传统影像与纹理分析对术前肾癌Fuhrman分级评估的应用价值
The Application Value of Traditional Image and Texture Analysis on Preoperative Fuhrman Grading Evaluation of Renal Carcinoma
DOI: 10.12677/ACM.2024.141080, PDF, HTML, XML, 下载: 166  浏览: 223 
作者: 妥亭萱, 赵 圆*:新疆医科大学第一附属医院影像中心,新疆 乌鲁木齐
关键词: 肾癌计算机断层扫描磁共振成像纹理分析Renal Carcinoma Computed Tomography Magnetic Resonance Imaging Texture Analysis
摘要: 肾癌是常见的泌尿系恶性肿瘤,肾癌的Fuhrman分级在诊疗过程中至关重要,因此能否在对肿瘤进行非侵袭性方法中准确评估其分级,对患者最优治疗方案的选择及预后都有着很大的影响。传统影像学检查包括计算机断层扫描(CT)、磁共振成像(MRI)及超声等技术,但这些技术在对肿瘤的检出及判断等方面存在一定的局限性,近年来影像组学技术中的纹理分析广泛应用于多个系统疾病的诊疗过程,为提高认识,本文以传统影像技术与纹理分析的方法在评估肾癌Fuhrman分级中应用的价值展开综述。
Abstract: Renal carcinoma is a common malignant tumor of urinary system. The Fuhrman grade of renal car-cinoma is very important in the process of diagnosis and treatment. Therefore, whether the grade can be accurately evaluated in the non-invasive method of tumor has a great impact on the selection of the optimal treatment plan and prognosis of patients. Traditional imaging examinations include computed tomography (CT), magnetic resonance imaging (MRI), ultrasound and other technologies, but these technologies have certain limitations in the detection and judgment of tumors. In recent years, texture analysis in imaging omics technology has been widely used in the diagnosis and treatment of multiple systems of diseases. This article reviews the value of traditional imaging techniques and texture analysis in evaluating Fuhrman grading of renal carcinoma.
文章引用:妥亭萱, 赵圆. 传统影像与纹理分析对术前肾癌Fuhrman分级评估的应用价值[J]. 临床医学进展, 2024, 14(1): 579-584. https://doi.org/10.12677/ACM.2024.141080

1. 引言

肾细胞癌,简称肾癌,是泌尿系统常见恶性肿瘤,对于肾癌诊断和治疗中最重要的任务之一是肿瘤分期和分级。Fuhrman分级系统在临床肿瘤学界得到广泛认可 [1] 。目前的研究表明,Fuhrman核分级与肿瘤的生长速度和患者的预后密切相关 [2] [3] [4] ,是体现肿瘤侵袭性和预后较重要的独立指标之一 [5] ,在肿瘤细胞核的形状、大小、染色以及有无核仁基础上进行病理等级。其中,I~II级为低,III~IV级为高 [6] 。高级别肿瘤具有更高的侵袭性能力、更高的转移可能性和较差的预后 [7] 。在过去十年中,已经设计过许多肾癌的无创治疗策略,包括射频消融、冷冻消融和主动监测,却仍然缺乏使用这些无创/微创治疗方法对患者进行管理的适当标准,因为大多数患者通常在诊断后接受手术治疗,而如今个体化治疗策略越来越被重视,例如根治性方法(例如手术)用于侵袭性或高级别肿瘤(III, IV),而保守治疗(例如主动监测)用于低级别(I, II)病变 [1] 。肾肿块活检可根据Fuhrman核分级进行风险分层,从而帮助选择适合进行主动监测的患者。此外,保留肾单位的手术越来越多地用于治疗T1期肾癌,但Fuhrman分级高的T1期肿瘤具有更高的复发及恶性可能性 [8] 。因此,术前确认肾癌的Fuhrman分级对于评估肿瘤侵袭性、做出合理临床决策、选择手术方案及评价预后都至关重要。然而肾肿块活检常会引起并发症,例如出血、感染等,所以基于医学图像的自动、无创和可重现的肾癌分级系统在不断地被探讨、研究及开发 [9] 。由此可见,基于医学影像评估肾癌分级至关重要,并且还可以在对影像学特征进行细致分析的基础上构建管理建议 [10] 。

