面向肿瘤早期诊断的延迟PET图像重建:多模态PET/CT核矩阵约束延迟成像算法

doi:10.11896/jsjkx.250400037

计算机科学 ›› 2025, Vol. 52 ›› Issue (9): 119-127.doi: 10.11896/jsjkx.250400037

面向肿瘤早期诊断的延迟PET图像重建:多模态PET/CT核矩阵约束延迟成像算法

宋志超^1,2, 张建平³, 张其阳², 方玺¹, 谢良¹, 宋少莉³, 胡战利²

1 武汉理工大学数学与统计学院武汉 430070
2 中国科学院深圳先进技术研究院医学人工智能中心广东深圳 518055
3 复旦大学附属肿瘤医院核医学科上海 200032

收稿日期:2025-04-09 修回日期:2025-06-20 出版日期:2025-09-15 发布日期:2025-09-11
通讯作者: 宋少莉(shaoli-song@163.com)
作者简介:(zcsong.math@whut.edu.cn)
基金资助:
国家自然科学基金(82372038)

Delayed PET Reconstruction for Early Tumor Diagnosis:Multimodal PET/CT Nuclear Matrix- constrained Delayed Imaging Algorithm

SONG Zhichao^1,2, ZHANG Jianping³, ZHANG Qiyang², FANG Xi¹, XIE Liang¹, SONG Shaoli³, HU Zhanli²

1 School of Mathematics and Statistics,Wuhan University of Technology,Wuhan 430070,China
2 Research Center for Medical AI,Shenzhen Institute of Advanced Technology,Chinese Academy of Sciences,Shenzhen,Guangdong 518055,China
3 Department of Nuclear Medicine,Fudan University Shanghai Cancer Center,Shanghai 200032,China

Received:2025-04-09 Revised:2025-06-20 Online:2025-09-15 Published:2025-09-11
About author:SONG Zhichao,born in 1999,postgra-duate.Her main research interests include medical image processing and survival analysis.
SONG Shaoli,born in 1972,Ph.D,chief physician,recipient of Shanghai Leading Talents(2023).Her main research interest is basic and cinical research in nuclear medicine of tumors.
Supported by:
National Natural Science Foudation of China(82372038).

摘要/Abstract

摘要： 正电子发射断层扫描(PET)延迟成像在肿瘤异质性分析和治疗评估中具有重要意义,但其临床应用受限于分辨率低、噪声高和定量不准确等问题。计算机断层扫描(CT)能够提供高分辨率的解剖信息,但在肿瘤评估中缺乏功能信息,难以区分良恶性病变和评估代谢活动。虽然动态PET/CT融合能提升图像质量,但多次CT扫描会增加患者累积辐射暴露,不利于长期随访。针对上述问题,提出了一种超分增强PET/CT多模态核矩阵约束算法(SR-PET/CT-KMC)。该算法基于Stable Diffusion对初始扫描PET图像进行超分增强,并将其与初始扫描CT图像的解剖先验信息相结合,建立了多模态PET/CT核矩阵约束的期望最大化(EM)迭代框架。Stable Diffusion用于提升初始扫描PET的分辨率,而多模态PET/CT先验信息则用于抑制噪声和伪影。通过利用初始扫描CT的结构信息,降低了延迟成像中CT扫描的需求,从而减少了患者累积辐射暴露。实验结果表明,SR-PET/CT-KMC与PET-KEM相比,PSNR提高了6.23%,SSIM提高了9.64%,NRMSE降低了33.3%,MSE降低了13.92%;与CT-KEM相比,PSNR提高了4.05%,SSIM提高了1.11%,NRMSE降低了33.3%,MSE降低了8.11%。这些结果表明,SR-PET/CT-KMC在提升延迟扫描PET图像分辨率和定量准确性方面具有优势,为肿瘤代谢追踪提供了一种新的成像范式,提高了延迟PET成像的临床可行性。

