基于锥形束CT图像的肿瘤放疗计划剂量计算可行性分析

随机选取40例(头颈部组、胸部组、腹部组、盆腔组各10例)不同部位的Rapid Arc病例,首先导出所有病例首次治疗时位置验证产生的锥形束电子计算机断层扫描(Cone-beam computed tomography,CBCT)原始投影文件,用自行开发的锥形束CT图像处理工具(Cone-beam CT imaging toolkit,CITK)对CBCT投影进行散射校准并三维重建为0.5 cm层厚的断层CBCT图像;然后用Eclipse软件将计划CT图像与该病例对应的CBCT图像进行配准并将计划CT上勾画的靶区等结构映射到CBCT图像上保存;接着用相同的处方剂量和优化条件设计放疗计划,采用各自的HU-ED标定曲线进行剂量计算后生成剂量体积直方图(Dose volume histogram,DVH)并导出;最后计算所有DVH的剂量分布指数(Dose distribution index,DDI)值,用配对t检验分析两种方法的结果是否存在差异。结果显示,4组病例中只有胸部组病例有统计学差异(t=2.284,p0.05),其余组病例均无统计学差异(p0.05)。实验结果提示,在对CBCT图像进行有效的散射校准并标定对应的HU-ED曲线后,其可用于肿瘤放疗计划的剂量计算。     

Three-dimensional weighting reconstruction algorithm for circular cone-beam CT under large scan angles

Improving imaging quality of cone-beam CT under large cone angle scan has been an important area of CT imaging research. Considering the idea of conjugate rays and making up missing data, we propose a three-dimensional(3D) weighting reconstruction algorithm for cone-beam CT. The 3D weighting function is added in the back-projection process to reduce the axial density drop and improve the accuracy of FDK algorithm. Having a simple structure, the algorithm can be implemented easily without rebinning the native cone-beam data into coneparallel beam data. Performance of the algorithm is evaluated using two computer simulations and a real industrial component, and the results show that the algorithm achieves better performance in reduction of axial intensity drop artifacts and has a wide range of application.     

中间函数法CT图像重建算法研究

CT自从被发明以来其技术已发生了很大的变化.这些变化不仅体现在计算机技术、探测器技术和X线技术方面,同时CT的图像重建算法也在不断地发展.本文提出的中间函数重建算法,是一种可以应用于扇束和锥束扫描的图像重建算法.研究它的目的是为将来的锥束扫描提供直接重建算法.     

基于异构多GPU的锥束CT图像重建研究

针对锥束CT图像重建系统中GPU型号不一致问题,提出了基于异构多GPU的重建模型。该模型基于FDK算法进行重建,采用了按计算能力进行任务分配的方法,确保各GPU计算平衡。采用数据流分解的方法,实现了海量数据的图像重建。给出了该重建模型基于CUDA的实现方法,包括采用流管理和异步函数来实现多GPU并行计算以及滤波和反投影核函数的流程设计。利用高精度工业CT系统进行模型的实验验证。结果表明：所建立的重建模型正确有效,能充分发挥系统中异构多GPU的计算能力,执行效率高。     

基于硬件实现的锥束CT图像重建系统的存储机制设计

为了实现三维锥束CT图像重建加速系统的小型化,建立了基于FPGA的三维图像重建系统。并对该系统中所采用的FDK重建算法所需的数据存储量和数据传输量以及由SDRAM、SRAM和FPGA内部缓存组成的存储系统的数据吞吐率进行了研究。首先,根据FDK算法的滤波与反投影两个步骤介绍了三维锥束CT图像重建系统的数据规模。接着,介绍了一种以SDRAM为外部主存,以SRAM为外部缓存和以FPGA内部SRAM资源作为内部高速缓存的存储机制。然后,介绍了该存储机制的实现方法以及测试方法。最后对该三维图像重建系统的数据吞吐能力进行了测试,并将之与FDK算法所需的数据传输量进行了对比分析。试验结果表明:该存储机制的数据连续存取速度为151.9MB/s,数据随机存取速度为100MB/s,基本满足小规模的三维图像重建的数据存储与传输带宽的要求。     