2. 使用须各类传统影像对肾癌Fuhrman分级评估的优势

术前评估肾癌的病理分级可以采用非侵入性成像的方法。一些传统的放射学特征,如肿瘤大小和计算机断层扫描(computed tomography, CT)增强模式已被证明与肿瘤分级相关 [11] 。有研究表明,肿块体积小的均质肾癌随着组织学分级的增加,内部异质性区域更为普遍,这种异质性可能是由于高级别肿瘤的瘤内坏死、出血、纤维化或囊性变性。未增强扫描中肿瘤衰减低是肿瘤分级低的另一个重要预测因素。已经证明,具有较低组织学分级的肾透明细胞癌显示出典型的透明细胞外观,而随着组织学等级的增加,细胞质的透明细胞特征和脂滴的数量会相应减少。有研究证明,小肾癌的CT成像特征预测肿瘤分级的准确率可超过70%,若将CT成像特征与其他临床变量相结合有望产生更准确的预测模型 [10] 。MRI目前在疑似RCC的诊断和术前计划中起着解决问题的作用,特别是对于识别肾脏病变内的增强软组织。此外,肾癌在MRI的独特影像学特征包括T2低信号、动态对比增强MRI各期的明显低强化以及对期成像信号丢失等 [12] 。由于水的运动程度受细胞或组织组织的影响,因此ADC值被认为与透明细胞肾癌的Fuhrman等级有关 [13] 。另外,Furhman分级与肿瘤代谢和大小呈显著正相关表明,随着大小和代谢参数的增加,肿瘤可能变得更加代谢活跃和侵袭,从而增加其等级。肾癌中FDG摄取量高(SUVmax)与肿瘤侵袭性有关,并提示高级别,肿瘤分级和大小与SUVmax之间存在显著相关性。F-18-FDG PET/CT扫描(SUV和MTV)上的代谢参数似乎是预测肾细胞癌等级的重要标志物。还有研究发现SUVmax在区分高级别肿瘤方面具有高度特异性,准确率为89%,可以可靠地将等级分为低和高 [14] 。

3. 纹理分析概念和原理

随着医学影像技术的发展,纹理分析已被用于肾癌分级的诊断和评估。纹理分析是一种从医学图像中自动提取定量特征的技术 [15] ,纹理分析通过评估灰度值的分布和空间关系,在像素水平上评估肿瘤异质性 [16] ,自动挖掘许多难以用肉眼识别的定量图像特征,并揭示具有潜在预后相关性的肿瘤内异质性的各个方面 [17] 。与活检相比,纹理分析有几个优点:可以获得整个肿瘤和微环境的空间信息,从而考虑肿瘤内的异质性和与微环境的相互作用;通过非侵入性方法获得信息 [18] 。MRTA作为一种无辐射的方式,将来可能会更广泛地使用,特别是在儿童和随访中 [16] 。获取纹理分析的方法中最常用的统计方法是基于图像的一阶直方图,这种方法是通过像素的强度来计算的,只需要考虑像素灰度的分布。当直方图类似于正太分布的时候,一、二阶统计量更有效。若直方图不是正太分布的时候,则需要高阶统计量来充分描述其分布。

4. 纹理分析在肿瘤中的应用

4.1. MRTA在临床肿瘤性疾病中的应用

从传统放射学特征获得的有限信息预测肾癌的病理分级一直具有挑战性。相比之下,涉及人眼无法辨别的数据的计算机化提取的纹理分析可以产生关于肿瘤纹理,形状和图像强度等。这种方法已成功用于癌症研究,具有识别肿瘤表型,病理分级和生物学行为的潜力。因此,纹理分析是一种潜在的有用方法,不仅可以用于评估肿瘤异质性,还可以用于评估病理分级以指导个性化癌症治疗 [11] 。近年来,MRTA在肿瘤性的疾病及非肿瘤性的疾病的诊疗中有大量应用,比如基因型的预测、淋巴结的转移、生存预后和治疗的反应评估等方面的应用,其中对肿瘤的研究一直是很多学者的关注焦点,在预测肿瘤的病理特征、治疗和预后等领域表现出良好的前景 [19] 。鉴于放射组学的非侵入性,图像的纹理分析可以表现为虚拟的活检,放射组学在肿瘤学中的主要诊断任务是采用无创的诊断方法来准确的鉴别肿瘤的良恶性 [20] 。MRTA的影像特征可以在一定的程度上反映并且放大肿瘤的异质性,并与生物异质性的不同组成部分相关,这有很大的潜力用于识别那些使用标准的治疗方法却结果较差的患者,甚至有可能改变肿瘤的治疗方式,并通过帮助预测肿瘤行为和对治疗的反应促进个性化治疗的进步,MRTA直方图分析在膀胱癌、乳腺癌、直肠癌等方面均有突显的优势。

