和cVRT评分在6组之间差异显著(P<0.05),均是A3组最高,且A3组与A2组之间差异无统计学意义。MIP评分和cVRT评分具有良好的一致性(ICC分别为0.810和0.633)。3种层厚/层间距组合数据量分别为8.74 MB、17.6 MB和35.7 MB。结论:基于320排640层宽体探测器CT全脑容积灌注数据提取4D-CTA图像,重建参数为0.5/0.5 mm或0.5/0.25 mm结合ADIR 3D图像质量最优,综合数据量大小,推荐0.5/0.5 mm联合ADIR 3D的重建组合。

Significance of an Advanced Image-Based Virtual Monoenergetic Reconstruction of Dual Source Dual-Energy CT Data at Low keV Increases Image Quality for Portal Vein System of Pancreatic Cancer Patients

Purpose: To explore the significance of dual-source computed tomography (DECT) virtual monoenergetic reconstructions technology in improving image quality for portal vein system of pancreatic cancer patients. Materials and methods: 47 patients with clinically suspected pancreatic cancer (all confirmed by pathology) were collected. Routine plain scan was performed with Siemens Force dual-source dual-energy CT followed by 3 scans respectively carried out in arterial phase, portal phase and delayed phase. Traditional virtual monoenergetic reconstructions (Mono_E) and new generation of virtual monoenergetic reconstructions (Mono+) were respectively performed on portal vein images to obtain virtual single energy images including Mono_ E70 keV, Mono_E 55 keV and Mono+ 70 keV and Mono+ 55 keV. The signal-to-noise ratio (SNR) and noise of portal vein, normal pancreatic tissues and pancreatic lesions of 100 kV, Mono_E and Mono+ images were compared. In addition, the contrast noise ratio of portal vein and lesions as well as pancreatic tissues and lesions (CNR PV, CNRtumor) were also compared. At the same time, two imaging physicians with rich clinical experiences read the films and scored the images of each group by using the 5-point scoring method. Results: Mono+ 55 keV images including SNRpv, SNRpanc, SNRtumor, Noise, CNRpv, CNRtumor were statistically different from 100 KV images and Mono_E images (P < 0.05). As for the subjective score, Mono+ 55 keV image score also had the highest score, which had statistical significance (P < 0.05). The results showed that Mono+ 55 keV images had the best quality. Conclusion: The new generation of virtual Mono+ post-treatment can reduce image noise. Low energy Mono+ images can improve the contrast between pancreatic cancer lesions and portal of pancreatic cancer patients.

Randomized Kaczmarz algorithm for CT reconstruction

The order of the projection in the algebraic reconstruction technique(ART)method has great influence on the rate of the convergence.Although many scholars have studied the order of the projection,few theoretical proofs are given.Thomas Strohmer and Roman Vershynin introduced a randomized version of the Kaczmarz method for consistent,and over-determined linear systems and proved whose rate does not depend on the number of equations in the systems in 2009.In this paper,we apply this method to computed tomography(CT)image reconstruction and compared images generated by the sequential Kaczmarz method and the randomized Kaczmarz method.Experiments demonstrates the feasibility of the randomized Kaczmarz algorithm in CT image reconstruction and its exponential curve convergence.

Distributed and Consensus Optimization for Non-smooth Image Reconstruction

We present a primal-dual algorithm with consensus constraints that can effectively handle large-scale and complex data in a variety of non-smooth image reconstruction problems.In particular,we focus on the case that the data fidelity term can be decomposed into multiple relatively simple functions and deployed to parallel computing units to cooperatively solve for a consensual solution of the original problem.In this case,the subproblems usually have closed form solution or can be solved efficiently at local computing units,and hence the per-iteration computation complexity is very low.A comprehensive convergence analysis of the algorithm,including convergence rate,is established.

