基于期望最大化算法的扩散弛豫定量方法研究
Research on Diffusion-Relaxation Quantitative Analysis Method Based on Expectation-Maximization Algorithm
DOI: 10.12677/mos.2025.142182, PDF,   
作者: 高清宇, 王 磊：上海理工大学健康科学与工程学院，上海；海军军医大学第一附属医院影像医学科，上海；王 涛：海军军医大学第一附属医院核医学科，上海；王前锋：复旦大学类脑智能科学与技术研究院，上海；马 超*：海军军医大学第一附属医院影像医学科，上海；同济大学电子与信息工程学院，上海；王丽嘉*：上海理工大学健康科学与工程学院，上海
关键词: 超高场磁共振成像；混合扩散弛豫；无监督学习；期望最大化算法；胰腺癌；Ultra-High Field Magnetic Resonance Imaging； Hybrid Diffusion Relaxation； Unsupervised Learning； Expectation-Maximization Algorithm； Pancreatic Cancer
摘要: 目的：通过结合无监督机器学习算法和混合弛豫扩散磁共振成像(MRI)，实现对胰腺癌的虚拟病理特征预测。方法：构建胰腺癌裸鼠模型，并使用11.7-T MRI扫描仪获取了包含多个b值(0、150、500、1500 s/mm²)和多个回波时间(25、50、75、100 ms)的扩散加权成像(DWI)数据。利用高斯混合模型和期望最大化算法，建立了一种混合弛豫扩散定量分析方法，用以计算肿瘤中上皮、间质和管腔成分的体积分数参数图，并与病理结果进行比较，以验证所建立分析方法的可行性。结果：超高场MRI下胰腺癌裸鼠模型弛豫扩散成像图像分辨率达到0.5 × 0.5 × 1 mm³，胰腺癌肿瘤组织的T2值范围为39至90 ms。基于混合扩散弛豫模型，我们建立的期望最大化算法能够快速计算出肿瘤组织中上皮、间质和管腔三种组织成分的相应权重。对于两种不同纤维化程度的肿瘤，基于混合扩散弛豫参数图计算出的肿瘤上皮(58.63%/86.55%)、间质(41.00%/13.18%)和管腔(1.65%/1.19%)成分的体积分数与病理结果相似。结论：本研究提出的无监督混合扩散弛豫定量分析方法为肿瘤组织内间质、上皮和管腔比例的虚拟病理预测提供了一种新手段。
Abstract: Objective: The aim of this study is to predict the virtual pathology of pancreatic cancer by integrating unsupervised machine learning algorithms with hybrid relaxation diffusion magnetic resonance imaging (MRI). Methods: We established a pancreatic tumor-bearing nude mouse model and acquired diffusion-weighted imaging data with multiple b-values (0, 150, 500, 1500 s/mm2) and multiple echo times (25, 50, 75, 100 ms) using an 11.7-T MRI scanner. Utilizing a Gaussian mixture model and the Expectation-Maximization algorithm, we developed a quantitative analysis method for hybrid relaxation diffusion to calculate the volume fraction parameter maps of epithelial, stromal, and lumen components in tumors, and then compared with the pathological results to verify the feasibility of the method. Results: Relaxation diffusion imaging of the pancreatic cancer-bearing nude mouse model performed under ultra-high field MRI was feasible, with an image resolution of 0.5 × 0.5 × 1 mm3, and the T2 values of tumor tissue ranged from 39 to 90 ms. Based on the assumptions of the hybrid diffusion relaxation model, our established Expectation-Maximization algorithm could rapidly calculate the corresponding weights of the epithelial, stromal, and lumen components in tumor. In two tumors with different degrees of fibrosis, the volumes of epithelial (58.63%/86.55%), stromal (41.00%/13.18%), and lumen (1.65%/1.19%) components calculated based on the hybrid diffusion relaxation parameter maps were similar to the pathological results. Conclusion: The unsupervised hybrid diffusion relaxation quantitative analysis method proposed in this study provides a new means for the virtual pathological prediction of the proportions of stroma, epithelium, and lumen within tumor.
