3D PET/MRI image fusion based on improved structure tensor

doi:10.3969/j.issn.1000-1565.2023.05.012

Abstract

Abstract: The fusion of magnetic resonance imaging(MRI)and positron emission computed tomography(PET)can achieve the prevention and diagnosis of Alzheimers disease(AD). Although the current three-dimensional image fusion algorithm based on structural tensor has good spatial characteristics, it can also represent the structural information of the image. This paper proposes structure tensor based on 3D ISobel and double-objective marine predator algorithm(IST), which improves the reliability and stability of the decomposition and makes the detailed information fully expressed. In this paper, the edge region fusion rule of Joint Weight Sum of 3D Energy And Modified Laplacian And Trace Phase Consistency(ETPC)is proposed, which fully describes the local region feature distribution of the fused - DOI:10.3969/j.issn.1000-1565.2023.05.012基于改进结构张量的三维PET/MRI图像融合郑伟¹,夏雪婷¹,张晓丹²,马泽鹏³(1.河北大学电子信息工程学院,河北保定 071002;2.河北大学附属医院超声诊断科,河北保定 071000;3.河北大学附属医院放射科,河北保定 071000)摘要:将核磁共振成像(magnetic resonance imaging, MRI)与正电子发射断层成像(positron emission computed tomography, PET)进行融合,可以实现阿尔茨海默病(Alzheimers disease, AD)的预防与诊断.提出了基于3D ISobel与双目标海洋捕食者的结构张量(ISobel and double-objective marine predator algorithm based on structure tensor, IST)分解算法,提高了分解的可靠性与稳定性,使细节信息充分表达.提出了联合三维拉普拉斯加权能量的迹值相位一致性(joint weight sum of 3d energy and modified laplacian and trace phase congruency, ETPC)的边缘区融合规则,充分描述了融合图像局部区域特征分布.融合结果表明,基于改进结构张量的三维PET/MRI图像融合算法解决了结构张量分解欠缺可靠性、融合图像细节信息不充分、局部特征描述不完善的问题,在主观评价和客观评价方面都要优于其他的空间域和变换域的融合方法.关键词:图像融合;结构张量;海洋捕食者算法;三维PET/MRI;相位一致性中图分类号:TP391.9 文献标志码:A 文章编号:1000-1565(2023)05-0525-143D PET/MRI image fusion based on improved structure tensorZHENG Wei¹, XIA Xueting¹, ZHANG Xiaodan², MA Zepeng³(1.College of Electronic Information Engineering, Hebei University, Baoding 071002, China;2. Department of Ultrasound Diagnosis, Affiliated Hospital of Hebei University, Baoding 071000, China;3. Department of Radiology, Affiliated Hospital of Hebei University, Baoding 071000, China)Abstract: The fusion of magnetic resonance imaging(MRI)and positron emission computed tomography(PET)can achieve the prevention and diagnosis of Alzheimers disease(AD). Although the current three-dimensional image fusion algorithm based on structural tensor has good spatial characteristics, it can also represent the structural information of the image. This paper proposes structure tensor based on 3D ISobel and double-objective marine predator algorithm(IST), which improves the reliability and stability of the decomposition and makes the detailed information fully expressed. In this paper, the edge region fusion rule of Joint Weight Sum of 3D Energy And Modified Laplacian And Trace Phase Consistency(ETPC)is proposed, which fully describes the local region feature distribution of the fused - 收稿日期:2023-01-06 基金项目:河北省自然科学基金资助项目( F2020201025; H2020201021);河北省高等学校科学技术研究项目(BJ2020030);河北大学附属医院青年科研基金资助项目(2021Q021;2022QB28) 第一作者:郑伟(1972—),女,黑龙江兰西人,河北大学教授,博士,主要从事图像处理与分析、图像安全通信、图像加密和隐藏研究. E-mail:147685650@qq.com 通信作者:马泽鹏(1989—),男,河北保定人,河北大学附属医院放射科主治医师,主要从事脑部和心脏疾病的CT、MR、fMRI等的鉴别诊断研究.E-mail:mzpdan@163.com第5期郑伟等:基于改进结构张量的三维PET/MRI图像融合image. The 3D PET/MRI image fusion algorithm based on improved structural tensor solves the problems of the lack of reliability of structural tensor decomposition, insufficient detail information of fusion image, and imperfect description of local features, and it is better than other fusion methods of spatial domain and transform domain in terms of subjective evaluation and objective evaluation.

