Advances in Clinical Medicine
Vol. 13  No. 10 ( 2023 ), Article ID: 73267 , 6 pages
10.12677/ACM.2023.13102146

影像组学在肝细胞癌及肝硬化相关结节方面的研究进展

姚翠翠,袁振国*

山东第一医科大学附属省立医院放射科,山东 济南

收稿日期:2023年8月26日;录用日期:2023年9月19日;发布日期:2023年9月28日

摘要

肝细胞癌(Hepatocellular carcinoma, HCC)是危害人类生命的最常见的恶性肿瘤之一。随着医学成像技术的不断完善,人们可以早期发现病变并进行干预,为患者提供最大的生存机会。基于影像组学的最新进展,其为肝细胞癌(Hepatocellular Carcinoma, HCC)与肝硬化相关结节的准确鉴别及肝癌的早期精准诊断提供了可能性。本综述旨在分析肝硬化相关结节鉴别、肝癌的诊断、影像组学先进技术,通过对已发表文献的回顾,对肝硬化相关结节的鉴别、原发性肝癌的发展、分级、分子生物表达进行分析总结,以阐明影像组学在改善肝癌患者预后,提高患者生存率方面的作用。

关键词

肝细胞癌,放射组学,肝硬化相关结节分级

Research Progress of Radiomics in Hepatocellular Carcinoma and Cirrhosis-Related Nodules

Cuicui Yao, Zhenguo Yuan*

Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan Shandong

Received: Aug. 26th, 2023; accepted: Sep. 19th, 2023; published: Sep. 28th, 2023

ABSTRACT

Hepatocellular carcinoma (HCC) is one of the most common malignant tumors that harm human life. With the development of the medical imaging technology, people can detect lesions early and intervene which gives patients the greatest chance of survival. Based on the latest advances in radiomics, it provides the possibility for the accurate identification of nodules associated with Hepatocellular Carcinoma (HCC) and accurate diagnosis of liver cancer. The purpose of this review is to analyze the identification of cirrhosis associated nodules, the diagnosis of liver cancer, and the advanced techniques of imaging radiomics. Through a review of the published literature, analyzing and summarizing the identification of cirrhosis related nodules, the development, classification and molecular biological expression of primary liver cancer, to elucidate the role of imaging radiomics in improving the prognosis and survival rate of patients with liver cancer.

Keywords:Hepatocellular Carcinoma, Radiomics, Cirrhosis-Related Nodule Classification

Copyright © 2023 by author(s) and Hans Publishers Inc.

This work is licensed under the Creative Commons Attribution International License (CC BY 4.0).

http://creativecommons.org/licenses/by/4.0/

1. 引言

目前对于影像图像的观察,主要依赖于影像科医师肉眼所观察到的信号、大小、边缘等表观信息;而随着计算机技术的进步,影像组学现在可以以客观、可重复和高通量的方式将图像信息转换成许多人眼无法识别的量化特征 [1] 。影像组学工作流程是多学科及多步骤的,涉及到放射科医生、数据及计算机应用,并遵循病变分割、图像预处理、特征提取、模型构建与验证的逐步过程。实践表明放射组学在肿瘤实践中为患者无病生存、转移预测和治疗反应评估相关的精准医疗提供了新的视角。在分割之前对图像进行预处理,可以对肿瘤、肿瘤亚区域或肿瘤周围区域进行分割;而提取的特征通常描述感兴趣区域内像素的信号强度分布和空间关系;同时删除冗余和不可重现的特征,以提高模型性能及减少过度拟合。验证对于估计数据中的模型性能至关重要,并且可以通过使用内部或外部数据在数据集样本(交叉验证)或单独的保留数据集上迭代执行。