基于BP神经网络的堆芯三维功率重构方法研究

堆芯功率分布包含了堆芯内的大量信息,由于在反应堆运行过程中无法直接测量堆芯内所有位置的功率,因此需通过其他方法得到堆芯三维功率分布的情况。本文以秦山一期工程为对象,利用堆外中子探测器在不同棒位和不同功率下的计数及BP神经网络对堆芯三维功率分布进行重构计算,并利用REMARK程序对该计算结果进行验证。结果表明,该功率重构方法能在反应堆运行的50%～100%功率范围内,较好地呈现堆芯三维功率分布。     

Modelling radiation-induced phase changes in binary FeCu and ternary FeCuNi alloys using an artificial intelligence-based atomistic kinetic Monte Carlo approach

We apply a novel atomistic kinetic Monte Carlo model, which includes local chemistry and relaxation effects when assessing the migration energy barriers of point defects, to the study of the microchemical evolution driven by vacancy diffusion in FeCu and FeCuNi alloys. These alloys are of importance for nuclear applications because Cu precipitation, enhanced by the presence of Ni, is one of the main causes of hardening and embrittlement in reactor pressure vessel steels used in existing nuclear power plants. Local chemistry and relaxation effects are introduced using artificial intelligence techniques, namely a conveniently trained artificial neural network, to calculate the migration energy barriers of vacancies as functions of the local atomic configuration. We prove, through a number of results, that the use of the neural network is fully equivalent to calculating the migration energy barriers on-the-fly, using computationally expensive methods such as nudged elastic bands with an interatomic potential. The use of the neural network makes the computational cost affordable, so that simulations of the same type as those hitherto carried out using heuristic formulas for the assessment of the energy barriers can now be performed, at the same computational cost, using more rigorously calculated barriers. This method opens the way to properly treating more complex problems, such as the case of self-interstitial cluster formation, in an atomistic kinetic Monte Carlo framework.     

锥束CT重建FDK算法的两级并行计算研究

三维锥束CT图像的FDK算法重建由于运算量大,在重建高分辨率的图像时,重建所需时间通常难以满足实际应用的需求,集群并行计算是解决上述问题的常用方法之一.在一个SMP集群系统上,采用MPI和Pthreads两种模型相结合的方法,通过节点之间的消息传递和节点内部的共享内存,实现了FDK算法的两级并行.     

基于卷积神经网络的辐射图像降噪方法研究

为了抑制探测器中统计涨落引起的噪声,提出了一种基于卷积神经网络的辐射图像降噪方法。该方法利用引入残差网络结构的卷积神经网络模型,对训练集中的辐射图像样本进行了训练,拟合出含噪声图像和无噪声图像的映射关系。实验结果表明,本文方法在降低统计噪声的同时保留了图像的细节。与传统的降噪方法相比,本文方法在量化指标和视觉效果上均有较大的改善。     

An analytical simulation technique for cone-beam CT and pinhole SPECT

This study was aimed at developing an efficient simulation technique with an ordinary PC.The work involved derivation of mathematical operators,analytic phantom generations,and effective analytical projectors developing for cone-beam CT and pinhole SPECT imaging.The computer simulations based on the analytical projectors were developed by ray-tracing method for cone-beam CT and voxel-driven method for pinhole SPECT of degrading blurring.The 3D Shepp-Logan,Jaszczak and Defrise phantoms were used for simulation evaluations and image reconstructions.The reconstructed phantom images were of good accuracy with the phantoms.The results showed that the analytical simulation technique is an efficient tool for studying cone-beam CT and pinhole SPECT imaging.     

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.     

