分享:
分享到微信朋友圈
X
综述
MRI影像组学在脑胶质瘤中的研究进展
梁倩 张辉

Cite this article as: LIANG Q, ZHANG H. Research progression of MRI radiomics in glioma[J]. Chin J Magn Reson Imaging, 2024, 15(2): 192-197.本文引用格式梁倩, 张辉. MRI影像组学在脑胶质瘤中的研究进展[J]. 磁共振成像, 2024, 15(2): 192-197. DOI:10.12015/issn.1674-8034.2024.02.031.


[摘要] 脑胶质瘤是中枢神经系统最常见的原发性恶性肿瘤,实现脑胶质瘤的鉴别诊断、术前预测脑胶质瘤的病理分级、基因分型、肿瘤微环境和胶质瘤的预后评估对于实施个体化治疗具有重要临床意义。近年来,影像组学以其无创、精准的特点在脑胶质瘤的诊疗中取得了较大进展。本文就MRI影像组学在脑胶质瘤中的研究进展予以综述,以拓展MRI影像组学在脑胶质瘤精准诊疗中的新思路,从而为脑胶质瘤的诊断和个体化管理提供临床指导意义。
[Abstract] Glioma is the most common primary malignant tumor of the central nervous system. It is of great clinical significance to realize the differential diagnosis of glioma, preoperative prediction of pathological grade, genotyping, tumor microenvironment and prognosis evaluation of glioma for individualized treatment. In recent years, radiomics has made great progress in the diagnosis and treatment of glioma because of its noninvasive and accurate characteristics. This paper reviews the research progress of MRI radiomics in glioma, in order to expand the new ideas of MRI radiomics in the accurate diagnosis and treatment of glioma, so as to provide clinical guidance for the diagnosis and individualized management of glioma.
[关键词] 脑胶质瘤;影像组学;磁共振成像;鉴别诊断;术前分级;基因分型;预后评估;肿瘤微环境
[Keywords] glioma;radiomics;magnetic resonance imaging;preoperative classification;genotyping;prognosis assessment;tumor microenvironment

梁倩 1   张辉 1, 2*  

1 山西医科大学医学影像学院,太原 030001

2 山西医科大学第一医院影像科,太原 030001

通信作者:张辉,E-mail:zhanghui_mr@163.com

作者贡献声明::张辉拟定本综述的写作思路,指导撰写稿件,并对稿件重要的内容进行了修改,获得了国家自然科学基金的资助;梁倩起草和撰写稿件,获取、分析并解释本综述的参考文献;全体作者都同意最后的修改稿发表,都同意对本研究的所有方面负责,确保本综述的准确性和诚信。


基金项目: 国家自然科学基金项目 U21A20386,81971593
收稿日期:2023-11-01
接受日期:2024-02-02
中图分类号:R445.2  R730.264 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2024.02.031
本文引用格式梁倩, 张辉. MRI影像组学在脑胶质瘤中的研究进展[J]. 磁共振成像, 2024, 15(2): 192-197. DOI:10.12015/issn.1674-8034.2024.02.031.

0 引言

       脑胶质瘤起源于神经胶质细胞,是中枢神经系统最常见的原发性恶性肿瘤,约占颅内恶性肿瘤的81%,具有高度异质性、预后差等特点[1, 2]。影像组学以高通量的方式提取和分析大量图像特征,能够无创表征肿瘤潜在的病理生理学特点 [3],在脑胶质瘤的鉴别诊断、术前预测脑胶质瘤的病理分级、基因分型、肿瘤微环境以及胶质瘤的预后评估中应用广泛,对脑胶质瘤患者实施个体化治疗方案具有重要临床意义,并成为近年来的研究热点。本文就MRI影像组学在脑胶质瘤中的研究进展进行综述,以期为脑胶质瘤的诊断和个体化管理提供临床指导意义。

1 MRI影像组学鉴别诊断脑胶质瘤

       MRI是脑胶质瘤术前诊断的主要影像学检查方式,但是其影像表现与其他的肿瘤及非肿瘤性病变有一定程度的重叠。例如高级别胶质瘤(high grade glioma, HGG)与单发脑转移瘤(solitary brain metastases, SBM)、原发性中枢神经系统淋巴瘤(primary central nervous system lymphoma, PCNSL)、瘤样脱髓鞘病变等具有相似的影像征象,低级别胶质瘤(lower-grade gliomas, LGG)需要与脑炎、脑梗死等进行鉴别。诊断不同,患者治疗方式的选择也不同。因此,准确鉴别脑胶质瘤与其他病变对于指导临床治疗具有重要意义。

