Noise reduction for low-dose X-ray computed tomography with fuzzy filter

Low-dose CT imaging has been particularly used in modern medical practice for its advantage on reducing the radiation dose to patients. However, reconstructed images will distinctly degenerate along with the decrease of the radiation dose. A resolution is to deal with the noisy projection space by an effective filter. This study was performed to address this problem and a fuzzy-median filter was proposed in this paper according to the properties of the noise of low-dose CT images. Reconstructed images by FBP were acquired from the previous noisy sinogram filtered by different filters for comparison. This fuzzy-median filter in fact is a spatially variant one that can solve the streak artifacts. And simulations also indicated that this spatially variant filter could suppress noise and obviously decrease streak artifacts in reconstructed images.     

Noise reduction for low-dose X-ray CT based on fuzzy logical in stationary wavelet domain

Apparent streak-like artifacts will present in reconstructed images due to excessive quantum noise in low-dose X-ray imaging process. Dealing with the noisy sinogram before a filtered back-projection (FBP) is a useful solution to solve this noise problem. In this paper, we proposed a novel noise restoration method combining wavelet and fuzzy logical technology for low-dose computed tomography (CT) sinogram. The method first utilizes stationary wavelet transform on the noisy sinogram, namely decomposes the sinogram to different resolution levels. And then, at each decomposed resolution level, a fuzzy shrinkage filter is applied to restore the noise-contaminated wavelet coefficients. Simulations were performed and indicated that the proposed method could significantly suppress noise and reduced streak-like artifacts in reconstructed images while at the same time maintaining the image sharpness.     

A statistical iteration approach with energy minimization to sinogram noise reduction for low-dose X-ray CT

Low-dose CT imaging has been particularly used in modern medical practice for its advantage on reducing the radiation dose to patients. However, excessive quantum noise is present in low dose X-ray imaging along with the decrease of the radiation dose; thus, there are obvious streak-like artifacts in reconstructed images. The statistical iterative reconstruction approach applied to the noisy sinogram before a filtered back-projection (FBP) is a resolution to deal with the noisy problem. In this paper, the statistical property of the noise sinogram was considered to achieve a satisfactory image reconstruction and a statistical iterative method with energy minimization was proposed to address the problem of streak-like artifacts. Simulations were performed and indicated that the proposed method could suppress noise and obviously decrease streak-like artifacts in reconstructed images.     

Anisotropic total variation minimization approach in in-line phase-contrast tomography and its application to correction of ring artifacts

In-line phase-contrast computed tomography(IL-PC-CT) imaging is a new physical and biochemical imaging method.IL-PC-CT has advantages compared to absorption CT when imaging soft tissues. In practical applications, ring artifacts which will reduce the image quality are commonly encountered in IL-PC-CT, and numerous correction methods exist to either pre-process the sinogram or post-process the reconstructed image. In this study, we develop an IL-PC-CT reconstruction method based on anisotropic total variation(TV) minimization. Using this method, the ring artifacts are corrected during the reconstruction process. This method is compared with two methods: a sinogram preprocessing correction technique based on wavelet-FFT filter and a reconstruction method based on isotropic TV. The correction results show that the proposed method can reduce visible ring artifacts while preserving the liver section details for real liver section synchrotron data.     

CT图像“条状”伪影校正方法研究

在CT重建中,当断层射线衰减系数的变化呈非连续性的跳跃变化,如一种材质密度远远大于另一种材质的密度时,CT图像中高密度物质周围就会产生条状伪影。利用重投影技术分别获取高密度物质和低密度物质的正弦图,低密度物质正弦图中的数据缺失区用线性插值技术进行平滑过渡。利用两幅正弦图分别进行CT重建,然后将两幅CT图像进行相加,得到校正后的图像。实验结果表明,该方法能有效抑制条状伪影,同时提高了图像对比度。     

Beam hardening correction for a cone-beam CT system and its effect on spatial resolution

In this paper, we present a beam hardening correction (BHC) method in three-dimension space for a cone-beam computed tomography (CBCT) system in a mono-material case and investigate its effect on the spatial resolution. Due to the polychromatic character of the X-ray spectrum used, cupping and streak artifacts called beam hardening artifacts arise in the reconstructed CT images, causing reduced image quality. In addition, enhanced edges are introduced in the reconstructed CT images because of the beam hardening effect. The spatial resolution of the CBCT system is calculated from the edge response function (ERF) on different planes in space. Thus, in the CT images with beam hardening artifacts, enhanced ERFs will be extracted to calculate the modulation transfer function (MTF), obtaining a better spatial resolution that deviates from the real value. Reasonable spatial resolution can be obtained after reducing the artifacts. The 10% MTF value and the full width at half maximum (FWHM) of the point spread function with and without BHC are presented.     

