Material decomposition of spectral CT images via attention-based global convolutional generative adversarial network

作     者：Xiaodong Guo Peng He Xiaojie Lv Xuezhi Ren Yonghui Li Yuanfeng Liu Xiaohua Lei Peng Feng Hongming Shan Xiaodong Guo;Peng He;Xiaojie Lv;Xuezhi Ren;Yonghui Li;Yuanfeng Liu;Xiaohua Lei;Peng Feng;Hongming Shan

作者机构：The Key Lab of Optoelectronic Technology and Systems of the Education Ministry of ChinaChongqing UniversityChongqing 400044China Department of Vascular and Endovascular SurgeryFirst Affiliated Hospital of Zhengzhou UniversityZhengzhou 450003China Institute of Science and Technology for Brain-inspired Intelligence and MOE Frontiers Center for Brain ScienceFudan UniversityShanghai 200433China Shanghai Center for Brain Science and Brain-Inspired TechnologyShanghai 200031China 

出 版 物：《Nuclear Science and Techniques》 (核技术（英文）)

年 卷 期：2023年第34卷第3期

页      面：143-153页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0805[工学-材料科学与工程（可授工学、理学学位）] 080502[工学-材料学] 0802[工学-机械工程] 

基　　金：supported by National Natural Science Foundation of China (No.62101136) Shanghai Sailing Program (No.21YF1402800) Shanghai Municipal Science and Technology Major Project (No.2018SHZDZX01) ZJLab,Shanghai Municipal of Science and Technology Project (No.20JC1419500) Natural Science Foundation of Chongqing (No.CSTB2022NSCQ-MSX0360) Shanghai Center for Brain Science and Brain-inspired Technology 

主　　题：Photon-counting CT Material decomposition Attention mechanism GAN 

摘      要：Spectral computed tomography(CT)based on photon counting detectors can resolve the energy of every single photon interacting with the sensor layer and be used to analyze material attenuation information under different energy ranges,which can be helpful for material decomposition ***,there is a considerable amount of inherent quantum noise in narrow energy bins,resulting in a low signal-to-noise ratio,which can consequently affect the material decomposition performance in the image *** learning technology is currently widely used in medical image segmentation,denoising,and *** order to improve the results of material decomposition,we propose an attention-based global convolutional generative adversarial network(AGC-GAN)to decompose different materials for spectral ***,our network is a global convolutional neural network based on an attention mechanism that is combined with a generative adversarial *** global convolutional network based on the attention mechanism is used as the generator,and a patchGAN discriminant network is used as the ***,a clinical spectral CT image dataset is used to verify the feasibility of our proposed *** experimental results demonstrate that AGC-GAN achieves a better material decomposition performance than vanilla U-Net,fully convolutional network,and fully convolutional ***,the mean intersection over union,structural similarity,mean precision,PAcc,and mean F1-score of our method reach up to 87.31%,94.83%,93.22%,97.39%,and 93.05%,respectively.