关键词: 延迟成像, 超分辨率PET, 多模态核矩阵, 生物医学, 核方法

Abstract: Positron emission tomography(PET) delayed imaging is of great significance in tumor heterogeneity analysis and treatment evaluation,but its clinical application is limited by low resolution,high noise and inaccurate quantification.Computed tomography(CT) can provide high-resolution anatomical information,but lacks functional information in tumor assessment,making it difficult to distinguish benign and malignant lesions and evaluate metabolic activity.Although dynamic PET/CT fusion can improve image quality,multiple CT scans increase patient radiation exposure and are not conducive to long-term follow-up.To address the above problems,a super-resolution enhanced PET/CT multimodal kernel matrix constraint algorithm(SR-PET/CT-KMC) is proposed.The algorithm super-enhances the initial scan PET image based on stable diffusion,combines it with the anatomical prior information of the initial scan CT image,and establishes an expectation maximization(EM) iterative framework for multimodal PET/CT kernel matrix constraints.Stable diffusion is used to improve the resolution of the initial scan PET,while the multimodal PET/CT prior information is used to suppress noise and artifacts.By utilizing the structural information of the initial scan CT,the need for CT scans in delayed imaging is reduced,thereby reducing the patient’s cumulative radiation exposure.Experimental results show that compared with PET-KEM,the peak signal-to-noise ratio(PSNR) of SR-PET/CT-KMC is improved by 6.23%,the structural similarity index(SSIM) is improved by 9.64%,the normalized root mean square error(NRMSE) is reduced by 33.3%,and the mean square error(MSE) is reduced by 13.92%.Compared with CT-KEM,the PSNR is improved by 4.05%,the SSIM is improved by 1.11%,the NRMSE is reduced by 33.3%,and the MSE is reduced by 8.11%.These results show that this method has advantages in improving the resolution and quantitative accuracy of delayed scanning PET images,providing a new imaging paradigm for tumor metabolism tracking and improving the clinical feasibility of delayed PET imaging.

Key words: Delayed imaging, Super-resolution PET, Multi-modal kernel matrix, Biomedical, Kernel method

中图分类号:

TP391.41

宋志超, 张建平, 张其阳, 方玺, 谢良, 宋少莉, 胡战利. 面向肿瘤早期诊断的延迟PET图像重建:多模态PET/CT核矩阵约束延迟成像算法[J]. 计算机科学, 2025, 52(9): 119-127. https://doi.org/10.11896/jsjkx.250400037

SONG Zhichao, ZHANG Jianping, ZHANG Qiyang, FANG Xi, XIE Liang, SONG Shaoli, HU Zhanli. Delayed PET Reconstruction for Early Tumor Diagnosis:Multimodal PET/CT Nuclear Matrix- constrained Delayed Imaging Algorithm[J]. Computer Science, 2025, 52(9): 119-127. https://doi.org/10.11896/jsjkx.250400037