文章引用：高清宇, 王磊, 王涛, 王前锋, 马超, 王丽嘉. 基于期望最大化算法的扩散弛豫定量方法研究[J]. 建模与仿真, 2025, 14(2): 631-641. https://doi.org/10.12677/mos.2025.142182

参考文献

[1] Alexander, D.C., Dyrby, T.B., Nilsson, M. and Zhang, H. (2017) Imaging Brain Microstructure with Diffusion MRI: Practicality and Applications. NMR in Biomedicine, 32, e3841. [Google Scholar] [CrossRef] [PubMed]
[2] Panagiotaki, E., Schneider, T., Siow, B., Hall, M.G., Lythgoe, M.F. and Alexander, D.C. (2012) Compartment Models of the Diffusion MR Signal in Brain White Matter: A Taxonomy and Comparison. NeuroImage, 59, 2241-2254. [Google Scholar] [CrossRef] [PubMed]
[3] Hedouin, R., Barillot, C. and Commowick, O. (2021) Interpolation and Averaging of Diffusion MRI Multi-Compartment Models. IEEE Transactions on Medical Imaging, 40, 916-927. [Google Scholar] [CrossRef] [PubMed]
[4] Zhang, H., Schneider, T., Wheeler-Kingshott, C.A. and Alexander, D.C. (2012) NODDI: Practical in Vivo Neurite Orientation Dispersion and Density Imaging of the Human Brain. NeuroImage, 61, 1000-1016. [Google Scholar] [CrossRef] [PubMed]
[5] Lehmann, N., Aye, N., Kaufmann, J., Heinze, H., Düzel, E., Ziegler, G., et al. (2021) Longitudinal Reproducibility of Neurite Orientation Dispersion and Density Imaging (NODDI) Derived Metrics in the White Matter. Neuroscience, 457, 165-185. [Google Scholar] [CrossRef] [PubMed]
[6] Kim, D., Doyle, E.K., Wisnowski, J.L., Kim, J.H. and Haldar, J.P. (2017) Diffusion‐Relaxation Correlation Spectroscopic Imaging: A Multidimensional Approach for Probing Microstructure. Magnetic Resonance in Medicine, 78, 2236-2249. [Google Scholar] [CrossRef] [PubMed]
[7] Lampinen, B., Szczepankiewicz, F., Mårtensson, J., van Westen, D., Hansson, O., Westin, C., et al. (2020) Towards Unconstrained Compartment Modeling in White Matter Using Diffusion‐relaxation MRI with Tensor‐Valued Diffusion Encoding. Magnetic Resonance in Medicine, 84, 1605-1623. [Google Scholar] [CrossRef] [PubMed]
[8] Veraart, J., Novikov, D.S. and Fieremans, E. (2018) TE Dependent Diffusion Imaging (TEDDI) Distinguishes between Compartmental T2 Relaxation Times. NeuroImage, 182, 360-369. [Google Scholar] [CrossRef] [PubMed]
[9] Kleban, E., Tax, C.M.W., Rudrapatna, U.S., Jones, D.K. and Bowtell, R. (2020) Strong Diffusion Gradients Allow the Separation of Intra-and Extra-Axonal Gradient-Echo Signals in the Human Brain. NeuroImage, 217, Article ID: 116793. [Google Scholar] [CrossRef] [PubMed]
[10] de Almeida Martins, J.P., Tax, C.M.W., Szczepankiewicz, F., Jones, D.K., Westin, C. and Topgaard, D. (2020) Transferring Principles of Solid-State and Laplace NMR to the Field of in Vivo Brain MRI. Magnetic Resonance, 1, 27-43. [Google Scholar] [CrossRef] [PubMed]
