计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (10): 115-122.doi: 10.11896/jsjkx.240700135

• 计算机图形学&多媒体 • 上一篇    下一篇

基于成像切片连续性的PET直接生成CT的衰减校正算法

郑涵元1, 葛荣骏2, 何升级3, 李楠4   

  1. 1 南京航空航天大学人工智能学院 南京 211106
    2 东南大学仪器科学与工程学院 南京 210096
    3 赛诺联合医疗科技(北京)有限公司 北京 100192
    4 江苏赛诺格兰医疗科技有限公司 江苏 扬州 225200
  • 收稿日期:2024-07-22 修回日期:2024-11-01 出版日期:2025-10-15 发布日期:2025-10-14
  • 通讯作者: 葛荣骏(rongjun_ge@seu.edu.cn)
  • 作者简介:(njzhenghy@nuaa.edu.cn)
  • 基金资助:
    国家自然科学基金(62101249);江苏省双创博士项目(JSSCBS20220202);中国博士后科学基金(2022M721611,2021TQ0149)

Direct PET to CT Attenuation Correction Algorithm Based on Imaging Slice Continuity

ZHENG Hanyuan1, GE Rongjun2, HE Shengji3, LI Nan4   

  1. 1 College of Artificial Intelligence,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China
    2 College of Instrument Science and Engineering,Southeast University,Nanjing 210096,China
    3 SinoUnion Healthcare Inc.,Beijing 100192,China
    4 Jiangsu Sinogram Medical Technology Co.,Ltd.,Yangzhou,Jiangsu 225200,China
  • Received:2024-07-22 Revised:2024-11-01 Online:2025-10-15 Published:2025-10-14
  • About author:ZHENG Hanyuan,born in 2000,postgraduate.His main research interests include medical image reconstruction and deep learning.
    GE Rongjun,born in 1992,Ph.D,asso-ciate professor,is a member of CCF(No.13248M).His main research interests include deep learing,intelligent reconstruction and analysis of medical image.
  • Supported by:
    National Natural Science Foundation of China(62101249),Jiangsu Province Double Innovation Doctoral Program(JSSCBS20220202) and China Postdoctoral Science Foundation(2022M721611,2021TQ0149).

PDF (PC)

0

摘要： PET衰减校正技术在临床上具有重要意义,其能够有效降低癌症误诊率并制定更为精确的治疗计划。然而,传统的PET衰减校正方法面临扫描时间较长和后期处理引入误差等问题,限制了其应用范围。近期,基于PET直接生成CT的衰减校正方法凭借更短的扫描时长和无后期处理误差的优势,逐渐流行于临床应用。然而,由于PET与CT之间的语义差异巨大且不对齐,使得PET直接生成CT的衰减方法在模态生成中面临着巨大的挑战。针对这一挑战,提出了一种基于Cycle-S2SCT-Net生成网络的PET衰减校正方法。Cycle-S2SCT-Net在整体上借助循环生成对抗结构学习PET与CT分布变换映射,实现了PET与CT两种模态间的语义转换。在单个生成对抗网络内部,Cycle-S2SCT-Net集成了成像切片连续性模块,以增强网络的语义对齐能力,从而提高生成成像的连续性和准确性。此外,还引入了网络特征层损失函数(Layer Loss),以增强生成网络的特征提取能力。实验结果表明,Cycle-S2SCT-Net生成的CT及其衰减校正后的PET,在定量评估指标峰值信噪比(PSNR)、结构相似性(SSIM)、均方根误差(RMSE),以及可视化结果上均表现出色。

关键词: PET, 衰减校正方法, CT, 切片连续性, 循环生成对抗

Abstract: PET attenuation correction technology is of significant clinical importance,effectively reducing cancer misdiagnosis rates and enabling more precise treatment planning.However,traditional PET attenuation correction methods face challenges such as long scan times and errors introduced during post-processing,limiting their applicability.Recently,attenuation correction methods based on directly generating CT from PET have gained popularity in clinical settings due to shorter scan times and the advantage of avoiding post-processing errors.However,the substantial semantic differences and misalignment between PET and CT pose significant challenges for direct PET-to-CT attenuation methods in modal generation.To address this challenge,this paper proposes a PET attenuation correction method based on Cycle-S2SCT-Net.Cycle-S2SCT-Net utilizes a cyclic generative adversarial structure to learn the mapping between PET and CT distributions,facilitating semantic translation between the two modalities.Within a single generative adversarial network,Cycle-S2SCT-Net integrates an imaging slice continuity module to enhance the network's semantic alignment capability,thereby improving the continuity and accuracy of generated images.Addi-tionally,this paper introduces a network feature layer loss function(Layer Loss) to enhance the feature extraction capability of the generation network.The experimental results demonstrate that CT generated by Cycle-S2SCT-Net and its attenuation-corrected PET exhibit excellent performance in both quantitative evaluation metrics,such as peak signal-to-noise ratio(PSNR),structural similarity index(SSIM),root mean square error(RMSE),and visual quality.