四种品牌CT设备成像质量的研究

目的 该文研究了四种品牌CT设备在采用不同剂量进行扫描且采用两种不同种类重建算法的情况下,四台CT机产生的图像质量的差异,并分析不同品牌厂家的CT设备和重建算法对成像质量产生的影响。方法 本研究使用Catphan500模体作为扫描对象。管电压在120k V的条件下,分别使用临床常规mAs和100m As的扫描剂量下,选择四种品牌高端CT设备对Catphan500模体进行头部的轴位扫描。结果 图像质量对比结果:传统FBP重建、常规扫描剂量下,成像质量排序:256iCT>宝石CT、640CT;迭代重建、常规扫描剂量下,成像质量排序:新双源Force CT>宝石CT、640CT;传统FBP重建、低剂量下,成像质量排序:新双源Force CT>宝石CT;迭代重建、低剂量下,成像质量排序:宝石CT<其它三台CT。迭代算法的影响分析:在低剂量条件下,采用迭代算法与使用传统FBP算法重建相比,这四台CT设备的图像质量有显著性的提高;在常规剂量条件下,采用迭代算法与使用传统FBP算法重建相比,除256iCT以外,其它三台CT图像质量有显著性的提高。结论 在常规剂量的条件下,本研究建议选择西门子新双源Force CT、飞利浦Brilliance-256iCT;在低剂量(100m As)条件下,本研究建议选择飞利浦256iCT、西门子新双源Force CT和东芝640CT;为了获得最佳的成像质量,建议选择迭代算法进行图像重建。

基于最小区域的快速CT图像重建

由于工业CT重建目标的形状差异较大,针对传统CT图像重建算法均选取矩形重建区域,提出一种基于最小区域的图像重建方法.首先由扫描到的投影数据通过直线生成算法构建重建目标的最小区域包络图;然后提出一种快速区域填充算法,生成图像重建的最小区域.该方法将重建区域限定在最小区域内,减少了不必要的计算,提高了重建速度.最后通过仿真实验与传统的重建方法进行比较,表明了文中方法的有效性.

高分辨锥束CT并行重建算法在基于NVDIA GPU显卡计算平台上的实现

目的:探讨高分辨率锥束显微CT断层重建中引入并行计算的必要性及其加速效果。方法:在具有并行计算功能的GPU显卡(NVIDIA QUADRO K5000显卡,显存4G)中为投影图像和重建体数据分配显存空间,每一个像素分配一个线程进行投影图像的各种校正和滤波,再给每个体素分配一线程进行反投影重建,在显存中实现全部断层重建。程序使用C++面向对象方法实现,内核函数用CUDA实现。结果:重建体数据大小是2 048×2 048×128,每个体素用32位浮点数记录,实验采集1 800张投影,每张投影图像大小为2 048×1 536,重建时间小于9 min,是图像采集时间的1/3,是基于CPU重建耗时的2%。将GPU并行重建得到的图像和CPU单线程重建图像结果进行对比,数据结果一致,满足实验设计的要求。结论:并行计算引入高分辨锥束CT重建可大大提高重建速度,并且能实现采集与重建同步进行。

基于特征金字塔网络和密集网络的肺部CT图像超分辨率重建

针对肺部计算机断层扫描(CT)图像的超分辨率(SR)重建中需要加大对肺结节的关注度、满足重建后的特征具有客观存在性等问题,提出一种基于特征金字塔网络(FPN)和密集网络的肺部图像SR重建方法。首先,在特征提取层利用FPN提取特征;其次,在特征映射层设计基于残差网络的局部结构,再用特殊的密集网络连接此类局部结构;再次,在特征重建层利用卷积神经网络(CNN)将不同深度的卷积层逐渐降为图像大小;最后,利用残差网络融合初始低分辨率(LR)特征与重建的高分辨率(HR)特征,形成最终的SR图像。对比实验显示,FPN中2次特征融合和特征映射中5个局部结构连接的深度学习网络效果更佳。所提出的网络相较于超分辨率卷积神经网络(SRCNN)等经典网络重建SR图像的峰值信噪比(PSNR)更高,并且可以获得更好的视觉质量。