2. 影像组学在鉴别肝细胞癌与肝硬化相关结节的作用

肝细胞癌是最常见的恶性肿瘤之一,几乎80%的HCC发生在肝硬化患者中 [2] [3] 。肝硬化中的肝癌的发生通常表现为从良性结节、早期HCC和进展性HCC的多阶段进展 [4] 。HCC的早期发现,与良性肝硬化结节的鉴别,为患者提供了最大的长期生存机会。然而,由于影像学特征的重叠及不典型性,这些结节的完整表征仍然是一个困难的诊断过程 [5] [6] 。根据美国肝病研究协会的标准,动脉增强后(门脉期或平衡期)洗脱被定义为HCC的决定性诊断 [7] 。然而,这种典型的增强模式并不总是存在,特别是对于一些分化良好或小的HCC [8] [9] 。扩散加权成像可以通过提高诊断敏感性为常规动态MRI提供额外的价值 [10] 。扩散受限反映了组织的高细胞性,有助于HCC的诊断 [11] 。然而,一些小型肝细胞癌可能并没有扩散受限。近年来,加氧酸增强磁共振成像(Gd-EOB-MRI)作为一种肝细胞特异性摄取剂磁共振成像,可提供早期动态血管期和延迟肝胆期(HBP)信息,越来越多地应用于肝结节的表征。因肝胆期特异性低信号,Gd-EOB-MRI被证明在检测HCC时比常规动态MRI具有更高的灵敏度 [5] [12] [13] 。然而,日常影像诊断过程中,小肝细胞癌的特征性信号改变并不是普遍存在,例如一些小的肝癌在HBP图像上可能不显示低信号;相反,一些不典型增生结节却表现为低信号 [5] 。由于结节的生长阶段,不同影像图像上可有细微差别,普通影像学方法观察到的信息有限,且受诊断医师主观影响较大,故对于s-HCC及DNs的定性与精准鉴别存在一定难度。

Jiang [14] 等人通过钆酸增强磁共振构建一个基于多序列的三维肿瘤放射组学模型,并与欧洲肝脏研究协会(EASL)和美国肝病研究协会(LI-RADS)标准进行比较,结果显示影像组学的方法可以降低LI-RADS标准对影像特征评估的主观性。基于影像图像的纹理分析(TA)是良恶性疾病鉴别诊断的一种后处理方法 [15] 。Mokrane利用机器学习技术,对CT增强图像进行分析与处理,结果显示该方法可以无创性对患者病灶进行监控,使患者免于活检的不适和潜在发病率(EASL指南),病制定积极的随访策略 [16] 。Zhong等人的研究结果表明,基于扩散加权图像的纹理分析可以识别肝硬化背景下不典型强化的小肝癌和不典型增生结节,其疗效优于扩散加权图像的定性和定量分析;在肝硬化背景下区分小肝癌和不典型增生结节方面,基于T2WI的纹理分析(TA)优于加氧酸增强MR成像(Gd-EOB-MRI)和扩散加权成像(DWI);同时基于MRI (T1WI、T2WI及ADC图)的纹理分析对于肝脏成像报告和数据系统在鉴别小肝癌与良性结节方面具有附加价值 [17] [18] [19] 。相等人从平扫及增强图像中提取影像特征并分别建立了逻辑回归及支持向量机模型,结果显示基于T1WI增强期相所建立的两类模型在训练组及验证组的AUC值均大于0.96,从而得出T1WI增强期相是鉴别小肝癌与不典型增生结节的最优序列以及支持向量机分类器模型稳定性最佳的结论 [20] 。