Key words: PET,Attenuation correction methods,CT,Slice continuity,Cycle generative adversarial

中图分类号: 

  • TP391
[1]CHERRY S R,DAHLBOM M,PHELPS M E.PET:Physics,Instrumentation,and Scanners[M].New York:Springer,2006.
[2]WAHL R L.To AC or not to AC:that is the question[J].Jour-nal of Nuclear Medicine Official Publication Society of Nuclear Medicine,1999,40(12):2025-2028.
[3]KINAHAN P E,TOWNSEND D W,BEYER T,et al.Attenuation correction for a combined 3D PET/CT scanner[J].MedicalPhysics,1998,25(10):2046-2053.
[4]HOFMANN M,STEINKE F,SCHEEL V,et al.MRI-Based Attenuation Correction for PET/MRI:A Novel Approach Combining Pattern Recognition and Atlas Registration[J].Journal of Nuclear Medicine,2008,49(11):1875-1883.
[5]SCHULZ V,TORRES-ESPALLARDO I,RENISCH S,et al.Automatic,three-segment,MR-based attenuation correction for whole-body PET/MR data[J].European Journal of Nuclear Medicine and Molecular Imaging,2011,38(1):138-152.
[6]GAO Y,ZHAO J H.The research progress of PET/MR attenuation correction technology [J].China Medical Equipment,2015,30(7):75-78.
[7]HOFMANN M,STEINKE F,SCHEEL V,et al.MRI-Based Attenuation Correction for PET/MRI:A Novel Approach Combining Pattern Recognition and Atlas Registration[J].Journal of Nuclear Medicine,2008,49(11):1875-1883.
[8]UH J,MERCHANT T E,LI Y,et al.MRI-based treatmentplanning with pseudo CT generated through atlas registration[J].Medical Physics,2013,41(5):051711.
[9]ROY S,CARASS A,JOG A,et al.MR to CT registration ofbrains using image synthesis[C]//Proceedings of SPIE:the International Society for Optical Engineering.2014.
[10]ANDREASEN D,EDMUND J M,ZOGRAFOS V,et al.Computed tomography synthesis from magnetic resonance images in the pelvis using multiple random forests and auto-context features[C]//Medical Imaging 2016:Image Processing.SPIE,2016.
[11]GONG K,YANG J,KIM K,et al.Attenuation correction for brain PET imaging using deep neural network based on dixon and ZTE MR images[J].Physics in Medicine & Biology,2018,63(12):125011.
[12]LEYNES A P,YANG J,WIESINGER,F,et al.Zero-Echo-Time and Dixon Deep Pseudo-CT(ZeDD CT):Direct Generation of Pseudo-CT Images for Pelvic PET/MRI Attenuation Correction Using Deep Convolutional Neural Networks with Multiparametric MRI[J].Journal of Nuclear Medicine,2018,59(5):852-858.
[13]TORRADO-CARVAJAL A,VERA-OLMOS J,IZQUIERDO-GARCIA D,et al.Dixon-VIBE Deep Learning(DIVIDE) Pseudo-CT Synthesis for Pelvis PET/MR Attenuation Correction [J].Journal of Nuclear Medicine,2019,60(3):429-435.
[14]PAN Y,LIU F,JIANG C,et al.Revealing Anatomical Structures in PET to Generate CT for Attenuation Correction[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention.Cham:Springer,2023:24-33.
[15]KALENDER W A.X-ray computed tomography[J].Physics in Medicine & Biology,2006,51(13):R29.
[16]DONG X,WANG T,LEI Y,et al.Synthetic CT generation from non-attenuation corrected PET images for whole-body PET imaging[J].Physics in Medicine & Biology,2019,64(21):215016.
[17]DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al.Animage is worth 16x16 words:Transformers for image recognition at scale[J].arXiv:2010.11929,2020.
[18]ZHU J Y,PARK T,ISOLA P,et al.Unpaired image-to-image translation using cycle-consistent adversarial networks[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:2223-2232.
[19]GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial networks[J].Communications of the ACM,2020,63(11):139-144.
[20]VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[C]//Advances in Neural Information Processing Systems.2017.
[21]RONNEBERGER O,FISCHER P,BROX T.U-net:Convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Compu-ter-Assisted Intervention.Cham:Springer,2015:234-241.
[22]DENOTTER T D,SCHUBERT J.Hounsfield unit[M].StatPearls Publishing,2025.
[23]MIRZA M,OSINDERO S.Conditional generative adversarialnets[J].arXiv:1411.1784,2014.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发