Key words: image fusion, structure tensor, MPA, 3D PET/MRI, Phase congruency

CLC Number:

TP391.9

ZHENG Wei, XIA Xueting, ZHANG Xiaodan, MA Zepeng. 3D PET/MRI image fusion based on improved structure tensor[J]. Journal of Hebei University(Natural Science Edition), 2023, 43(5): 525-538.

References

[1] GAO F. Integrated positron emission tomography/magnetic resonance imaging in clinical diagnosis of Alzheimer’s disease[J]. Eur J Radiol, 2021, 145: 110017. DOI: 10.1016/j.ejrad.2021.110017.
[2] 刘思睿,有慧,冯逢.PET和PET/MRI在阿尔茨海默病中的应用[J].国际医学放射学杂志, 2019, 42(1): 62-65. DOI: 10.19300/j.2019.Z6091.
[3] LI H F, LI X S, YU Z T, et al. Multifocus image fusion by combining with mixed-order structure tensors and multiscale neighborhood[J]. Inf Sci, 2016, 349/350: 25-49. DOI: 10.1016/j.ins.2016.02.030.
[4] FU Z Z, ZHAO Y F, XU Y W, et al. Gradient structural similarity based gradient filtering for multi-modal image fusion[J]. Inf Fusion, 2020, 53: 251-268. DOI: 10.1016/j.inffus.2019.06.025.
[5] 肖倩.基于改进结构张量的PET和MRI图像融合[D].保定: 河北大学, 2020.
[6] WANG N, ZHANG W Y, LI D W. GIMI: a new evaluation index for 3D multimodal medical image fusion[C] //2018 14th International Conference on Computational Intelligence and Security(CIS). Hangzhou, 2018: 25-29. DOI: 10.1109/CIS2018.2018.00014.
[7] HARRIS C G, STEPHENS M J. A combined corner and edge detector[C] //Alvey Vision Conferece, 1988.DOI: 10.5244/C.2.23.
[8] 郑伟,安晓林,李涵,等.基于3D NSDST和改进空间频率相结合的MRI/PET图像融合[J].微电子学与计算机, 2021, 38(11): 53-60. DOI: 10.3969/j.issn.1003-501X.2016.10.008.
[9] FARAMARZI A, HEIDARINEJAD M, MIRJALILI S, et al. Marine predators algorithm: a nature-inspired metaheuristic[J]. Expert Syst Appl, 2020, 152: 113377. DOI: 10.1016/j.eswa.2020.113377.
[10] HE D K, WANG L, CHEN B, et al. Improved select space multi-objective particle swarm optimization[C] //2010 Sixth International Conference on Natural Computation, Yantai, 2010: 2764-2768. DOI: 10.1109/ICNC.2010.5583807.
[11] IOANNIDIS Y E, WONG E. Query optimization by simulated annealing[J]. SIGMOD Rec, 1987, 16(3): 9-22. DOI: 10.1145/38713.38722.
[12] KIRKPATRICK S, GELATT C D Jr, VECCHI M P. Optimization by simulated annealing[J]. Science, 1983, 220(4598): 671-680. DOI: 10.1126/science.220.4598.671.
[13] MIRJALILI S, LEWIS A. The whale optimization algorithm[J]. Adv Eng Softw, 2016, 95(C): 51-67. DOI: 10.1016/j.advengsoft.2016.01.008.
[14] 李涵.基于改进的像素级多尺度变换的PET和MRI图像融合[D].保定:河北大学, 2021.
[15] ZHU Z Q, ZHENG M Y, QI G Q, et al. A phase congruency and local Laplacian energy based multi-modality medical image fusion method in NSCT domain[J]. IEEE Access, 2019, 7: 20811-20824. DOI: 10.1109/ACCESS.2019.2898111.
[16] 余艳,杨学志,方帅,等.基于相位一致性与PCNN的SAR和多光谱图像融合算法[J].合肥工业大学学报(自然科学版), 2022, 45(3): 324-331. DOI: 10.3969/j.issn.1003-5060.2022.03.007.
[17] SAZAK Ç, OBARA B. Contrast-independent curvilinear structure enhancement in 3D biomedical images[C] //2017 IEEE 14th International Symposium on Biomedical Imaging(ISBI 2017). Melbourne, VIC, Australia, 2017: 1165-1168. DOI: 10.1109/ISBI.2017.7950723.
[18] ZHANG Q, SHEN X Y, XU L, et al. Rolling guidance filter[M] //Computer Vision-ECCV 2014. Cham: Springer International Publishing, 2014: 815-830.
[19] LI X S, ZHOU F Q, TAN H S, et al. Multimodal medical image fusion based on joint bilateral filter and local gradient energy[J]. Inf Sci, 2021, 569: 302-325. DOI: 10.1016/j.ins.2021.04.052. (