[11] Reymbaut, A., Critchley, J., Durighel, G., Sprenger, T., Sughrue, M., Bryskhe, K., et al. (2020) Toward Nonparametric Diffusion‐Characterization of Crossing Fibers in the Human Brain. Magnetic Resonance in Medicine, 85, 2815-2827. [Google Scholar] [CrossRef] [PubMed]
[12] Wang, S., Peng, Y., Medved, M., Yousuf, A.N., Ivancevic, M.K., Karademir, I., et al. (2013) Hybrid Multidimensional T2 and Diffusion‐Weighted MRI for Prostate Cancer Detection. Journal of Magnetic Resonance Imaging, 39, 781-788. [Google Scholar] [CrossRef] [PubMed]
[13] Sadinski, M., Karczmar, G., Peng, Y., Wang, S., Jiang, Y., Medved, M., et al. (2016) Pilot Study of the Use of Hybrid Multidimensional T2-Weighted Imaging-DWI for the Diagnosis of Prostate Cancer and Evaluation of Gleason Score. American Journal of Roentgenology, 207, 592-598. [Google Scholar] [CrossRef] [PubMed]
[14] Chatterjee, A., Bourne, R.M., Wang, S., Devaraj, A., Gallan, A.J., Antic, T., et al. (2018) Diagnosis of Prostate Cancer with Noninvasive Estimation of Prostate Tissue Composition by Using Hybrid Multidimensional MR Imaging: A Feasibility Study. Radiology, 287, 864-873. [Google Scholar] [CrossRef] [PubMed]
[15] Zhang, Z., Wu, H.H., Priester, A., Magyar, C., Afshari Mirak, S., Shakeri, S., et al. (2020) Prostate Microstructure in Prostate Cancer Using 3-T MRI with Diffusion-Relaxation Correlation Spectrum Imaging: Validation with Whole-Mount Digital Histopathology. Radiology, 296, 348-355. [Google Scholar] [CrossRef] [PubMed]
[16] Chatterjee, A., Mercado, C., Bourne, R.M., Yousuf, A., Hess, B., Antic, T., et al. (2022) Validation of Prostate Tissue Composition by Using Hybrid Multidimensional MRI: Correlation with Histologic Findings. Radiology, 302, 368-377. [Google Scholar] [CrossRef] [PubMed]
[17] Slator, P.J., Hutter, J., Palombo, M., Jackson, L.H., Ho, A., Panagiotaki, E., et al. (2019) Combined Diffusion‐Relaxometry MRI to Identify Dysfunction in the Human Placenta. Magnetic Resonance in Medicine, 82, 95-106. [Google Scholar] [CrossRef] [PubMed]
[18] Melbourne, A., Aughwane, R., Sokolska, M., Owen, D., Kendall, G., Flouri, D., et al. (2018) Separating Fetal and Maternal Placenta Circulations Using Multiparametric MRI. Magnetic Resonance in Medicine, 81, 350-361. [Google Scholar] [CrossRef] [PubMed]
[19] 李天骄, 叶龙云, 金凯舟, 等. 2023年度胰腺癌研究及诊疗新进展[J]. 中国癌症杂志, 2024, 34(1): 1-13.
[20] Donoho, D. and Stodden, V. (2003) When Does Non-Negative Matrix Factorization Give a Correct Decomposition into Parts? Advances in Neural Information Processing Systems, 16, 1141-1148.
[21] Choi, S., Cichocki, A., Park, H.M. and Lee, S.Y. (2005) Blind Source Separation and Independent Component Analysis: A Review. Neural Information Processing-Letters and Reviews, 6, 1-57.