       多项研究探讨了MRI影像组学在HGG和SBM、PCNSL中的鉴别价值。尹浩霖等[4]通过对T1加权成像(T1 weighted imaging, T1WI)、T2加权成像(T2 weighted imaging, T2WI)和液体衰减反转恢复(fluid attenuated inversion recovery, FLAIR)序列进行高通量纹理分析,建立了鉴别SBM和HGG的影像组学诊断模型,在训练组中的准确率和曲线下面积(area under the curve, AUC)分别为84.5%和0.939,在验证组中结果相似。张少茹等[5]探讨基于T1对比增强序列(contrast-enhanced T1-weighted, T1CE)和T2-FLAIR的多参数MRI影像组学模型鉴别HGG与PCNSL的价值,研究结果表明联合模型的预测性能最佳,优于单一序列模型,在训练集和验证集的AUC分别为0.978和0.983。此外,一些研究建立的基于功能磁共振成像(functional magnetic resonance imaging, fMRI)的影像组学模型也显示出了良好的鉴别诊断效能。BATHLA等[6]比较了多个影像组学模型的诊断性能以区分胶质母细胞瘤(glioblastoma, GBM)和PCNSL,基于表观弥散系数(apparent diffusion coefficient, ADC)、FLAIR和T1CE的组合模型预测效能最佳,AUC为0.977。SARTORETTI等[7]建立了基于酰胺质子转移加权(amide proton transfer weighted, APTw)成像的影像组学模型鉴别GBM和SBM,AUC为0.836,但是这项研究的样本量较少,有必要进一步收集数据验证模型的效能。

       以上研究表明基于常规MRI和fMRI的影像组学模型在HGG的鉴别诊断中显示出良好的预测效能,且基于多序列的影像组学模型表现更好。但是,目前MRI影像组学在LGG的鉴别诊断中应用较少,应开发高性能模型为临床管理提供指导。

2 MRI影像组学预测脑胶质瘤病理分级

       世界卫生组织(World Health Organization, WHO)根据胶质瘤的组织学表现和侵袭程度将其分为1~4级。胶质瘤的分级越高,预后越差。胶质瘤的术前分级对于临床制订治疗方案以及预后评估至关重要,而目前胶质瘤的分级主要依赖于术后病理学检查[8, 9]。因此,需要一项能够在术前无创预测脑胶质瘤病理分级的技术。ZHOU等[10]构建基于T1CE的影像组学模型用于胶质瘤分级,在训练集中AUC和准确率分别为0.95、84.8%,在验证集中AUC和准确率分别为0.952、93.9%。LIN等[11]基于多参数MRI序列建立了包含影像组学特征、定量参数和临床特征的3种影像组学列线图模型,所有模型均表现出较高的准确性,T1CE模型在单序列中性能最高(AUC=0.92),联合模型的性能更高(AUC=0.97)。SU等[12]结合多参数MRI影像组学特征,获得区分Ⅱ级与Ⅲ级、Ⅱ级与Ⅳ级的最佳预测模型,AUC分别为0.858、0.966,对于区分Ⅲ级和Ⅳ级胶质瘤,平均弥散峰度(mean kurtosis, MK)影像组学特征获得的模型诊断效能最高(AUC=0.947)。

       以上研究表明MRI影像组学在术前无创预测脑胶质瘤病理分级中发挥着重要作用,越来越多的研究表明基于多参数MRI影像组学模型在术前预测胶质瘤病理分级中具有一定意义。由于2021年第五版世界卫生组织中枢神经系统肿瘤分类(the fifth edition of the WHO Classification of Tumors of the Central Nervous System, WHO CNS 5)[13]引入了新的基因分型能够改变脑胶质瘤的分级,有必要按照新版的中枢神经系统分类开发术前精准预测脑胶质瘤分级的MRI影像组学模型。

3 MRI影像组学预测胶质瘤基因分型

       2016年WHO中枢神经系统肿瘤分类首次增加了胶质瘤的基因分型,对胶质瘤做出整合组织学表型和基因分型的诊断,以提高胶质瘤诊断的准确性。2021年WHO CNS 5更加突出了分子标志物对于胶质瘤诊断和分级的重要性,这对于患者的个体化治疗和预后评估具有重要价值[13]。异柠檬酸脱氢酶(isocitrate dehydrogenase, IDH)突变、O6-甲基鸟嘌呤DNA甲基转移酶(O6-methylguanine-DNA methyltransferase, MGMT)启动子甲基化、1号染色体短臂及19号染色体长臂(1p/19q)共缺失等是脑胶质瘤的关键分子标志物,在胶质瘤的发生发展中起着重要作用,并与患者的预后密切相关,术前无创预测胶质瘤的基因突变状态对于胶质瘤患者治疗方案的制订具有指导意义。