A Monte Carlo method for the model-based estimation of nuclear reactor dynamics

The safe operation and control of a nuclear system requires the accurate estimation of its dynamic state in real time. This can be pursued starting from a model of the system dynamics and on related measurements, which are typically affected by noise. In practice, the nonlinearity of the model and non-Gaussianity of the noise are such that classical approximate approaches, e.g. the extended-Kalman, Gaussian-sum and grid-based filters, often lead to inaccurate results and/or are too computationally expensive for real-time applications. On the contrary, Monte Carlo estimation methods, also called particle filters, can be very effective. The present paper investigates the use of a Monte Carlo method, called sampling importance resampling (SIR), for the estimation of the nonlinear dynamics of a nuclear reactor, as described by a simplified model of literature.     

Derivative-Hilbert-Backprojection based image reconstruction from truncated projections in helical cone-beam CT

In helical cone-beam computed tomography(CT), Feldkamp-Davis-Kress(FDK) based image reconstruction algorithms are by far the most popular. However, artifacts are commonly met in the presence of lateral projection truncation. The reason is that the ramp filter is global. To restrain the truncation artifacts, an approximate reconstruction formula is proposed based on the Derivative-Hilbert-Backprojection(DHB) framework. In the method, the first order derivative filter is followed by the Hilbert transform. Since the filtered projection values are almost zero by the first order derivative filter, the following Hilbert transform has little influence on the projection values, even though the projections are laterally truncated. The proposed method has two main advantages. First, it has comparable computational efficiency and image quality as well as the conventional helical FDK algorithm for non-truncated projections. The second advantage is that images can be reconstructed with acceptable quality and much lower computational cost in comparison to the Laplace operator based algorithm in cases with truncated projections. To point out the advantages of our method, simulations on the computer and real data experiments on our laboratory industrial cone-beam CT are conducted. The simulated and experimental results demonstrate that the method is feasible for image reconstruction in the case of projection truncation.     

Performance evaluation of the backprojection filtered （BPF） algorithm in circular fan-beam and cone-beam CT

In this article we introduce an exact backprojection filtered （BPF） type reconstruction algorithm for cone-beam scans based on Zou and Pan＇s work. The algorithm can reconstruct images using only the projection data passing through the parallel PI-line segments in reduced scans. Computer simulations and practical experiments are carded out to evaluate this algorithm. The BPF algorithm has a higher computational efficiency than the famous FDK algorithm. The BPF algorithm is evaluated using the practical CT projection data on a 450 keV X-ray CT system with a flat-panel detector （FPD）. From the practical experiments, we get the spatial resolution of this CT system. The algorithm could achieve the spatial resolution of 2.4 lp/mm and satisfies the practical applications in industrial CT inspection.     

Filter-based energy-resolved X-ray computed tomography with a clinical imager

Filter-based energy-resolved X-ray computed tomography (CT) is an approach for implementing energy-resolved CT imaging using a flat-panel-detector-based cone-beam system. In this study, we performed experiments with a 20-cm-diameter phantom on a clinical X-ray imager. The material density results showed good agreement with the ideal values. We also propose an improved method for obtaining the detector response function and the X-ray spectrum, which requires fewer measurements and will be practical for future clinical use. Issues such as scatter and image noise remain to be addressed.     

Iterative reconstruction with attenuation compensation fromcone-beam projections acquired via nonplanar orbits

Single photon emission computed tomography (SPECT) imaging with cone-beam collimators provides improved sensitivity and spatial resolution for imaging small objects with large field-of-view detectors. It is known that Tuy's cone-beam data sufficiency condition must be met to obtain artifact-free reconstructions. Even though Tuy's condition was derived for an attenuation-free situation, the authors hypothesize that an artifact-free reconstruction can be obtained even if the cone-beam data are attenuated, provided the imaging orbit satisfies Tuy's condition and the exact attenuation map is known. In the authors' studies, emission data are acquired using nonplanar circle-and-line orbits to acquire cone-beam data for tomographic reconstructions. An extended iterative ML-EM (maximum likelihood-expectation maximization) reconstruction algorithm is derived and used to reconstruct projection data with either a pre-acquired or assumed attenuation map. Quantitative accuracy of the attenuation corrected emission reconstruction is significantly improved     

