Sinogram denoising via attention residual dense convolutional neural network for low-dose computed tomography

The widespread use of computed tomography(CT)in clinical practice has made the public focus on the cumulative radiation dose delivered to patients.Low-dose CT(LDCT)reduces the X-ray radiation dose,yet compromises quality and decreases diagnostic performance.Researchers have made great efforts to develop various algorithms for LDCT and introduced deep-learning techniques,which have achieved impressive results.However,most of these methods are directly performed on reconstructed LDCT images,in which some subtle structures and details are readily lost during the reconstruction procedure,and convolutional neural network(CNN)-based methods for raw LDCT projection data are rarely reported.To address this problem,we adopted an attention residual dense CNN,referred to as AttRDN,for LDCT sinogram denoising.First,it was aided by the attention mechanism,in which the advantages of both feature fusion and global residual learning were used to extract noise from the contaminated LDCT sinograms.Then,the denoised sinogram was restored by subtracting the noise obtained from the input noisy sinogram.Finally,the CT image was reconstructed using filtered back-projection.The experimental results qualitatively and quantitatively demonstrate that the proposed AttRDN can achieve a better performance than state-of-the-art methods.Importantly,it can prevent the loss of detailed information and has the potential for clinical application.

Sinogram Interpolation Method for Sparse-Angle Tomography

In sparse-angle X-ray tomography reconstruction, where only a small number of projection images are taken around the object, appropriate sinogram interpolation has a significant impact on image quality. A novel sinogram interpolation method is introduced for extreme sparse tomographic reconstruction where only nine measured projection images are available. The sinogram is interpolated by solving characteristics of the so-called warps, which can be considered as approximation sine waves in a limited region. The numerical evidence suggests that this approach gives superior results over standard interpolation methods when the tomographic data are extremely sparse and noisy.

Image Noise Analysis of a Large Ring PET Scanner

Image noise analysis of a large ring PET scanner “macro PET” performed using two different phantoms, namely a Jaszczak SPECT phantom and a uniform cylindrical phantom. In the present work, simple 2D filtered back projection was used to reconstruct all the images, and in almost all the cases a Hamming filter of cutoff frequency 0.4 and a 256 by 256 matrix with zoom factors from 1 to 4 were used in order to investigate the imaging capabilities of the new scanner and the influence of filter and cut-off frequency on the filtered back projected images. Results indicate that 11.1 mm cold rod in the Jaszczak phantom images can consistently be seen. The Coefficient of variation (CV) results for Hann and Hamming filters are very similar and increase approximately in linear fashion with higher cutoff frequency. The value of CV for the Parsen filter is lower than the value for Hann and Hamming filters. It concludes that all filters with low cut off-frequency (0.6) would suppress image noise but decrease contrast.

Evaluation of Delivery Analysis to Detect Intrafractional Motion during Tomotherapy

The purpose of this study was to investigate the ability of a management system (Delivery Analysis: DA) to detect intrafractional motion during intensity-modulated radiation therapy (IMRT) in tomotherapy mode. Tomotherapy has made it possible to manage internal movements during treatment using software DA, which quantifies using the information of the passing dose obtained during the radiation treatment of patients. First, three treatment plans for the test were created (lumbar spine, prostate, and femur). Second, a pelvis phantom was moved in the X, Y, and Z directions, and a sinogram was acquired. The magnitudes of the movements were 3 mm, 5 mm, and 10 mm, respectively. Finally, the ability of DA to detect the motion was evaluated by comparing the sinogram obtained by moving the phantom with a reference sinogram obtained without movement. The sensitivity of DA could be detected with a shift amount of 3 mm (gamma analysis tolerance 0.3 mm/0.3%). The average gamma analysis of each direction at 0.3 mm/0.3% tolerance at each treatment site was 96.1% for the prostate, 93.5% for the lumbar spine, and 94.4% for the femur. Additionally, the average gamma pass rate results for the pelvic phantom in the X, Y, Z directions for a 10 mm shift were 96.2%, 96.3%, and 95.9%, respectively. DA is a powerful tool with high detection sensitivity and ability to detect body movement during treatment.

Improved computer-aided detection of pulmonary nodules via deep learning in the sinogram domain

Computer aided detection(CADe)of pulmonary nodules plays an important role in assisting radiologists’diagnosis and alleviating interpretation burden for lung cancer.Current CADe systems,aiming at simulating radiologists’examination procedure,are built upon computer tomography(CT)images with feature extraction for detection and diagnosis.Human visual perception in CT image is reconstructed from sinogram,which is the original raw data acquired from CT scanner.In this work,different from the conventional image based CADe system,we propose a novel sinogram based CADe system in which the full projection information is used to explore additional effective features of nodules in the sinogram domain.Facing the challenges of limited research in this concept and unknown effective features in the sinogram domain,we design a new CADe system that utilizes the self-learning power of the convolutional neural network to learn and extract effective features from sinogram.The proposed system was validated on 208 patient cases from the publicly available online Lung Image Database Consortium database,with each case having at least one juxtapleural nodule annotation.Experimental results demonstrated that our proposed method obtained a value of 0.91 of the area under the curve(AUC)of receiver operating characteristic based on sinogram alone,comparing to 0.89 based on CT image alone.Moreover,a combination of sinogram and CT image could further improve the value of AUC to 0.92.This study indicates that pulmonary nodule detection in the sinogram domain is feasible with deep learning.

“汉字”的译名问题

本文从汉字国际化的背景下讨论'汉字'一名的英译问题。共分三个部分:(1)'汉字'译成Chinese Character带来的问题。论述了这一译名对理解汉语性质、汉字教学、汉语研究带来的影响。(2)'汉字'英译的种种选择。提出了writing,script,graph,zi,Sinogram,Sinograph等几种可能的选择。(3)'汉字'译成Sinogram,Sinograph的合理性及其分工,分别对应于grammar和graphics,实现了语言研究和文字研究的分工。

LIMITED TOMOGRAPHY RECONSTRUCTION VIA TIGHT FRAME AND SIMULTANEOUS SINOGRAM EXTRAPOLATION

X-ray computed tomography （CT） is one of widely used diagnostic tools for medical and dental tomographic imaging of the human body. However, the standard filtered back- projection reconstruction method requires the complete knowledge of the projection data. In the case of limited data, the inverse problem of CT becomes more ill-posed, which makes the reconstructed image deteriorated by the artifacts. In this paper, we consider two dimensional CT reconstruction using the projections truncated along the spatial direc- tion in the Radon domain. Over the decades, the numerous results including the sparsity model based approach has enabled the reconstruction of the image inside the region of interest （ROI） from the limited knowledge of the data. However, unlike these existing methods, we try to reconstruct the entire CT image from the limited knowledge of the sinogram via the tight frame regularization and the simultaneous sinogram extrapolation. Our proposed model shows more promising numerical simulation results compared with the existing sparsity model based approach.