硬化伪影的新型表现形式及其校正

X射线硬化是导致工业CT重建图像质量下降的物理原因之一。硬化伪影通常表现为两种形式,即杯状伪影和带状伪影。描述并证实了硬化伪影的一种新型表现形式,这种伪影与真实结构相关,且分布规则,容易造成伪影与真实结构的混淆。采用线性化校正方法对该伪影进行抑制,提高了重建图像质量,改善了通过CT重建图像进行几何测量的精度。     

多通道相位图像的改进型自适应重建算法

自适应重建(Adaptive Reconstruction,AR)算法被广泛应用于磁共振图像的多通道合并问题上.AR算法不需要直接采集各个线圈的灵敏度信息,而是通过通道间信号及噪声相关矩阵,估算出各个通道的灵敏度,从而保证了合并的幅值图像具有较高的信噪比(Signal-to-Noise Ratio,SNR).然而,由于AR算法没有针对相位图像的合并问题进行优化,导致重建出的相位图像具有不确定性.另外,受各通道之间相位偏移及低信噪比相位图像的影响,重建结果可能包含伪影.该文提出了一种改进型AR算法,估算并移除了各通道之间的相位偏移,同时对多通道数据的相位进行质量评估及通道重排,用以进行后续自适应重建.仿体及在体实验表明,该方法可以有效提升AR算法稳定性、消除重建图像中存在的伪影,同时保持合并后幅值图像及相位图像的高信噪比.     

基于自适应正则化的超分辨率重建方法

传统正则化超分辨重建得到的图像往往存在过度平滑或伪信息残留的问题,结合超分辨重建模型对重建图像伪信息的产生进行了分析,针对传统方法的不足提出了基于图像区域信息自适应的正则化方法,通过图像的区域信息将图像划分为平滑区与非平滑区域,对不同区域选用不同的先验模型进行约束。同时考虑人眼的视觉感知特性,结合区域信息实现正则化参数的自适应选取。实验结果表明该方法在抑制重建图像伪信息的同时能有效保护细节,效果要优于传统方法与单一的先验模型约束,对于红外与可见光图像重建效果的提升提供了一定的理论参考。     

扩散噪声下协方差矩阵重构的语音分离与降噪

针对传统多通道语音分离算法在扩散噪声下性能下降的问题,提出了一种用于语音分离及降噪的空间协方差模型及参数估计方法。该方法将扩散噪声视为独立声源,利用由导向矢量重构的空间协方差矩阵建模目标声源的空间特性,并通过空间协方差分析方法估计用于语音分离的多通道维纳滤波器。同时,还提出了一种联合该方法的后置滤波器参数框架,为输出信号降噪和失真的折中提供了更多选择。在扩散噪声下的单目标和多目标实验中,所提方法的语音提取和分离性能都优于对比算法,联合参数的后置滤波器可提供更为符合人们要求的降噪语音,验证了所提模型与参数估计方法的有效性。      

Dictionary learning based sinogram inpainting for CT sparse reconstruction

In CT (computed tomography), reconstruction from undersampling projection data is often ill-posed and suffers from severe artifact in the reconstructed images. To overcome this problem, this paper proposes a sinogram inpainting method based on recently rising sparse representation technology. In this approach, a dictionary learning based inpainting is used to estimate the missing projection data. The final image can be reconstructed by the analytic filtered back projection (FBP) reconstruction. We conduct experiments using both simulated and real phantom data. Compared to the comparative interpolation method, visual and numerical results validate the clinical potential of the proposed method.     

Application of autoregressive modelling in magnetic resonance imaging to remove noise and truncation artifacts

Magnetic resonance imaging data is conventionally reconstructed using two dimensional discrete Fourier transforms. However, there is growing interest in other types of spectral estimation which minimize noise and artifacts due to truncated data. This note presents preliminary results--showing the improvement obtainable using a modified autoregressive model, the Transient Error method.     

The statistical sinogram smoothing via adaptive-weighted total variation regularization for low-dose X-ray CT

Though clinically desired, low-dose X-ray computed tomography (CT) images tend to be degraded by the noise-contaminated sinogram data. Preprocessing the noisy sinogram before filtered back-projection (FBP) is an effective way to solve this problem. This paper presents a statistical sinogram smoothing approach for low-dose CT reconstruction. The approach is obtained by minimizing an energy function consisting of an adaptive-weighted total variation (AWTV) regularization term and a data fidelity term based on the Markov random fields (MRF) framework. The AWTV regularization term can make our algorithm automatically adjust the smoothing degree according to the feature and the level of noise of the smoothed pixel. The experimental results indicate that the proposed approach has the excellent performance in visual effects and quantitative analysis.     

Deep learning for image reconstruction in thermoacoustic tomography

Microwave-induced thermoacoustic tomography(TAT)is a rapidly-developing noninvasive imaging technique that integrates the advantages of microwave imaging and ultrasound imaging.While an image reconstruction algorithm is critical for the TAT,current reconstruction methods often creates significant artifacts and are computationally costly.In this work,we propose a deep learning-based end-to-end image reconstruction method to achieve the direct reconstruction from the sinogram data to the initial pressure density image.We design a new network architecture TAT-Net to transfer the sinogram domain to the image domain with high accuracy.For the scenarios where realistic training data are scarce or unavailable,we use the finite element method(FEM)to generate synthetic data where the domain gap between the synthetic and realistic data is resolved through the signal processing method.The TAT-Net trained with synthetic data is evaluated through both simulations and phantom experiments and achieves competitive performance in artifact removal and robustness.Compared with other state-of-the-art reconstruction methods,the TAT-Net method can reduce the root mean square error to 0.0143,and increase the structure similarity and peak signal-to-noise ratio to 0.988 and 38.64,respectively.The results obtained indicate that the TAT-Net has great potential applications in improving image reconstruction quality and fast quantitative reconstruction.     