3. 影像组学在预测肝细胞癌组织学分级中的应用

目前临床获得病变的病理及组织学分级只能在有创的条件下进行,而随着计算机技术的发展,影像组学使得临床在术前无创获得病理相关信息成为一种可能。基于影像组学可靠的术前分级可以帮助患者制定治疗计划,还可以有效降低复发率和不良反应的发生率 [21] 。Wu等人提取T1WI及T2WI序列的放射组学特征,并将其与临床因素相结合构建联合模型,评估了放射组学特征模型(基于T1WI、T2WI和T1WI与T2WI联合图像)、临床模型和临床与放射组学联合特征模型的预测价值,他们的结果表明,基于T1WI或T2WI图像的放射组学特征在预测肝细胞癌分级方面表现出相似的性能 [22] 。Zhou等人选择动脉期进行纹理分析,以直接表征均值及灰度行程不均匀(GLN)索引形态结构,结果表明Gd-DTPA增强磁共振图像中由均值和GLN标记的纹理特征与肝细胞癌的组织学分级相关 [23] 。Mao等人对动脉期及肝胆期进行放射特征的提取,分别及两两联合构建了神经网络及逻辑回归模型,并进行比较,发现磁共振图像动脉期和肝胆期的一些直方图特征低级别肿瘤明显高于高级别肿瘤,而神经网络在预测肝细胞癌组织学分级方面优于逻辑回归模型,同时动脉期及肝胆期无论是在神经网络模型亦或是在逻辑回归模型在区分肝细胞癌病理分级方面无明显差异 [24] 。Feng等人对T2WI和DCE的动脉期和门静脉期的直方图衍生特征进行研究,结果表明动脉期成像信号强度越高,分化程度越高 [25] 。Choi等人在肿瘤最大直径的T2WI、动脉期、ADC图上提取纹理特征,在T2WI的ROC曲线分析中,熵是HCC组织学分级的潜在预测参数,以此证明了T2WI在肝细胞癌组织学分级预测中的可行性 [26] 。Geng等人证明了从磁敏感图像中提取放射组学特征来评价肝细胞癌病理分级的可行性 [27] 。赵等人提取了动脉期、静脉期和延迟期磁共振图像的影像组学特征,并通过逻辑回归模型计算影像组学评分,得出了静脉期模型更能反映肿瘤异质性的结论 [28] 。Brancato等人术前对T2WI、动脉期、门脉期及延迟期图像构建放射组学模型,对肝细胞癌进行了病变及病理分级的预测,得出HCC分级预测贡献最大的组织特征是GLCM、GLDM和GLSZM特征的结论 [29] 。Han等人从两个研究中心获得图像样本,从T1WI、T2WI、门静脉期及肝胆期提取特征,联合临床特征构建了8种模型,在预测肝细胞癌组织学分级中,只有HBP模型AUC最好,这进一步说明了彰显了Gd-EOB-MRI不可替代的价值 [30] 。近年研究发现大梁–巨块(MTM)亚型是一种与血管生成相关的肝细胞癌形态变异,治疗后预后不良。Feng等人构建了一个基于机器学习的CT影像组学模型来无创性地预测肝癌患者的MTM亚型,其基于散装和单细胞RNA的测序揭示了与放射组学模型相关的潜在免疫渗透模式主要是体液免疫缺陷,为下一步肝癌的治疗提供了新的思路 [31] 。

4. 影像组学在预测肝细胞癌Ki-67中的应用

Wu等人的研究先后表明放射组学特征可以有效识别肝细胞癌中Ki-67的高低 [32] ,并将CT放射组学特征与临床因素相结合,建立了诺模图预测了肝细胞癌Ki-67的表达,同时发现在临床特征中甲胎蛋白水平及肝细胞癌病理分级是Ki-67蛋白表达的独立预测因子 [33] 。Ye等人通过一项前瞻性研究,从多模态磁共振图像中提取纹理特征,构建了纹理特征模型、临床模型及组合诺莫图,结果表明将纹理特征及临床因素结合起来可以提高对Ki-67预测性能的潜力 [34] 。Li等人从肝胆期(HBP)、T2WI、动脉期(AP)、门静脉期(PVP)提取纹理特征,得出肝胆期、动脉期和门静脉期的纹理分析有助于预测Ki67的表达 [35] 。Fan等人进行了同样的研究,结果表明基于动脉期的放射组学模型比肝胆期及T2WI的放射组学模型有更多的净效益,具有更好的预测性能 [36] 。Dong等人基于全氟丁烷造影剂增强超声造影图像的Kupffer-phase特征构建放射学模型,得出基于S-CEUS图像的Kupffer期构建的放射组学模型具有预测HCC患者Ki-67表达和组织学分期的潜力 [10] 。

5. 挑战与展望

基于获得的初步结果,影像组学可能是肝细胞癌患者个性化治疗的合适工具。这种方法可以在非侵入性的情况下补充或取代肿瘤活检,也可以用于开发新的预后生物标志物,对肝细胞癌的检测和分级有重要作用,而不需要侵入性手术。然而,将放射组学的结果转化为临床应用是很困难的,这主要是由于放射组学工作流程缺乏标准化以及由此导致的肝细胞癌放射组学研究的异质性。此外,由于大多数放射组学模型的学习是回顾性的,并不总是有说服力,所以只使用解释性模型仍然是不明智的,因为没有一个模型适用于所有的临床决策场景 [31] 。在未来,研究人员必须继续进行前瞻性研究,以验证本综述中描述的方法的临床实用性,此外还需对更一致的患者样本进行分析,这将使在验证集上验证模型和测试不同的机器学习模型成为可能。此外,对于肝细胞癌的诊断和分级,建立可重复和可解释的放射学标记物,并将放射学数据与临床/实验室信息和其他组学数据(如基因组和病理数据)结合起来,将非常重要 [13] [26] 。不同尺度(放射学、病理学、分子学)定量数据的整合肯定会提高对肝细胞癌的诊断和分子知识,这将对临床决策过程产生直接影响。此外,这可能有助于在临床实践中验证作为“虚拟活检”的放射学方法,并发现基因型–表型相关性 [27] 。