参考文献

[1]STAFFORD W C,PENG X,OLOFSSON M H,et al.Irreversible inhibition of cytosolic thioredoxin reductase 1 as a mechanistic basis for anticancer therapy[J].Science Translational Medicine,2018,10(428):eaaf7444.
[2]SCHWENCK J,SONANINI D,COTTON J M,et al.Advances in PET imaging of cancer[J].Nature Reviews Cancer,2023,23(7):474-490.
[3]CROTEAU E,RENAUD J M,RICHARD M A,et al.PET Metabolic Biomarkers for Cancer:Supplementary Issue:Biomarkers and their Essential Role in the Development of Personalised Therapies(A)[J].Biomarkers in cancer,2016,8:BIC.S27483.
[4]GRASSO V,WILLUMEIT-RÖMER R,JOSE J.Superpixelspectral unmixing framework for the volumetric assessment of tissue chromophores:A photoacoustic data-driven approach[J].Photoacoustics,2022,26:100367.
[5]KHARE S K,MARCH S,BARUA P D,et al.Application of data fusion for automated detection of children with developmental and mental disorders:A systematic review of the last decade[J].Information Fusion,2023,99:101898.
[6]BAGCI U,UDUPA J K,MENDHIRATTA N,et al.Joint segmentation of anatomical and functional images:Applications in quantification of lesions from PET,PET-CT,MRI-PET,and MRI-PET-CT images[J].Medical Image Analysis,2013,17(8):929-945.
[7]BONNEY L M,KALISVAART G M,VAN VELDEN F H P,et al.Deep learning image enhancement algorithms in PET/CT imaging:a phantom and sarcoma patient radiomic evaluation[J].European Journal of Nuclear Medicine and Molecular Imaging,2025:1-12.
[8]DIMITRAKOPOULOU-STRAUSS A,PAN L,SACHPEKIDISC.Kinetic modeling and parametric imaging with dynamic PET for oncological applications:general considerations,current clinical applications,and future perspectives[J].European journal of nuclear medicine and molecular imaging,2021,48:21-39.
[9]ZONDERVAN R L,HAHN P F,SADOW C A,et al.Frequent body CT scanning of young adults:indications,outcomes,and risk for radiation-induced cancer[J].Journal of the American College of Radiology,2011,8(7):501-507.
[10]MATAAC M T,LI X,REHANI M M.What proportion of CT scan patients are alive or deceased after 10 years?[J].European Journal of Radiology,2024,178:111629.
[11]ALBERTS I,SARI H,MINGELS C,et al.Long-axial field-of-view PET/CT:perspectives and review of a revolutionary development in nuclear medicine based on clinical experience in over 7000 patients[J].Cancer Imaging,2023,23(1):28.
[12]MOSTAFAPOUR S,GREUTER M,VAN SNICK J H,et al.Ultra-low dose CT scanning for PET/CT[J].Medical Physics,2024,51(1):139-155.
[13]SHEN C,TIE C,YANG Z,et al.Bidirectional condition diffu-sion probabilistic models for pet image denoising[J].IEEE Transactions on Radiation and Plasma Medical Sciences,2024,8(4):402-415.
[14]WANG J,YUE Z,ZHOU S,et al.Exploiting diffusion prior for real-world image super-resolution[J].International Journal of Computer Vision,2024,132(12):5929-5949.
[15]AI X,HUANG B,CHEN F,et al.RED:Residual estimation diffusion for low-dose PET sinogram reconstruction[J].Medical Image Analysis,2025,102:103558.
[16]ZHANG Q,HU Y,ZHOU C,et al.Reducing pediatric total-body PET/CT imaging scan time with multimodal artificial intelligence technology[J].European Journal of Nuclear Medicine and Molecular Imaging physics,2024,11(1):1.
[17]FILIPOVIĆ M,BARAT E,DAUTREMER T,et al.PET reconstruction of the posterior image probability,including multimodal images[J].IEEE Transactions on Medical Imaging,2018,38(7):1643-1654.
[18]ZAIDI H,BECKER M.The promise of hybrid PET\/MRI:technical advances and clinical applications[J].IEEE Signal Processing Magazine,2016,33(3):67-85.
[19]CHEN Z,LIU Y,ZHANG Y,et al.Enhanced Multimodal Low-rank Embedding based Feature Selection Model for Multimodal Alzheimer’s Disease Diagnosis[J].IEEE Transactions on Medical Imaging,2025,44(2):815-827.
[20]WANG G,QI J.PET image reconstruction using kernel method[J].IEEE Transactions on Medical Imaging,2014,34(1):61-71.
[21]MULLER F M,LI E J,DAUBE-WITHERSPOON M E,et al.Impact of deep learning denoising on kinetic modelling for low-dose dynamic PET:application to single-and dual-tracer imaging protocols[J].European Journal of Nuclear Medicine and Mole-cular Imaging,2025:1-19.
[22]ZHOU S,CHAN K,LI C,et al.Towards robust blind face restoration with codebook lookup transformer[J].Advances in Neural Information Processing Systems,2022,35:30599-30611.
[23]DRAELOS R L,DOV D,MAZUROWSKI M A,et al.Machine-learning-based multiple abnormality prediction with large-scale chest computed tomography volumes[J].Medical Image Analysis,2021,67:101857.
[24]NUYTS J,MICHEL C,DUPONT P.Maximum-likelihood expectation-maximization reconstruction of sinograms with arbitrary noise distribution using NEC-transformations[J].IEEE Transactions on Medical Imaging,2001,20(5):365-375.
[25]HU R,CUI J,LI C,et al.Dynamic low-count PET image reconstruction using spatio-temporal primal dual network[J].Physics in Medicine & Biology,2023,68(13):135015.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

面向肿瘤早期诊断的延迟PET图像重建:多模态PET/CT核矩阵约束延迟成像算法

Delayed PET Reconstruction for Early Tumor Diagnosis:Multimodal PET/CT Nuclear Matrix- constrained Delayed Imaging Algorithm

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 0