3.1 MRI影像组学预测IDH基因型

       胶质瘤的预后与IDH突变状态相关,IDH突变型胶质瘤较IDH野生型胶质瘤侵袭性弱、预后好,对替莫唑胺更为敏感。目前,胶质瘤基因分型的诊断依赖于外科手术,肿瘤相关标志物的非侵入性评估有助于确定胶质瘤的生物学特征,并指导治疗[14, 15]。WANG等[16]基于动态磁敏感对比增强(dynamic susceptibility contrast, DSC)和弥散加权成像(diffusion weighted imaging, DWI)建立的支持向量机模型在预测IDH1突变方面具有良好的诊断性能,训练集和验证集的AUC分别为0.939、0.880。YAN等[17]构建了基于T1CE和ADC影像组学特征的图像融合模型在预测IDH时AUC为0.884,准确率为82.4%。PENG等[18]结合T1CE、T2WI、动脉自旋标记(arterial spin labeling, ASL)的纹理特征构建模型用于预测IDH突变,AUC和准确率分别为0.819、76.1%。这表明多参数MRI影像组学特征可以预测胶质瘤的IDH基因型。

3.2 MRI影像组学预测MGMT启动子甲基化

       MGMT是一种DNA修复酶,MGMT启动子甲基化会导致MGMT对胶质瘤细胞的保护作用丧失,从而使肿瘤对于替莫唑胺等烷化剂更为敏感,预后也较好[19]。因此,MGMT启动子甲基化状态是替莫唑胺化疗效果的重要预测因素[20]。先前已有研究表明[21, 22],常规MRI影像组学能够术前预测MGMT甲基化状态。CHEN等[23]的结果表明肿瘤核心感兴趣区的T1CE与ADC模型相结合预测MGMT甲基化状态的性能最佳,平均准确率和AUC分别为91%、0.90,优于其他单个或多个MRI序列,纳入ADC值后基于常规MRI的模型的预测性能有所提升。而这与WEI等[24]的研究结果不一致,WEI等[24]的研究表明在融合影像组学特征中加入水肿特征和ADC值会导致预测性能轻度下降。故ADC值的纳入对影像组学模型预测脑胶质瘤MGMT启动子甲基化的性能的影响还有待评估。

3.3 MRI影像组学预测1p/19q状态

       1p/19q共缺失被认为是LGG的诊断、预后和预测性生物标志物[25],1p/19q共缺失的LGG比1p/19q非共缺失的脑胶质瘤患者的生存预后、治疗疗效更好,因而识别1p/19q共缺失对于LGG患者的临床管理具有重要作用[26]。FAN等[27]从MRI图像中提取影像特征,开发了一个嵌套交叉验证机器学习模型,运用组学分析WHO Ⅱ级胶质瘤的1p/19q状态,其AUC为0.8079,准确率为75.8%。CASALE等[28]从T2WI和T1CE中提取影像组学特征,建立MRI衍生影像组学模型预测1p/19q共缺失状态,在训练集和验证集中均表现出良好的预测性能,表明基于T2WI和T1CE的MRI影像组学模型可为预测LGG的1p/19q状态提供可靠的无创技术。KHA等[26]以七个最优化的影像组学特征构建最终预测模型预测LGG患者的1p/19q缺失状态,在训练集和外部测试集中准确率分别达到了87%和82.8%,并且该模型在预测不同分级胶质瘤的1p/19q缺失状态具有较好效能,WHO Ⅱ级和WHO Ⅲ级的AUC分别为0.876、0.847。以上多项研究表明MRI影像组学在预测胶质瘤1p/19q状态中具有较好的诊断性能。