从RSNA2012看CT低剂量技术的发展

本文回顾了刚刚结束的第98届北美放射学会年会(RSNA2012)上关于CT低剂量研究和临床的新进展。这些新技术包括各种迭代重建软件,更新型探测器等的推出,大幅度降低患者的辐射剂量,拓宽了CT的临床应用,如全器官CT灌注成像,大范围多期扫描等。合理使用低剂量ALARA原则(As Low As Reasonably Achievable)同样适用于CT检查。

基于可分离替代函数优化的CT重建

针对应用迭代坐标下降算法,且基于q广义高斯马尔科夫随机场正则化模型的计算机断层成像(computed tomography, CT)目标函数优化中存在的耗时过长问题,提出一种基于可分离替代函数的优化方法.根据目标函数特性,采用优化转换策略,将目标函数的优化转换为可分离替代函数的优化,使得待重建CT图像各个像素可以同时更新,再结合图形处理器(graphics processing unit, GPU)技术并行运算能力,最终极大降低了迭代重建所需时间.实验结果表明,与迭代坐标下降算法相比,通过本研究算法重建的CT图像具有相同的空间分辨率和噪声水平,计算耗时仅为迭代坐标下降算法的1/20.本算法在极低的计算耗时下,保持了迭代坐标下降算法优异的图像性能,为临床应用所面临的耗时过长问题提供解决途径.

基于GPU并行计算的X射线动态显微CT的快速重构

上海光源X射线成像线站的X射线动态显微计算机断层扫描(Computed Tomography,CT)的用户实验产生了海量投影数据,采用基于CPU串行计算的CT重构方法速度慢,将导致实验数据的大量积压。图形处理器(Graphics Processing Unit,GPU)因其多核架构、可大规模并行化的特点,在高性能计算领域得到广泛应用,将其用于CT重构,有望实现X射线动态显微CT的快速重构。采用CUDA编程标准,将滤波反投影(Filtered Back Projection,FBP)CT重构算法的反投影运算进行并行化设计,并在NVDIA RTX2080的GPU上成功实现了CT重构的并行计算。与基于CPU串行计算的CT重构时间的对比结果表明:基于GPU并行计算的CT重构获得了200倍左右的加速比,一套CT数据所需的重构时间由分钟级降低到了秒级。借助GPU强大的计算能力可以实现X射线动态显微CT的快速重构。

基于模型的CT三维医学图像重建仿真

2DShepp-Logan头部模型是CT二维医学图像重建领域普遍采用的经典模型。提出一种思路—以3DShepp-Logan头部模型作为CT三维医学图像重建进行仿真实验和算法性能评价的基本参考模型。首先介绍3DShepp-Logan头部模型的设计以及仿真投影数据的计算,进而描述所设计的CT三维医学图像重建仿真计算过程。数值实验部分给出了基于3DShepp-Logan头部模型的CT三维医学图像重建仿真实例。实验结果表明了该思路的可行性和模型计算的准确性。

基于并行计算和多层次B样条的肺部CT-PET图像配准

为了提高医学图像配准的效率,提出一种基于并行计算和多层次B样条函数的图像配准方法。该方法提取图像均匀内部点及轮廓点作为参考图像特征点,采用并行计算技术在浮动图像上选择特征点对;在形变过程采用B样条曲面函数插值来完成,通过多层次B样条函数逐步优化形变效果。实验结果表明,该方法适合于肺器官CT-PET图像快速配准。

ASiR与ASiR-V重建算法对CT图像质量影响的体模研究

目的:通过模型研究,评价自适应统计迭代重建(adaptive statistical iterative reconstruction,ASiR)算法与基于多模型的迭代重建算法ASiR-V对图像质量的影响。方法:以不同管电压(100、120 kVp)、不同管电流(50、100、150、200、250、300、350、400 mA)对腹部体模进行扫描,分别采用滤波反投影(filtered back projection,FBP)、ASiR 50%、ASiR-V 50%算法重建。测量肝脏、左侧肾脏平均CT值、图像噪声、对比噪声比(contrast to noise ratio,CNR),并记录容积CT剂量指数(CT dose index volume,CTDI_(vol))。计算并比较ASiR 50%、ASiR-V 50%算法相比FBP算法降低噪声、提高CNR的能力以及降低辐射剂量的能力。结果:与FBP相比,ASiR 50%可降低图像噪声约30%,提高CNR约43%;而ASiR-V 50%可降低图像噪声约39%,提高CNR约65%。在维持图像质量相近的前提下(CTDI_(vol)SD≈9或10 HU),ASiR 50%与ASiR-V 50%较FBP可分别降低辐射剂量约52%、68%。结论:ASiR 50%与ASiR-V 50%能显著降低CT图像噪声并提高密度分辨力,且ASiR-V 50%降噪能力更为显著,在降低辐射剂量方面更具优势。