文章引用

姚翠翠,袁振国. 影像组学在肝细胞癌及肝硬化相关结节方面的研究进展
Research Progress of Radiomics in Hepatocellular Carcinoma and Cirrhosis-Related Nodules[J]. 临床医学进展, 2023, 13(10): 15335-15340. https://doi.org/10.12677/ACM.2023.13102146

参考文献

  1. 1. Jiang, L., Miao, Z., Chen, H., et al. (2023) Radiomics Analysis of Diffusion-Weighted Imaging and Long-Term Unfa-vorable Outcomes Risk for Acute Stroke. Stroke, 54, 488-498. https://doi.org/10.1161/STROKEAHA.122.040418

  2. 2. El-Serag, H.B., Davila, J.A., Petersen, N.J., et al. (2003) The Continuing Increase in the Incidence of Hepatocellular Carcinoma in the United States: An Update. Annals of Inter-nal Medicine, 139, 817. https://doi.org/10.7326/0003-4819-139-10-200311180-00009

  3. 3. McGlynn, K.A. and London, W.T. (2005) Epi-demiology and Natural History of Hepatocellular Carcinoma. Best Practice & Research Clinical Gastroenterology, 19, 3-23. https://doi.org/10.1016/j.bpg.2004.10.004

  4. 4. Choi, B.I., Takayasu, K. and Han, M.C. (1993) Small Hepa-tocellular Carcinomas and Associated Nodular Lesions of the Liver: Pathology, Pathogenesis, and Imaging Findings. American Journal of Roentgenology, 160, 1177-1187. https://doi.org/10.2214/ajr.160.6.8388618

  5. 5. Di Martino, M., Anzidei, M., Zaccagna, F., et al. (2016) Qualitative Analysis of Small (≤ 2 cm) Regenerative Nodules, Dysplastic Nodules and Well-Differentiated HCCs with Gadoxetic Acid MRI. BMC Medical Imaging, 16, Article No. 62. https://doi.org/10.1186/s12880-016-0165-5

  6. 6. Park, H.J., Choi, B.I., Lee, E.S., et al. (2018) How to Differentiate Borderline Hepatic Nodules in Hepatocarcinogenesis: Emphasis on Imaging Diagnosis. Liver Cancer, 6, 189-203. https://doi.org/10.1159/000455949

  7. 7. Bruix, J. and Sherman, M. (2011) Management of Hepatocellular Carcinoma: An Update. Hepatology, 53, 1020-1022. https://doi.org/10.1002/hep.24199

  8. 8. Jang, H.J., Kim, T.K., Burns, P.N. and Wilson, S.R. (2007) Enhancement Patterns of Hepatocellular Carcinoma at Contrast-Enhanced Us: Comparison with Histologic Differentiation. Radiology, 244, 898-906. https://doi.org/10.1148/radiol.2443061520

  9. 9. Li, C.S., Chen, R.C., Tu, H.Y., et al. (2006) Imaging Well-Differentiated Hepatocellular Carcinoma with Dynamic Triple-Phase Helical Computed Tomography. The British Journal of Radiology, 79, 659-665. https://doi.org/10.1259/bjr/12699987

  10. 10. Park, M.S., Kim, S., Patel, J., et al. (2012) Hepatocellular Carcinoma: Detection with Diffusion-Weighted versus Contrast-Enhanced Magnetic Resonance Imaging in Pretransplant Patients. Hepatology, 56, 140-148. https://doi.org/10.1002/hep.25681

  11. 11. Taouli, B. and Koh, D.M. (2010) Diffusion-Weighted MR Imaging of the Liver. Radiology, 254, 47-66. https://doi.org/10.1148/radiol.09090021