3.4 MRI影像组学预测其他基因型

       2021年WHO CNS 5[29]指出对于成人型IDH野生型弥漫性星形胶质细胞瘤,若存在端粒酶逆转录酶(telomerase reverse transcriptase, TERT)启动子突变、表皮生长因子受体(epidermal growth factor receptor, EGFR)扩增、7号染色体增加和10号染色体完全丢失(+7/-10)中的一个或多个基因型的改变,直接分类为CNS WHO 4级,并且将细胞周期依赖性激酶抑制基因2A/B(cyclin-dependent kinase inhibitor 2A/B, CDKN2A/B)纯合缺失的IDH突变型星形细胞瘤归为CNS WHO 4级,CDKN2A/B纯合缺失成为诊断高级别星形细胞瘤的标准之一[30]。TERT启动子突变、EGFR扩增、+7/-10和CDKN2A/B纯合缺失能够增加胶质瘤的分级,影响患者的预后,因而对脑胶质瘤的分级诊断和预后评估至关重要。

       LU等[31]的研究表明基于MRI影像组学特征的列线图在预测LGG患者TERT突变状态方面具有良好性能。FANG等[32]建立基于影像组学特征的机器学习算法预测WHO Ⅱ级胶质瘤患者TERT启动子突变,AUC为0.8446。LI等[33]分析了270例LGG患者的T2WI图像,从这些图像中提取并选择了最优影像组学特征,采用逻辑回归模型预测LGG的EGFR表达状态,训练集和验证集的AUC分别为0.90和0.95,研究结果表明基于常规MRI影像组学模型可用于预测EGFR表达状态。CALABRESE等[34]对400名CNS WHO 4级弥漫性星形细胞瘤成年患者的术前T1WI、T2WI、T2-FLAIR、T1CE、DWI、磁敏感加权成像(susceptibility weighted imaging, SWI)和ASL等MRI数据进行了回顾性分析,开发了使用影像组学特征、卷积神经网络特征和两者的组合模型用于预测分子标志物的状态,结果显示影像组学模型预测CDKN2A/B纯合缺失、+7/-10、EGFR扩增和TERT启动子突变的AUC分别为0.76、0.79、0.77、0.87。但是,此研究结果缺乏外部验证。此外,GAO等[35]建立了基于T1CE和T2-FLAIR的影像组学模型确定CDKN2A/B纯合缺失状态,训练集和验证集的AUC分别为0.916和0.886。YANG等[36]也证明了影像组学特征对CDKN2A/B纯合缺失的预测具有较高的准确性。

       综上,MRI影像组学在预测胶质瘤分子分型中取得了较大进展,亦有多项研究 [37, 38, 39]表明MRI影像组学对多种分子分型成功进行了联合预测,这有利于胶质瘤患者治疗方案的调整与制订,并对患者的预后具有指导意义。然而,CDKN2A/B纯合缺失和+7/-10的MRI影像组学预测模型仅得到初步开发,模型性能还有待在以后的研究中进一步验证。

4 MRI影像组学预测生存期

       脑胶质瘤是最常见的颅内恶性肿瘤,尽管经过标准化治疗,HGG的生存率仍然较低,GBM的中位总生存期仅为10~14个月[40],部分LGG患者的生存率也较低,建立术前预测胶质瘤患者生存期的无创方法将对患者的预后及临床管理提供指导意义。已有多项研究[41, 42, 43]表明基于MRI的影像组学预测模型可以无创有效地应用于胶质瘤的生存预测。JIA等[40]基于多参数MRI影像组学特征,构建并验证标准治疗后GBM患者术前生存分层的影像组学列线图,训练集和验证集的AUC分别为0.877和0.919,对GBM患者的个体化生存分层实现了满意的术前预测。CHOI等[44]的研究证明了将影像组学与常规临床和遗传预后模型相结合后,可以提高GBM患者总生存期和无进展生存期的预测价值。XU等[45]建立了GBM患者的影像组学列线图,预测长期生存率和短期生存率的准确率分别为0.878和0.875。

       综上,基于MRI的影像组学在预测脑胶质瘤患者生存期方面显示出良好的预测效能,纳入临床因素和遗传因素后预测效能有一定程度的提高。病理组学在癌症预后评估中的应用是近年来的研究热点,已有研究[46]表明病理组学预测脑胶质瘤患者总生存期和无进展生存期的价值。因此,建立包含影像、基因、病理及临床信息的多组学融合模型将在脑胶质瘤患者的生存分析中具有更广阔的前景。