4.2. MRI纹理分析在肾癌中的应用

多参数磁共振成像及其衍生体征和参数可鉴别肾癌和RCC,但诊断效果因文献而异。最近出现的影像组学和纹理分析(TA)为评估肿瘤特征引入了新的范式。纹理分析是一种图像处理算法,用于量化病变异质性和人眼可能无法察觉的某些模式。近年来,越来越多的研究证明,肾脏成像的纹理特征具有预测肿瘤亚型、分期、核分级和预后的潜力 [21] 。少数初步研究评估了RCC亚型的MRI纹理分析,并显示出鉴别RCC亚型的良好的诊断性能,但这些研究要么样本量小,要么包括较宽的RCC分期,Wang [22] 等人通过研究得出MR纹理分析可能有助于区分小(≤4 cm)和非常小(≤2 cm) RCC亚型的结论,这种非侵入性方法可能为局部RCC治疗和监测策略提供更多信息。Ankur等人在2019年报道了包括T2WI上的熵(AUC 0.807)在内的几个纹理参数能够区分ccRCC和非ccRCC [23] 。后来,Wei等人发现MR影像组学参数可用于鉴别ccRCC和pRCC。在MR T0加权序列上使用基于机器学习的模型,他们的AUC为63.57,准确率为4.50%,灵敏度为0.81%,特异性为8.2% [24] 。Uyen等人使用多相造影剂增强MR纹理来区分小的肾脏肿块,如pRCC和ccRCC [25] 。

4.3. MRI纹理分析在其他泌尿系恶性肿瘤中的应用

泌尿系恶性肿瘤的组织病理资料对准确的诊断和预测预后都具有着重要的意义,最常见的是肾癌和膀胱癌,外科活检所选取的部分组织并不一定足以评价整个肿瘤的特征,这个时候就需要其他辅助工具来辅助诊断,比如MRI纹理分析。MRI对膀胱癌局部分期的评估主要依赖于T2加权序列的成像。具有推导的表观扩散系数(apparent diffusion coefficient, ADC)的DWI可能有助于评估生物学行为,包括细胞性和增殖潜力 [26] 。虽然目前不推荐所有接受过TURBT的膀胱癌患者进行MRI检查,但MRI可能仍有帮助 [27] 。已经探索了膀胱癌的T2加权和DWI MRI纹理分析在T2加权图像上区分正常膀胱壁和膀胱癌的作用 [28] 。Yang等人证明,定量多参数MRI可以准确识别基于顺铂的NAC敏感性,使MIBC的敏感性提高75% [29] ,这证明了基于MRI的定量分析在评估MIBC肿瘤分子异质性方面的能力。Meng [30] 等人提出与常规DWI相比,DKI可以提供更多的定量参数来反映肿瘤微观结构的变化,更好地反映肿瘤微观结构的复杂性,这样一来应用于DKI的全肿瘤纹理分析可能在区分NMIBC与MIBC以及低级别和高级别膀胱癌方面产生比传统ADC值更稳健的值。

5. 局限性和挑战性

尽管纹理分析已成为近年来广泛应用的图像数据后处理方法,可回顾性地应用于医学图像,以评估肿瘤复杂性、亚体素水平的异质性,并获得更多的病变信息 [30] ,但仍然存在一定的局限性。广泛使用影像组学的主要局限性在于,所执行的组织纹理分析类型、使用的分割类型、后处理方法以及纹理对象输出的数量和质量在不同平台和研究中差异很大,因此很难比较结果。目前,还没有统一的影像组学参数和组织纹理测量标准。尽管结果具有统计学意义,但已发表的数据存在很大差异 [31] 。另外还会与研究设计有关,事实上质构分析已大部分用于回顾性研究,需要在第一阶段为进一步提供概念验证调查。回顾性分析缺乏对数据采集和管理的完全控制,这会对结果的可重复性和稳健性产生负面影响 [32] 。此外,MRI通常不是实性肾肿块的首选影像学检查方式,并且由于T2加权成像中的切片厚度,存在非常小的肿瘤的部分体积效应。手动分割小肾肿瘤也可能影响纹理分析的重现性和可重复性。进一步使用更薄的切片图像可能会克服对小肿瘤的限制 [21] [33] [34] [35] [36] [37] 。

