SWShijie Wang

Shijie Wang

*This is an algorithmically generated author record based on publicly available information.

Shijie Wang has a total of 53 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Data-consistent Unsupervised Diffusion Model for Metal Artifact Reduction

    by Zhan Tong, Zhan Wu, Yang Yang, Weilong Mao, Shijie Wang, Yinsheng Li, Yang Chen

    Computed Tomography (CT) is an imaging technique widely used in clinical diagnosis. However, high-attenuation metallic implants result in the obstruction of low-energy Xrays and further lead to metal artifacts in the reconstructed CT images. Deep supervised model-based metal artifact reduction(MAR) approaches are limited in clinical applications due to the difficulty in obtaining paired artifact-affected and artifactfree data. Furthermore, these model-based methods lack the consideration of data consistency in the sinogram-domain to perform exact metal trace inpainting. To address these challenges, we propose a Data-consistent unsupErVised diffusiOn model for meTal artifact rEDuction, called DEVOTED-Net. First, DEVOTED-Net leverages prior knowledge to guide the conditional diffusion model for fine-grained metal trace inpainting. Second, an unsupervised MAR framework is designed in the reverse process for the unknown metal traces restoration in the sinogram domain. Third, to further enhance the sinogram-domain data consistency, physics-based consistency constraint loss including conjugateray consistency loss and accumulation-ray consistency loss is designed. Extensive experiments are carried out to verify the performance of our algorithm on the publicly available dataset and clinical experimental dataset. This efficient, accurate, and reliable MAR approach holds great potential in clinics.


  • Cas-PRED: A Limited-Angle Cone Beam CT Reconstruction Research Based on Prior Information

    by Yuan Gao, Liu Zhang, Wenxue Yu, Shijie Wang, Pinzheng Zhang, Cheng Xue, Chunfeng Yang, Shipeng Xie, Gouenou Coatrieux, Jean-Louis Coatrieux, et al.

    Due to limitations in radiation dose or scanning conditions, there are instances where CT data can only be collected within a restricted scanning angle. However, the use of limited-angle projection data often results in reconstructed images that exhibit significant noise and artifacts, which can significantly impact diagnostic accuracy. Conventional iterative reconstruction algorithms, such as the total variation algorithm, consume considerable time simulating the cone beam forward and backward projection processes. Furthermore, these algorithms often struggle to address the challenges of limited-angle reconstruction due to their inherent limitations. To tackle these challenges and achieve superior reconstruction outcomes, we introduce a novel approach known as the Cascaded Prior-Based Residual Encoder-Decoder Network (Cas-PRED). This method employs a multi-level cascaded network as its core architecture, leveraging full-angle CT reconstruction results as prior information to guide the deep learning network’s training process. Our approach consistently delivers superior results compared to existing reconstruction methods, with impressive performance metrics, including a peak signal-to-noise ratio (PSNR) of 33.69±0.90 dB, a structural similarity index (SSIM) of 0.9557±0.0029, and a root mean squared error (RMSE) of 22.73±1.56 for the result of [0, 150°] reconstruction, highlighting its effectiveness in enhancing image quality and accuracy.


  • Polystyrene-supported imidazolium acidic ionic liquids: highly efficient catalysts for the synthesis of bisphenols

    by Dan Su, Fei Xu, Heng Wang, Jingxue Xie, Shijie Wang, Ming Jiang, Mi Feng, Zhencai Zhang, Zhiqiang Song, Na Liu

    A series of reusable polystyrene-supported imidazolium acidic ionic liquid catalysts were synthesized to replace industrial sulfuric acid for the synthesis of bisphenol compounds. These ionic liquids were covalently bonded to the support, reaching a high loading of 1.62–1.80 mmol g−1, respectively. As expected, they achieved a catalytic effect comparable to that of sulfuric acid in the condensation reaction of 9-fluorenone with phenol. Under the optimum reaction parameters, a high yield and high selectivity of 99.9% and 99.7% were achieved for 9,9-bis(4-hydroxyphenyl)fluorene (BHPF). Moreover, this catalyst was also employed for synthesizing other bisphenol compounds and 99.0% conversion of the substrate was achieved. Based on the experimental results and density functional theory (DFT) calculations, a possible reaction mechanism was proposed, which indicated that 9-fluorenone and phenol were simultaneously activated through the synergistic catalysis of anions and cations in a polystyrene-supported ionic liquid catalyst. The recycling experiment results suggested that [PS-PsIm][CF3SO3] maintained a high catalytic activity in five reaction runs, making this a promising catalyst for the synthesis of bisphenol compounds.