5 MRI影像组学鉴别复发和假性进展

       GBM的标准治疗方案为外科手术切除后辅以放化疗,胶质瘤复发和假性进展通常发生于放化疗后的三个月或更长时间内。假性进展通常无需特殊治疗,而复发的胶质瘤因其进展迅速、预后差,可能需要重复手术,故二者的治疗策略和预后完全不同。因此,准确区分复发和假性进展对于胶质瘤患者的生存预后是极为重要的[47, 48]。KIM等[49]将DWI、灌注加权成像(perfusion weighted imaging, PWI)纳入多参数影像组学模型用于区分GBM患者放化疗后的复发和假性进展,AUC为0.90,且其诊断性能明显优于任何ADC或脑血容量(cerebral blood volume, CBV)参数和使用常规MRI的单个影像组学模型。PARK等[50]建立了基于ADC序列的影像组学模型用于鉴别GBM复发和假性进展,AUC和准确性分别为0.80、78%,诊断性能高于基于常规MRI的影像组学模型,表明扩散影像组学模型有助于区分GBM的复发和假性进展。WANG等[51]从术后18F-氟脱氧葡萄糖正电子发射断层扫描(positron emission computed tomography, PET)、11C-蛋氨酸PET和磁共振图像中提取纹理特征建立了综合影像组学模型,展示出了良好的诊断性能(AUC=0.988)。

       脑胶质瘤术后复发和假性进展的鉴别仍然是临床上亟待解决的难题。以上研究表明基于常规MRI和fMRI的影像组学模型在区分胶质瘤复发和假性进展中表现出良好的诊断性能,尤其是fMRI表现出了独特的优势。但现有研究的样本量较小,未来在多中心收集数据并运用深度学习方法训练模型将为肿瘤复发和假性进展的鉴别诊断提供更重要的价值。

6 MRI影像组学预测胶质瘤肿瘤微环境

       肿瘤微环境(tumor microenvironment, TME)由肿瘤细胞、免疫细胞、基质细胞及其各种分泌因子组成,在肿瘤的发生和进展过程中有着重要影响[52]。巨噬细胞是胶质瘤中最常见的免疫浸润细胞,包括肿瘤相关巨噬细胞在内的免疫抑制微环境不利于HGG患者的免疫治疗[53]。因此,无创评估TME对于治疗方案的选择至关重要。先前已有研究[54]表明基于常规MRI的影像组学特征能够预测LGG患者的免疫细胞浸润水平。LI等[41]构建的术前T2WI影像组学模型有助于术前评估胶质瘤中巨噬细胞浸润的程度。FAN等[55]的研究揭示了MRI和巨噬细胞浸润的显著相关性,表明构建基于MRI的巨噬细胞浸润术前定量预测模型具有较大的潜力。KIM等[56]对肿瘤浸润免疫细胞的绝对定量证实了高级别胶质瘤免疫微环境的异质性,并且影像组学特征在预测免疫表型中表现出良好性能,其中ADC图尤为重要。胶质瘤干细胞(glioma stem cells, GSC)具有促进肿瘤发生、分化和自我更新的能力,是胶质瘤治疗的新靶点[57]。CD44和CD133是GSC的常见标记物,在肿瘤细胞的增殖、转移和存活中起调节作用,CD44和CD133表达水平的增加与胶质瘤患者的预后不良显著相关[58]。WANG等[58]的研究表明基于T2-FLAIR序列的影像组学特征在术前能够有效预测LGG患者的CD44和CD133表达水平,有助于患者的个体化治疗。

       目前多数研究都是基于常规MRI建立影像组学模型,fMRI与TME的相关性较少有人研究,fMRI能够从多方面反映肿瘤的内在信息,在影像组学中纳入fMRI对于预测胶质瘤TME的价值值得我们关注。

7 局限性与前景展望

       MRI影像组学在脑胶质瘤的鉴别诊断、脑胶质瘤病理分级、基因分型、肿瘤微环境的术前预测以及胶质瘤的预后评估等方面有着深入研究,但是要应用于临床实践中仍存在有一些问题需要解决。首先,大多数MRI影像组学在脑胶质瘤中的研究是单中心、回顾性研究,得出的研究结论有限,因此需要建立多中心数据库,在大规模数据集中训练模型以提高模型性能,与此同时也要完善相关法律法规制度,避免患者隐私数据的泄露。其次,MRI图像质量与设备的参数设置、重建算法及对比剂用量等因素有关,要获得大量标准化的数据需要各个机构协同制订一致的成像协议。此外,影像组学工作流程中的每个步骤(如图像分割、特征提取、建模等)在不同的研究中差异较大,这使得研究结果的比较和可重复性具有挑战性,深度学习技术可以为影像组学工作流程中的每个步骤提供潜在的解决方案[59]。现有的影像组学模型预测脑胶质瘤基因分型、病理分级的研究大多是依据2016年的中枢神经系统分类,之后的研究应立足于2021年WHO CNS 5,为脑胶质瘤的精准诊疗提供更大助力。