6. 小结与展望

纹理分析已在不同肿瘤类型领域进行了运用,包括脑胶质瘤、乳腺癌、软骨骨肿瘤、宫颈癌、膀胱癌和肾细胞癌等,纹理分析代表了一种越来越流行的后处理定量评估技术,可以潜在地用作诊断成像的辅助手段,以及可能的成像生物标志物 [38] 。

NOTES

*通讯作者。

参考文献

[1] Nazari, M., Shiri, I., Hajianfar, G., et al. (2020) Noninvasive Fuhrman Grading of Clear Cell Renal Cell Carcinoma Us-ing Computed Tomography Radiomic Features and Machine Learning. La radiologia medica, 125, 754-762.
https://doi.org/10.1007/s11547-020-01169-z
[2] Minardi, D., Lucarini, G., Mazzucchelli, R., Milanese, G., Natali, D., Galosi, A.B., et al. (2005) Prognostic Role of Fuhrman Grade and Vascular Endothelial Growth Factor in pT1a Clear Cell Carcinoma in Partial Nephrectomy Specimens. Journal of Urology, 174, 1208-1212.
https://doi.org/10.1097/01.ju.0000173078.57871.2d
[3] Li, X.S., Yao, L., Gong, K., Yu, W., He, Q., Zhou, L.Q., et al. (2012) Growth Pattern of Renal Cell Carcinoma (RCC) in Patients with Delayed Surgical Intervention. Journal of Cancer Research and Clinical Oncology, 138, 269-274.
https://doi.org/10.1007/s00432-011-1083-0
[4] Sahni, V.A. and Silverman, S.G. (2014) Imaging Management of Incidentally Detected Small Renal Masses. Seminars in Interventional Radiology, 31, 9-19.
https://doi.org/10.1055/s-0033-1363838
[5] Sun, J., Pan, L., Zha, T., Xing, W., Chen, J. and Duan, S. (2021) The Role of MRI Texture Analysis Based on Susceptibility-Weighted Imaging in Predicting Fuhrman Grade of Clear Cell Renal Cell Carcinoma. Acta Radiologica, 62, 1104-1111.
https://doi.org/10.1177/0284185120951964
[6] Fuhrman, S.A., Lasky, L.C. and Limas, C. (2004) Prognostic Significance of Morphologic Parameters in Renal Cell Carcinoma. International Journal of Clinical Practice, 58, 333-336.
https://doi.org/10.1111/j.1368-5031.2004.00008.x
[7] Donat, S.M., Diaz, M., Bishoff, J.T., Coleman, J.A., Dahm, P., Derweesh, I.H., et al. (2013) Follow-Up for Clinically Localized Renal Neoplasms: AUA Guideline. Journal of Urology, 190, 407-416.
https://doi.org/10.1016/j.juro.2013.04.121
[8] Suzuki, K., Mizuno, R., Mikami, S., Tanaka, N., Kanao, K., Kiku-chi, E., et al. (2012) Prognostic Significance of High Nuclear Grade in Patients with Pathologic T1a Renal Cell Carcino-ma. Japanese Journal of Clinical Oncology, 42, 831-835.
https://doi.org/10.1093/jjco/hys109
[9] Hussain, M.A., Hamarneh, G. and Garbi, R. (2021) Learnable Image Histograms-Based Deep Radiomics for Renal Cell Carcinoma Grading and staging. Computerized Medical Imaging and Graphics, 90, Article ID: 101924.
https://doi.org/10.1016/j.compmedimag.2021.101924
[10] Choi, S.Y., Sung, D.J., Yang, K.S., et al. (2016) Small (<4 cm) Clear Cell Renal Cell Carcinoma: Correlation between CT Findings and Histologic Grade. Abdominal Radiology, 41, 1160-1169.
https://doi.org/10.1007/s00261-016-0732-9
[11] Yi, X., Xiao, Q., Zeng, F., Yin, H., Li, Z., Qian, C., Wang, C., Lei, G., Xu, Q., Li, C., Li, M., Gong, G., Zee, C., Guan, X., Liu, L. and Chen, B.T. (2021) Computed Tomography Radiomics for Predicting Pathological Grade of Renal Cell Carcinoma. Frontiers in Oncology, 10, Article 570396.
https://doi.org/10.3389/fonc.2020.570396
[12] Chiarello, M.A., Mali, R.D. and Kang, S.K. (2018) Diag-nostic Accuracy of MRI for Detection of Papillary Renal Cell Carcinoma: A Systematic Review and Meta-Analysis. American Journal of Roentgenology, 211, 812-821.
https://doi.org/10.2214/AJR.17.19462