  • Open Access

    Fast raypath separation based on low-rank matrix approximation in a shallow-water waveguide

    by Longyu Jiang, Wenbo Song, Zhe Zhang, Chunfeng Yang, Shijie Wang, Philippe Roux

    Subspace algorithms based on higher-order cumulants were developed to achieve high-resolution separation in non-Gaussian processes. However, singular value decomposition (SVD) of a huge matrix is an unavoidable step of these algorithms. The memory space and running time required by the decomposition are super-linear with respect to the size of the matrix, which is prohibitive in terms of practical applications. Thus, in this paper, a fast raypath separation algorithm based on low-rank matrix approximation is proposed in a shallow-water waveguide. The experimental results illustrate that the proposed algorithm dramatically reduces the consumption of time and space, with arbitrarily small error, compared to conventional higher-order cumulant-based algorithms.


  • Metal artifact reduction in CT using fusion based prior image

    by Jun Wang, Shijie Wang, Yang Chen, Jiasong Wu, Jean-Louis Coatrieux, Limin Luo

    Purpose: In computed tomography, metallic objects in the scanning field create the so‐called metal artifacts in the reconstructed images. Interpolation‐based methods for metal artifact reduction (MAR) replace the metal‐corrupted projection data with surrogate data obtained from interpolation using the surrounding uncorrupted sinogram information. Prior‐based MAR methods further improve interpolation‐based methods by better estimating the surrogate data using forward projections from a prior image. However, the prior images in most existing prior‐based methods are obtained from segmented images and misclassification in segmentation often leads to residual artifacts and tissue structure loss in the final corrected images. To overcome these drawbacks, the authors propose a fusion scheme, named fusion prior‐based MAR (FP‐MAR). Methods: The FP‐MAR method consists of (i) precorrect the image by means of an interpolation‐based MAR method and an edge‐preserving blur filter; (ii) generate a prior image from the fusion of this precorrected image and the originally reconstructed image with metal parts removed; (iii) forward project this prior image to guide the estimation of the surrogate data using well‐developed replacement techniques. Results: Both simulations and clinical image tests are carried out to show that the proposed FP‐MAR method can effectively reduce metal artifacts. A comparison with other MAR methods demonstrates that the FP‐MAR method performs better in artifact suppression and tissue feature preservation. Conclusions: From a wide range of clinical cases to which FP‐MAR has been tested (single or multiple pieces of metal, various shapes, and sizes), it can be concluded that the proposed fusion based prior image preserves more tissue information than other segmentation‐based prior approaches and can provide better estimates of the surrogate data in prior‐based MAR methods.


  • Potential biomarkers for esophageal carcinoma detected by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

    by Li-Hua Liu, Bao-En Shan, Zi-Qiang Tian, Mei-Xiang Sang, Jun Ai, Ze-Feng Zhang, Jun Meng, Hui Zhu, Shi-Jie Wang

    Currently, no satisfactory biomarkers are available to screen for esophageal squamous cell carcinoma (ESCC). The goal of this study was to find biomarkers and establish a serum protein fingerprint model for early diagnosis of ESCC using the ClinProt protocol of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Serum samples were collected from 62 patients with ESCC, nine patients with esophageal adenocarcinoma (EA) and 38 healthy individuals. Proteomic spectra of mass to charge ratio (m/z) were generated following the application of plasma to weak cationic-exchanger magnetic beads (WCX-MB). The spectral data were analyzed using a support vector machine, and potential biomarkers were chosen for system training and used to construct diagnostic models. Three differential patterns were established using MALDI-TOF MS. Pattern 1, consisting of 11 protein peaks, separated ESCC patients from the healthy individuals with a sensitivity of 90.0% and a specificity of 88.4%. Pattern 2, consisting of eight protein peaks, separated ESCC in stage I and stage II from stage III and stage IV with a sensitivity of 92.9% and a specificity of 82.3%. Pattern 3, consisting of seven protein peaks, separated ESCC from EA with a sensitivity of 91.3% and a specificity of 80.0%. These results suggested that MALDI-TOF MS combined with MB separation yields significantly higher sensitivity and specificity for the detection of serum protein in patients with ESCC.