8 小结

       MRI影像组学在脑胶质瘤的诊疗中发挥着重要价值。尽管目前仍有许多挑战需要克服,但随着人工智能技术的发展,未来建立庞大的标准化数据集并结合深度学习技术建立和验证影像组学预测模型将在脑胶质瘤的精准诊疗中提供更大助力。

[1]
XU H, ZHANG A, HAN X, et al. ITGB2 as a prognostic indicator and a predictive marker for immunotherapy in gliomas[J]. Cancer Immunol Immunother, 2022, 71(3): 645-660. DOI: 10.1007/s00262-021-03022-2.
[2]
DU Q, LIN Y, ZHANG W, et al. Bioinformatics analysis of LMAN1 expression, clinical characteristics, and its effects on cell proliferation and invasion in glioma[J/OL]. Brain Res, 2022, 1789: 147952 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/35623391/. DOI: 10.1016/j.brainres.2022.147952.
[3]
PARK Y W, KIM S, PARK C J, et al. Adding radiomics to the 2021 WHO updates may improve prognostic prediction for current IDH-wildtype histological lower-grade gliomas with known EGFR amplification and TERT promoter mutation status[J]. Eur Radiol, 2022, 32(12): 8089-8098. DOI: 10.1007/s00330-022-08941-x.
[4]
尹浩霖, 李冬宝, 蒋宇, 等. 高通量纹理分析鉴别脑内单发转移瘤和高级别胶质瘤[J]. 中华肿瘤学杂志, 2018, 40(11): 841-846. DOI: 10.3760/cma.j.issn.0253-3766.2018.11.009.
YIN H L, LI D B, JIANG Y, et al. High-throughput texture analysis in the distinction of single metastatic brain tumors from high-grade gliomas[J]. Chin J Oncol, 2018, 40(11): 841-846. DOI: 10.3760/cma.j.issn.0253-3766.2018.11.009.
[5]
张少茹, 周云舒, 张若弟, 等. 多参数MRI影像组学模型鉴别高级别胶质瘤与原发性中枢神经系统淋巴瘤的价值[J]. 磁共振成像, 2023, 14(10): 54-57, 64. DOI: 10.12015/issn.1674-8034.2023.10.010.
ZHANG S R, ZHOU Y S, ZHANG R D, et al. The value of multi-parametric MRI-based radiomics models indistinguishing primary central nervous system lymphoma from high-grade glioma[J]. Chin J Magn Reson Imaging, 2023, 14(10): 54-57, 64. DOI: 10.12015/issn.1674-8034.2023.10.010.
[6]
BATHLA G, PRIYA S, LIU Y, et al. Radiomics-based differentiation between glioblastoma and primary central nervous system lymphoma: a comparison of diagnostic performance across different MRI sequences and machine learning techniques[J]. Eur Radiol, 2021, 31(11): 8703-8713. DOI: 10.1007/s00330-021-07845-6.
[7]
SARTORETTI E, SARTORETTI T, WYSS M, et al. Amide proton transfer weighted (APTw) imaging based radiomics allows for the differentiation of gliomas from metastases[J/OL]. Sci Rep, 2021, 11(1): 5506 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33750899/. DOI: 10.1038/s41598-021-85168-8.
[8]
GAO M, HUANG S, PAN X, et al. Machine learning-based radiomics predicting tumor grades and expression of multiple pathologic biomarkers in gliomas[J/OL]. Front Oncol, 2020, 10: 1676 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33014836/. DOI: 10.3389/fonc.2020.01676.
[9]
ZHUGE Y, NING H, MATHEN P, et al. Automated glioma grading on conventional MRI images using deep convolutional neural networks[J]. Med Phys, 2020, 47(7): 3044-3053. DOI: 10.1002/mp.14168.
[10]
ZHOU H, XU R, MEI H, et al. Application of Enhanced T1WI of MRI Radiomics in Glioma Grading[J/OL]. Int J Clin Pract, 2022, 2022: 3252574 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/35685548/. DOI: 10.1155/2022/3252574.
[11]
LIN K, CIDAN W, QI Y, et al. Glioma grading prediction using multiparametric magnetic resonance imaging-based radiomics combined with proton magnetic resonance spectroscopy and diffusion tensor imaging[J]. Med Phys, 2022, 49(7): 4419-4429. DOI: 10.1002/mp.15648.
[12]
SU C, CHEN X, LIU C, et al. T2-FLAIR, DWI and DKI radiomics satisfactorily predicts histological grade and Ki-67 proliferation index in gliomas[J]. Am J Transl Res, 2021, 13(8): 9182-9194.
[13]
BJORKBLOM B, WIBOM C, ERIKSSON M, et al. Distinct metabolic hallmarks of WHO classified adult glioma subtypes[J]. Neuro Oncol, 2022, 24(9): 1454-1468. DOI: 10.1093/neuonc/noac042.