[22] Kolda, T.G. and Bader, B.W. (2009) Tensor Decompositions and Applications. SIAM Review, 51, 455-500. [Google Scholar] [CrossRef] 
[23] Benjamini, D. and Basser, P.J. (2020) Multidimensional correlation MRI. NMR in Biomedicine, 33, e4226. [Google Scholar] [CrossRef] [PubMed]
[24] Slawski, M. and Hein, M. (2013) Non-Negative Least Squares for High-Dimensional Linear Models: Consistency and Sparse Recovery without Regularization. Electronic Journal of Statistics, 7, 3004-3056. [Google Scholar] [CrossRef] 
[25] Benjamini, D. and Basser, P.J. (2016) Use of Marginal Distributions Constrained Optimization (MADCO) for Accelerated 2D MRI Relaxometry and Diffusometry. Journal of Magnetic Resonance, 271, 40-45. [Google Scholar] [CrossRef] [PubMed]
[26] Molina‐Romero, M., Gómez, P.A., Sperl, J.I., Czisch, M., Sämann, P.G., Jones, D.K., et al. (2018) A Diffusion Model‐free Framework with Echo Time Dependence for Free‐water Elimination and Brain Tissue Microstructure Characterization. Magnetic Resonance in Medicine, 80, 2155-2172. [Google Scholar] [CrossRef] [PubMed]
[27] Zhao, Y., Fu, C., Zhang, W., Ye, C., Wang, Z. and Ma, H. (2022) Automatic Segmentation of Cervical Cells Based on Star-Convex Polygons in Pap Smear Images. Bioengineering, 10, Article 47. [Google Scholar] [CrossRef] [PubMed]
[28] Chatterjee, A., Watson, G., Myint, E., Sved, P., McEntee, M. and Bourne, R. (2015) Changes in Epithelium, Stroma, and Lumen Space Correlate More Strongly with Gleason Pattern and Are Stronger Predictors of Prostate ADC Changes than Cellularity Metrics. Radiology, 277, 751-762. [Google Scholar] [CrossRef] [PubMed]
[29] Do, C.B. and Batzoglou, S. (2008) What Is the Expectation Maximization Algorithm? Nature Biotechnology, 26, 897-899. [Google Scholar] [CrossRef] [PubMed]
[30] Barral, M., Taouli, B., Guiu, B., Koh, D., Luciani, A., Manfredi, R., et al. (2015) Diffusion-Weighted MR Imaging of the Pancreas: Current Status and Recommendations. Radiology, 274, 45-63. [Google Scholar] [CrossRef] [PubMed]
[31] Hutter, J., Slator, P.J., Christiaens, D., Teixeira, R.P.A.G., Roberts, T., Jackson, L., et al. (2018) Integrated and Efficient Diffusion-Relaxometry Using Zebra. Scientific Reports, 8, Article No. 15138. [Google Scholar] [CrossRef] [PubMed]
[32] Kjaer, L., Thomsen, C., Iversen, P. and Henriksen, O. (1987) In Vivo Estimation of Relaxation Processes in Benign Hyperplasia and Carcinoma of the Prostate Gland by Magnetic Resonance Imaging. Magnetic Resonance Imaging, 5, 23-30. [Google Scholar] [CrossRef] [PubMed]
[33] Panagiotaki, E., Walker-Samuel, S., Siow, B., Johnson, S.P., Rajkumar, V., Pedley, R.B., et al. (2014) Noninvasive Quantification of Solid Tumor Microstructure Using VERDICT MRI. Cancer Research, 74, 1902-1912. [Google Scholar] [CrossRef] [PubMed]
[34] White, N.S., Leergaard, T.B., D'Arceuil, H., Bjaalie, J.G. and Dale, A.M. (2012) Probing Tissue Microstructure with Restriction Spectrum Imaging: Histological and Theoretical Validation. Human Brain Mapping, 34, 327-346. [Google Scholar] [CrossRef] [PubMed]
[35] Westin, C., Szczepankiewicz, F., Pasternak, O., Özarslan, E., Topgaard, D., Knutsson, H., et al. (2014) Measurement Tensors in Diffusion MRI: Generalizing the Concept of Diffusion Encoding. In: Golland, P., Hata, N., Barillot, C., Hornegger, J. and Howe, R., Eds., Medical Image Computing and Computer-Assisted Intervention—MICCAI 2014, Springer, 209-216. [Google Scholar] [CrossRef] [PubMed]
[36] Lemberskiy, G., Rosenkrantz, A.B., Veraart, J., Taneja, S.S., Novikov, D.S. and Fieremans, E. (2017) Time-Dependent Diffusion in Prostate Cancer. Investigative Radiology, 52, 405-411. [Google Scholar] [CrossRef] [PubMed]

为你推荐