Recovery of saturated signal waveform acquired from high-energy particles with artificial neural networks

Artificial neural networks(ANNs) are a core component of artificial intelligence and are frequently used in machine learning. In this report, we investigate the use of ANNs to recover the saturated signals acquired in highenergy particle and nuclear physics experiments. The inherent properties of the detector and hardware imply that particles with relatively high energies probably often generate saturated signals. Usually, these saturated signals are discarded during data processing, and therefore, some useful information is lost. Thus, it is worth restoring the saturated signals to their normal form. The mapping from a saturated signal waveform to a normal signal waveform constitutes a regression problem. Given that the scintillator and collection usually do not form a linear system, typical regression methods such as multi-parameter fitting are not immediately applicable. One important advantage of ANNs is their capability to process nonlinear regression problems.To recover the saturated signal, three typical ANNs were tested including backpropagation(BP), simple recurrent(Elman), and generalized radial basis function(GRBF)neural networks(NNs). They represent a basic network structure, a network structure with feedback, and a network structure with a kernel function, respectively. The saturated waveforms were produced mainly by the environmental gamma in a liquid scintillation detector for the China Dark Matter Detection Experiment(CDEX). The training and test data sets consisted of 6000 and 3000 recordings of background radiation, respectively, in which saturation was simulated by truncating each waveform at 40% of the maximum signal. The results show that the GBRF-NN performed best as measured using a Chi-squared test to compare the original and reconstructed signals in the region in which saturation was simulated. A comparison of the original and reconstructed signals in this region shows that the GBRF neural network produced the best performance. This ANN demonstrates a powerful efficacy in terms of solving the saturation recovery problem. The proposed method outlines new ideas and possibilities for the recovery of saturated signals in high-energy particle and nuclear physics experiments. This study also illustrates an innovative application of machine learning in the analysis of experimental data in particle physics.     

锥束CT平行双丝法获取几何参数研究

几何参数的精确获取是锥束CT高质量成像的关键条件,因此锥束CT几何参数的精确获取是CT技术研究的重点之一。本文设计了一种易于制作、操作简便、集成化的模体,提出了锥束CT几何参数获取方法：平行双丝法。首先介绍了平行双丝法获取重要参数的原理及公式推导,通过仿真计算,论证了方法的鲁棒性。然后对该方法进行了实验验证,结果表明,平行双丝法具有参数获取精度高、鲁棒性好等优点,是一种较好的锥束CT几何参数获取方法。     

基于FDK算法的锥束CT重建近似算法性能比较

相比通常的扇束和平行束CT重建,锥束CT有诸多优点(如空间分辨率高,扫描速度快等)。近似锥束重建算法理论结构简单,重建速度快,易于实际应用。对新出现的几种近似锥束重建算法作了简单介绍,并利用这些算法对3D Shepp-logan模型的仿真模拟对这些算法的特点作了比较和讨论。     

Integrating neural network technology and noise analysis

The integrated use of neural network and noise analysis technologies offers advantages not available by the use of either technology alone. The application of neural network technology to noise analysis offers an opportunity to expand the scope of problems where noise analysis is useful and unique ways in which the integration of these technologies can be used productively. The two-sensor technique, in which the responses of two sensors to an unknown driving source are related, is used to demonstration such integration. The relationship between power spectral densities (PSDs) of accelerometer signals is derived theoretically using noise analysis to demonstrate its uniqueness. This relationship is modeled from experimental data using a neural network when the system is working properly, and the actual PSD of one sensor is compared with the PSD of that sensor predicted by the neural network using the PSD of the other sensor as an input. A significant deviation between the actual and predicted PSDs indicate that system is changing (i.e., failing). Experiments carried out on check valves and bearings illustrate the usefulness of the methodology developed.