[14]
CHOI Y S, BAE S, CHANG J H, et al. Fully automated hybrid approach to predict the IDH mutation status of gliomas via deep learning and radiomics[J]. Neuro Oncol, 2021, 23(2): 304-313. DOI: 10.1093/neuonc/noaa177.
[15]
DECUYPER M, BONTE S, DEBLAERE K, et al. Automated MRI based pipeline for segmentation and prediction of grade, IDH mutation and 1p19q co-deletion in glioma[J/OL]. Comput Med Imaging Graph, 2021, 88: 101831 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33482430/. DOI: 10.1016/j.compmedimag.2020.101831.
[16]
WANG J, HU Y, ZHOU X, et al. A radiomics model based on DCE-MRI and DWI may improve the prediction of estimating IDH1 mutation and angiogenesis in gliomas[J/OL]. Eur J Radiol, 2022, 147: 110141 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34995947/. DOI: 10.1016/j.ejrad.2021.110141.
[17]
YAN J, ZHANG B, ZHANG S, et al. Quantitative MRI-based radiomics for noninvasively predicting molecular subtypes and survival in glioma patients[J/OL]. NPJ Precis Oncol, 2021, 5(1): 72 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34312469/. DOI: 10.1038/s41698-021-00205-z.
[18]
PENG H, HUO J, LI B, et al. Predicting isocitrate dehydrogenase (IDH) mutation status in gliomas using multiparameter MRI radiomics features[J]. J Magn Reson Imaging, 2021, 53(5): 1399-1407. DOI: 10.1002/jmri.27434.
[19]
YOGANANDA C G B, SHAH B R, NALAWADE S S, et al. MRI-based deep-learning method for determining glioma MGMT promoter methylation status[J]. AJNR Am J Neuroradiol, 2021, 42(5): 845-852. DOI: 10.3174/ajnr.A7029.
[20]
KATSIGIANNIS S, GRAU S, KRISCHEK B, et al. MGMT-positive vs MGMT-negative patients with glioblastoma: Identification of prognostic factors and resection threshold[J/OL]. Neurosurgery, 2021, 88(4): E323-E329 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33432978/. DOI: 10.1093/neuros/nyaa562.
[21]
XI Y B, GUO F, XU Z L, et al. Radiomics signature: A potential biomarker for the prediction of MGMT promoter methylation in glioblastoma[J]. J Magn Reson Imaging, 2018, 47(5): 1380-1387. DOI: 10.1002/jmri.25860.
[22]
JIANG C, KONG Z, LIU S, et al. Fusion radiomics features from conventional MRI predict MGMT promoter methylation status in lower grade gliomas[J/OL]. Eur J Radiol, 2019, 121: 108714 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/31704598/. DOI: 10.1016/j.ejrad.2019.108714.
[23]
CHEN S, XU Y, YE M, et al. Predicting MGMT promoter methylation in diffuse gliomas using deep learning with radiomics[J/OL]. J Clin Med, 2022, 11(12): 3445 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/35743511/. DOI: 10.3390/jcm11123445.
[24]
WEI J, YANG G, HAO X, et al. A multi-sequence and habitat-based MRI radiomics signature for preoperative prediction of MGMT promoter methylation in astrocytomas with prognostic implication[J]. Eur Radiol, 2019, 29(2): 877-888. DOI: 10.1007/s00330-018-5575-z.
[25]
KONG Z, JIANG C, ZHANG Y, et al. Thin-slice magnetic resonance imaging-based radiomics signature predicts chromosomal 1p/19q co-deletion status in grade Ⅱ and Ⅲ gliomas[J/OL]. Front Neurol, 2020, 11: 551771 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33192984/. DOI: 10.3389/fneur.2020.551771.
[26]
KHA Q H, LE V H, HUNG T N K, et al. Development and validation of an efficient MRI radiomics signature for improving the predictive performance of 1p/19q co-deletion in lower-grade gliomas[J/OL]. Cancers (Basel), 2021, 13(21): 5398 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34771562/. DOI: 10.3390/cancers13215398.
[27]