[13] Woo, S., Suh, C.H., Kim, S.Y., Cho, J.Y. and Kim, S.H. (2017) Diag-nostic Performance of DWI for Differentiating High- From Low-Grade Clear Cell Renal Cell Carcinoma: A Systematic Review and Meta-Analysis. American Journal of Roentgenology, 209, W374-W381.
https://doi.org/10.2214/AJR.17.18283
[14] Singh, H., Arora, G., Nayak, B., Sharma, A., Singh, G., Kumari, K., Jana, S., Patel, C., Pandey, A.K., Seth, A. and Kumar, R. (2020) Semi-Quantitative F-18-FDG PET/Computed Tomog-raphy Parameters for Prediction of Grade in Patients with Renal Cell Carcinoma and the Incremental Value of Diuretics. Nuclear Medicine Communications, 41, 485-493.
https://doi.org/10.1097/MNM.0000000000001169
[15] Yu, W., Liang, G., Zeng, L., Yang, Y. and Wu, Y. (2021) Accuracy of CT Texture Analysis for Differentiating Low-Grade and High-Grade Renal Cell Carcinoma: Systematic Review and Meta-Analysis. BMJ Open, 11, e051470.
https://doi.org/10.1136/bmjopen-2021-051470
[16] Goyal, A., Razik, A., Kandasamy, D., et al. (2019) Role of MR Texture Analysis in Histological Subtyping and Grading of Renal Cell Carcinoma: A Preliminary Study. Abdominal Ra-diology, 44, 3336-3349.
https://doi.org/10.1007/s00261-019-02122-z
[17] Xing, J., Liu, Y., Wang, Z., Xu, A., Su, S., Shen, S. and Wang, Z. (2023) Incremental Value of Radiomics with Machine Learning to the Existing Prognostic Models for Predicting Out-come in Renal Cell Carcinoma. Frontiers in Oncology, 13, Article 1036734.
https://doi.org/10.3389/fonc.2023.1036734
[18] Greco, F., Beomonte Zobel, B., Di Gennaro, G. and Mallio, C.A. (2023) Advanced CT Imaging, Radiomics, and Artificial Intelligence to Evaluate Immune Checkpoint Inhibitors’ Effects on Metastatic Renal Cell Carcinoma. Applied Sciences, 13, Article 3779.
https://doi.org/10.3390/app13063779
[19] Yu, G., Mao, N., Song, X., et al. (2022) Radiomics Model for Predict-ing TP53 Status Using CT and Machine Learning Approach in Laryngeal Squamous Cell Carcinoma. Frontiers in On-cology, 12, Article 823428.
https://doi.org/10.3389/fonc.2022.823428
[20] Litvin, A.A., Burkin, D.A., Kropinov, A.A., et al. (2021) Radi-omics and Digital Image Texture Analysis in Oncology (Review). Sovremennye Tehnologii v Medicine, 13, 97-104.
https://doi.org/10.17691/stm2021.13.2.11
[21] Gillies, R.J., Kinahan, P.E. and Hricak, H. (2016) Radiomics: Im-ages Are More than Pictures, They Are Data. Radiology, 278, 563-577.
https://doi.org/10.1148/radiol.2015151169
[22] Wang, Y., Zhang, X., Zhang, J., et al. (2023) MR Texture Analysis in Differentiation of Small and Very Small Renal Cell Carcinoma Subtypes. Abdominal Radiology, 48, 1044-1050.
https://doi.org/10.1007/s00261-022-03794-w
[23] Ankur, G., Abdul, R., Devasenathipathy, K., et al. (2019) Role of MR Texture Analysis in Histological Subtyping and Grading of Renal Cell Carcinoma: A Preliminary Study. Ab-dominal Radiology, 44, 3336-3349.
https://doi.org/10.1007/s00261-019-02122-z
[24] Wei, W., Cao, K.M., Jin, S.M., et al. (2020) Differentiation of Renal Cell Carcinoma Subtypes through MRI-Based Radiomics Analysis. European Radiology, 30, 5738-5747.
https://doi.org/10.1007/s00330-020-06896-5
[25] Uyen, N.H., Mirmomen, S.M., Osorio, M., et al. (2018) As-sessment of Multiphasic Contrast-Enhanced MR Textures in Differentiating Small Renal Mass Subtypes. Abdominal Ra-diology, 43, 3400-3409.
https://doi.org/10.1007/s00261-018-1625-x