腹部动脉能谱CT混合能量与单能量图像的比较研究

目的通过比较腹部动脉能谱CT成像中混合能量重建模式图像与单能量重建模式图像的质量,筛选适宜观察腹部动脉的图像重建模式。方法 80例患者采用GE公司的宝石CT能谱成像技术,获得腹部动脉混合能量图像(140kVp)和40-140keV范围内的单能量图像。对混合能量图像和单能量图像进行感兴趣区CT值的测量,计算腹主动脉、肝动脉的SNR和CNR;各组图像由三名有经验的放射医师采用盲法进行图像质量主观评分。最后进行腹部动脉CT值、SNR、CNR和图像质量主观评分的综合比较。所获数据采用SPSS 19.0软件进行统计学分析。结果随着keV值的降低,腹部动脉CT值逐渐升高,图像噪声同时发生改变。根据腹部动脉图像质量的客观指标(CT值、SNR和CNR)和主观评分,对混合能量图像与各组单能量图像进行综合比较,结果显示50-60keV单能量组腹部动脉图像质量优于混合能量组,同时单能量成像对腹主动脉和肝动脉图像质量的影响存在差别。结论在腹部能谱CT常规增强扫描所获得的腹部动脉不同类型重建图像中,50-60keV单能量重建模式的图像质量优于混合能量重建模式,推荐其作为腹部动脉能谱CT扫描的常规重建模式。

CT快速二维反投影算法

CT图像重建过程中 ,标准的二维反投影运算计算量为O(N^3)。本研究提出一种快速二维反投影算法 ,其计算量仅为O(N^2 log_2~N )。该快速算法可以并行实现 ,处理器阵列规模为O(N^2 )时 ,计算量为O(log_2~N )。本研究还分析得到快速算法的误差上界 ,并提出一种改进的快速二维反投影算法以获得更高的计算精度。最后 ,对算法进行了仿真实验。理论分析及仿真实验结果都表明 ,本研究的二维反投影算法在CT图像重建过程中有着更高的计算效率 ,并且具有良好的计算精度。

等渗低浓度对比剂联合能谱CT在优化颈动脉CTA成像中的应用研究

目的评价等渗低浓度对比剂下能谱单能量联合自适应统计迭代重建(ASiR)算法在优化颈动脉能谱单能量CTA成像中的应用价值,以寻求等渗低浓度对比剂下图像质量最佳的单能量和ASiR比组合,优化颈部血管显示。方法共30例使用等渗低浓度对比剂并经能谱颈动脉CTA扫描的患者入组。一次能谱成像获得11组单能量图像(40~140 keV,间隔10 keV),同时每一组单能量图像分别进行不同ASiR比例(0%,20%,40%,60%,80%和100%)重组。分别对各组单能量图像进行感兴趣区域的SNR值、CNR值及噪声值测定,并由两位高年资影像诊断医生依据血管清晰度行4分法盲法评分。结果 60 keV时CNR和SNR值最高;噪声值随着ASiR比例升高而下降,ASiR 100%时噪声降至最低;图像质量主观评分显示60 keV联合ASiR 60%重建时的评分最高;最后,综合主观及客观评价指标,确定60 keV+ASiR 60%重建参数组合时颈动脉CTA图像质量最佳。结论应用等渗低浓度对比剂行颈动脉能谱CTA扫描,60 keV单能量图像联合ASiR 60%重建时可以明显降低噪声,提高图像质量,最优化显示颈部血管。