Novel correction method for X-ray beam energy fluctuation of high energy DR system with a linear detector

A high energy digital radiography (DR) testing system has generated diverse scientific and technological interest in the field of industrial non-destructive testing. However, due to the limitations of manufacturing technology for accelerators, an energy fluctuation of the X-ray beam exists and leads to bright and dark streak artifacts in the DR image. Here we report the utilization of a new software-based method to correct the fluctuation artifacts. The correction method is performed using a high pass filtering operation to extract the high frequency information that reflects the X-ray beam energy fluctuation, and then subtracting it from the original image. Our experimental results show that this method is able to rule out the artifacts effectively and is readily implemented on a practical scanning system.     

A statistical method for characterizing the noise in nonlinearly reconstructed images from undersampled MR data: The POCS example

The projection-onto-convex-sets (POCS) algorithm is a powerful tool for reconstructing high-resolution images from undersampled k-space data. It is a nonlinear iterative method that attempts to estimate values for missing data. The convergence of the algorithm and its other deterministic properties are well established, but relatively little is known about how noise in the source data influences noise in the final reconstructed image. In this paper, we present an experimental treatment of the statistical properties in POCS and investigate 12 stochastic models for its noise distribution beside its nonlinear point spread functions. Statistical results show that as the ratio of the missing k-space data increases, the noise distribution in POCS images is no longer Rayleigh as with conventional linear Fourier reconstruction. Instead, the probability density function for the noise is well approximated by a lognormal distribution. For small missing data ratios, however, the noise remains Rayleigh distributed. Preliminary results show that in the presence of noise, POCS images are often dominated by POCS-enhanced noise rather than POCS-induced artifacts. Implicit in this work is the presentation of a general statistical method that can be used to assess the noise properties in other nonlinear reconstruction algorithms.     

基于深度神经网络的时空编码磁共振成像超分辨率重建方法

单扫描时空编码磁共振成像是一种新型超快速磁共振成像技术,它对磁场不均匀和化学位移伪影有较强的抵抗性,但是其固有的空间分辨率较低,因此通常需要进行超分辨率重建,以在不增加采样点数的情况下提高时空编码磁共振图像的空间分辨率.然而,现有的重建方法存在迭代求解时间长、重建结果有混叠伪影残留等问题.为此,本文提出了一种基于深度神经网络的单扫描时空编码磁共振成像超分辨率重建方法.该方法采用模拟样本训练深度神经网络,再利用训练好的网络模型对实际采样信号进行重建.数值模拟、水模和活体鼠脑的实验结果表明,该方法能快速重建出无残留混叠伪影、纹理信息清楚的超分辨率时空编码磁共振图像.适当增加训练样本数量以及在训练样本中加入适当的随机噪声水平,有助于改善重建效果.     

A priori information in a regularized sinogram‐based method for removing ring artefacts in tomography

Ring artefacts in X‐ray computerized tomography reconstructions are considered. The authors propose a ring artefact removal method based on a priori information regarding the sinogram including smoothness along the horizontal coordinate, symmetry of the first and the final rows and consideration of small perturbations during acquisition. The method does not require prior reconstruction of the original or corrected sinograms. Its numerical implementation is based on quadratic programming. Its efficacy is examined with regard to experimental data sets collected on graphite and bone.     

基于极大似然模型和期望最大化算法的闪光图像重建

针对闪光照相图像低信噪比的特点,研究了带约束的基于极大似然模型和期望最大化算法（ML-EM）的闪光照相图像重建算法。该算法在ML-EM迭代重建算法的基础上,根据闪光照相图像低信噪比的特点及被重建客体具有分段平滑的先验信息,在迭代重建的过程中进行了相应的噪声抑制,抑制了噪声对重建结果的影响,同时很好地保持了客体的边界特征。数值模拟结果表明,基于噪声约束的ML-EM重建算法能取得较好的重建效果。     

Artifact Reduction Based on Sinogram Interpolation for the 3D Reconstruction of Nanoparticles Using Electron Tomography

Electron tomography is a well‐known technique providing a 3D characterization of the morphology and chemical composition of nanoparticles. However, several reasons hamper the acquisition of tilt series with a large number of projection images, which deteriorate the quality of the 3D reconstruction. Here, an inpainting method that is based on sinogram interpolation is proposed, which enables one to reduce artifacts in the reconstruction related to a limited tilt series of projection images. The advantages of the approach will be demonstrated for the 3D characterization of nanoparticles using phantoms and several case studies.