[26] Lim, C.S., Tirumani, S., van der Pol, C.B., Alessandrino, F., Sonpavde, G.P., Silverman, S.G. and Shinagare, A.B. (2019) Use of Quantitative T2-Weighted and Apparent Diffusion Coefficient Texture Features of Bladder Cancer and Extravesical Fat for Local Tumor Staging after Transurethral Resec-tion. American Journal of Roentgenology, 212, 1060-1069.
https://doi.org/10.2214/AJR.18.20718
[27] van der Pol, C.B., Shinagare, A.B., Tirumani, S.H., Preston, M.A., Vangel, M.G. and Silverman, S.G. (2018) Bladder Cancer Local Staging: Multiparametric MRI Performance following Transurethral Resection. Abdominal Radiology, 43, 2412-2423.
https://doi.org/10.1007/s00261-017-1449-0
[28] Xu, X., Zhang, X., Tian, Q., et al. (2017) Three-Dimensional Texture Features from Intensity and High-Order Derivative Maps for the Discrimination between Bladder Tumors and Wall Tissues via MRI. International Journal of Computer Assisted Radiology and Surgery, 12, 645-656.
https://doi.org/10.1007/s11548-017-1522-8
[29] Yang, X., Yuan, B.R., Zhang, Y.D., Zhuang, J.T., Cai, L.K., Wu, Q.K., Cao, Q., Li, P.C., Lu, Q. and Sun, X.Y. (2022) Quantitative Multiparametric MRI as a Promising Tool for the As-sessment of Early Response to Neoadjuvant Chemotherapy in Bladder Cancer. European Journal of Radiology, 157, Ar-ticle ID: 110587.
https://doi.org/10.1016/j.ejrad.2022.110587
[30] Meng, X., Li, S., He, K., Hu, H., Feng, C., Li, Z. and Wang, Y. (2023) Evaluation of Whole-Tumor Texture Analysis Based on MRI Diffusion Kurtosis and Biparametric VI-RADS Model for Staging and Grading Bladder Cancer. Bioengineering, 10, Article 745.
https://doi.org/10.3390/bioengineering10070745
[31] Song, J., Yin, Y., Wang, H., Chang, Z., Liu, Z. and Cui, L. (2020) A Review of Original Articles Published in the Emerging Field of Radiomics. European Journal of Radiology, 127, Article ID: 108991.
https://doi.org/10.1016/j.ejrad.2020.108991
[32] Scalco, E. and Rizzo, G. (2017) Texture Analysis of Medical Im-ages for Radiotherapy Applications. The British Journal of Radiology, 90, Article ID: 20160642.
https://doi.org/10.1259/bjr.20160642
[33] Stephan, U., Lucian, B., Annemarie, B., et al. (2020) Radiomics of Computed Tomography and Magnetic Resonance Imaging in Renal Cell Carcinoma—A Systematic Review and Me-ta-Analysis. European Radiology, 30, 3558-3566.
https://doi.org/10.1007/s00330-020-06666-3
[34] Cui, E., Li, Z., Ma, C., et al. (2020) Predicting the ISUP Grade of Clear Cell Renal Cell Carcinoma with Multiparametric MR and Multiphase CT Radiomics. European Radiology, 30, 2912-2921.
https://doi.org/10.1007/s00330-019-06601-1
[35] Yao, Z., Shuai, W., Yan, C. and Du, H.Q. (2021) Deep Learning with a Convolutional Neural Network Model to diFferentiate Renal Parenchymal Tumors: A Preliminary Study. Ab-dominal Radiology, 46, 3260-3268.
https://doi.org/10.1007/s00261-021-02981-5
[36] Xi, I.L., Zhao, Y., Wang, R., et al. (2020) Deep Learning to Dis-tinguish Benign from Malignant Renal Lesions Based on Routine MR Imaging. Clinical Cancer Research, 26, 1944-1952.
https://doi.org/10.1158/1078-0432.CCR-19-0374
[37] Meghan, G.L. (2020) Radiomics and Artificial Intelligence for Renal Mass Characterization. Radiologic Clinics, 58, 995-1008.
https://doi.org/10.1016/j.rcl.2020.06.001
[38] Buch, K., Kuno, H., Qureshi, M.M., Li, B. and Sakai, O. (2018) Quantitative Variations in Texture Analysis Features Dependent on MRI Scanning Parameters: A Phantom Model. Journal of Applied Clinical Medical Physics, 19, 253-264.
https://doi.org/10.1002/acm2.12482