  12. 12. Ahn, S.S., Kim, M.J., Lim, J.S., et al. (2010) Added Value of Gadoxetic Acid-Enhanced Hepatobiliary Phase MR Imaging in the Diagnosis of Hepatocellular Carcinoma. Radiology, 255, 459-466. https://doi.org/10.1148/radiol.10091388

  13. 13. Kawada, N., Ohkawa, K., Tanaka, S., et al. (2010) Im-proved Diagnosis of Well-Differentiated Hepatocellular Carcinoma with Gadolinium Ethoxybenzyl Diethylene Triamine Pentaacetic Acid-Enhanced Magnetic Resonance Imaging and Sonazoid Contrast-Enhanced Ultrasonography. Hepatology Research, 40, 930-936. https://doi.org/10.1111/j.1872-034X.2010.00697.x

  14. 14. Jiang, H., Liu, X., Chen, J., et al. (2019) Man or Machine? Prospective Comparison of the Version 2018 EASL, LI-RADS Criteria and a Radiomics Model to Diagnose Hepatocel-lular Carcinoma. Cancer Imaging, 19, Article No. 84. https://doi.org/10.1186/s40644-019-0266-9

  15. 15. Castellano, G., Bonilha, L., Li, L.M. and Cendes, F. (2004) Tex-ture Analysis of Medical Images. Clinical Radiology, 59, 1061-1069. https://doi.org/10.1016/j.crad.2004.07.008

  16. 16. Mokrane, F.Z., Lu, L., Vavasseur, A., et al. (2020) Radiomics Machine-Learning Signature for Diagnosis of Hepatocellular Carcinoma in Cirrhotic Patients with Indeterminate Liver Nodules. European Radiology, 30, 558-570. https://doi.org/10.1007/s00330-019-06347-w

  17. 17. Zhong, X., Tang, H., Lu, B., et al. (2020) Differentiation of Small Hepatocellular Carcinoma from Dysplastic Nodules in Cirrhotic Liver: Texture Analysis Based on MRI Improved Performance in Comparison over Gadoxetic Acid-En- hanced MR and Diffusion-Weighted Imaging. Frontiers in On-cology, 9, Article 1382. https://doi.org/10.3389/fonc.2019.01382

  18. 18. Zhong, X., Li, J.S., Chen, Z.J., Yin, J.X., Gui, S., Sun, Z.Q. and Tang, H.S. (2020) [Texture Analysis of Diffusion- Weighted Magnetic Resonance Imaging to Identify Atypically En-hanced Small Hepatocellular Carcinoma and Dysplastic Nodules under the Background of Cirrhosis]. Chinese Journal of Hepatology, 28, 37-42.

  19. 19. Zhong, X., Guan, T., Tang, D., et al. (2021) Differentiation of Small (≤ 3 cm) Hepatocellu-lar Carcinomas from Benign Nodules in Cirrhotic Liver: The Added Additive Value of MRI-Based Radiomics Analysis to LI-RADS Version 2018 Algorithm. BMC Gastroenterology, 21, Article No. 155. https://doi.org/10.1186/s12876-021-01710-y

  20. 20. 相悦. 影像组学在评估小肝癌与不典型增生结节的价值研究[D]: [硕士学位论文]. 大连: 大连医科大学, 2020. https://doi.org/10.26994/d.cnki.gdlyu.2021.000914

  21. 21. Ganesan, P. and Kulik, L.M. (2023) Hepatocellular Car-cinoma. Clinics in Liver Disease, 27, 85-102. https://doi.org/10.1016/j.cld.2022.08.004

  22. 22. Wu, M., Tan, H., Gao, F., et al. (2019) Predicting the Grade of Hepatocellular Carcinoma Based on Non-Contrast- Enhanced MRI Radiomics Signature. European Radiology, 29, 2802-2811. https://doi.org/10.1007/s00330-018-5787-2

  23. 23. Zhou, W., Zhang, L., Wang, K., et al. (2017) Malig-nancy Characterization of Hepatocellular Carcinomas Based on Texture Analysis of Contrast-Enhanced MR Images. Journal of Magnetic Resonance Imaging, 45, 1476-1484. https://doi.org/10.1002/jmri.25454

  24. 24. Mao, Y., Wang, J., Zhu, Y., et al. (2022) GD-EOB-DTPA-Enhanced MRI Radiomic Features for Predicting Histological Grade of Hepatocellular Carcinoma. Hepatobiliary Surgery and Nutrition, 11, 13-24. https://doi.org/10.21037/hbsn-19-870

  25. 25. Feng, M., Zhang, M., Liu, Y., et al. (2020) Texture Analysis of MR Images to Identify the Differentiated Degree in Hepatocellular Carcinoma: A Retrospective Study. BMC Cancer, 20, Ar-ticle No. 611. https://doi.org/10.1186/s12885-020-07094-8

  26. 26. Choi, J.M., Yu, J.S., Cho, E.S., et al. (2020) Texture Analysis of Hepatocellular Carcinoma on Magnetic Resonance Imaging: Assessment for Performance in Predicting Histopathologic Grade. Journal of Computer Assisted Tomography, 44, 901-910. https://doi.org/10.1097/RCT.0000000000001087

  27. 27. Geng, Z., Zhang, Y., Wang, S., et al. (2021) Radiomics Analysis of Susceptibility Weighted Imaging for Hepatocellular Carcinoma: Exploring the Correlation between Histo-pathology and Radiomics Features. Magnetic Resonance in Medical Sciences, 20, 253-263. https://doi.org/10.2463/mrms.mp.2020-0060

  28. 28. 赵莹, 刘爱连, 武敬君, 郭妍, 宋清伟, 李昕, 吴艇帆, 崔景景. 基于增强MRI影像组学术前预测肝细胞癌病理分化程度[J]. 中国医学影像学杂志, 2021, 29(6): 570-576.

  29. 29. Brancato, V., Garbino, N., Salvatore, M. and Cavaliere, C. (2022) MRI-Based Radiomic Features Help Identify Lesions and Predict Histopathological Grade of Hepatocellular Carcinoma. Diagnostics, 12, Article 1085. https://doi.org/10.3390/diagnostics12051085

  30. 30. Han, Y.E., Cho, Y., Kim, M.J., et al. (2022) Hepatocellular Car-cinoma Pathologic Grade Prediction Using Radiomics and Machine Learning Models of Gadoxetic Acid-Enhanced MRI: A Two-Center Study. Abdominal Radiology, 48, 244-256. https://link.springer.com/10.1007/s00261-022-03679-y

  31. 31. Feng, Z., Li, H., Liu, Q., et al. (2023) CT Radiomics to Predict Macrotrabecular-Massive Subtype and Immune Status in Hepatocellular Carcinoma. Radiology, 307, e221291. https://doi.org/10.1148/radiol.221291

  32. 32. Wu, H., Han, X., Wang, Z., et al. (2020) Prediction of the Ki-67 Marker Index in Hepatocellular Carcinoma Based on CT Radiomics Features. Physics in Medicine & Biology, 65, Article ID: 235048. https://doi.org/10.1088/1361-6560/abac9c

  33. 33. Wu, C., Chen, J., Fan, Y., et al. (2022) Nomogram Based on CT Radiomics Features Combined with Clinical Factors to Predict Ki-67 Expression in Hepatocellular Carcinoma. Frontiers in Oncology, 12, Article 943942. https://doi.org/10.3389/fonc.2022.943942

  34. 34. Ye, Z., Jiang, H., Chen, J., et al. (2019) Texture Analysis on Gadoxetic Acid Enhanced-MRI for Predicting Ki-67 Status in Hepatocellular Carcinoma: A Prospective Study. Chinese Journal of Cancer Research, 31, 806-817. https://doi.org/10.21147/j.issn.1000-9604.2019.05.10

  35. 35. Li, Y., Yan, C., Weng, S., et al. (2019) Texture Analy-sis of Multi-Phase MRI Images to Detect Expression of Ki67 in Hepatocellular Carcinoma. Clinical Radiology, 74, 813.e19-813.e27.

  36. 36. Fan, Y., Yu, Y., Wang, X., Hu, M.J. and Hu, C.H. (2021) Radiomic Analysis of GD-EOB-DTPA-Enhanced MRI Predicts Ki-67 Expression in Hepatocellular Carcinoma. BMC Medical Imaging, 21, Article No. 100. https://doi.org/10.1186/s12880-021-00633-0

  37. NOTES

    *通讯作者。

期刊菜单