FAN Z, SUN Z, FANG S, et al. Preoperative Radiomics Analysis of 1p/19q Status in WHO Grade Ⅱ Gliomas[J/OL]. Front Oncol, 2021, 11: 616740 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34295805/. DOI: 10.3389/fonc.2021.616740.
[28]
CASALE R, LAVROVA E, SANDULEANU S, et al. Development and external validation of a non-invasive molecular status predictor of chromosome 1p/19q co-deletion based on MRI radiomics analysis of low grade glioma patients[J/OL]. Eur J Radiol, 2021, 139: 109678 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33848780/. DOI: 10.1016/j.ejrad.2021.109678.
[29]
LOUIS D N, PERRY A, WESSELING P, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary[J]. Neuro Oncol, 2021, 23(8): 1231-1251. DOI: 10.1093/neuonc/noab106.
[30]
ŚLEDZIŃSKA P, BEBYN M G, FURTAK J, et al. Prognostic and predictive biomarkers in gliomas[J/OL]. Int J Mol Sci, 2021, 22(19): 10373 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34638714/. DOI: 10.3390/ijms221910373.
[31]
LU J, LI X, LI H. A radiomics feature-based nomogram to predict telomerase reverse transcriptase promoter mutation status and the prognosis of lower-grade gliomas[J/OL]. Clin Radiol, 2022, 77(8): e560-e567 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/35595562/. DOI: 10.1016/j.crad.2022.04.005.
[32]
FANG S, FAN Z, SUN Z, et al. Radiomics features predict telomerase reverse transcriptase promoter mutations in world health organization grade ii gliomas via a machine-learning approach[J/OL]. Front Oncol, 2020, 10: 606741 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33643908/. DOI: 10.3389/fonc.2020.606741.
[33]
LI Y, LIU X, XU K, et al. MRI features can predict EGFR expression in lower grade gliomas: A voxel-based radiomic analysis[J]. Eur Radiol, 2018, 28(1): 356-362. DOI: 10.1007/s00330-017-4964-z.
[34]
CALABRESE E, RUDIE J D, RAUSCHECKER A M, et al. Combining radiomics and deep convolutional neural network features from preoperative MRI for predicting clinically relevant genetic biomarkers in glioblastoma[J/OL]. Neurooncol Adv, 2022, 4(1): vdac060 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/35611269/. DOI: 10.1093/noajnl/vdac060.
[35]
GAO J, LIU Z, PAN H, et al. Preoperative discrimination of CDKN2A/B homozygous deletion status in isocitrate dehydrogenase‐mutant astrocytoma: A deep learning‐based radiomics model using MRI[J/OL]. J Magn Reson Imaging, 2023 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/37555723/. DOI: 10.1002/jmri.28945.
[36]
YANG J, LI L, LUO T, et al. Cyclin-dependent kinase inhibitor 2A/B homozygous deletion prediction and survival analysis[J/OL]. Brain Sciences, 2023, 13(4): 548. https://pubmed.ncbi.nlm.nih.gov/37190513/. DOI: 10.3390/brainsci13040548.
[37]
REN Y, ZHANG X, RUI W, et al. Noninvasive prediction of IDH1 mutation and ATRX expression loss in low-grade gliomas using multiparametric MR radiomic features[J]. J Magn Reson Imaging, 2019, 49(3): 808-817. DOI: 10.1002/jmri.26240.
[38]
SUN C, FAN L, WANG W, et al. Radiomics and qualitative features from multiparametric MRI predict molecular subtypes in patients with lower-grade glioma[J/OL]. Front Oncol, 2021, 11: 756828 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/35127472/. DOI: 10.3389/fonc.2021.756828.
[39]
ZHANG S, SUN H, SU X, et al. Automated machine learning to predict the co-occurrence of isocitrate dehydrogenase mutations and O(6)-methylguanine-DNA methyltransferase promoter methylation in patients with gliomas[J]. J Magn Reson Imaging, 2021, 54(1): 197-205. DOI: 10.1002/jmri.27498.
[40]
JIA X, ZHAI Y, SONG D, et al. A multiparametric MRI-based radiomics nomogram for preoperative prediction of survival stratification in glioblastoma patients with standard treatment[J/OL]. Front Oncol, 2022, 12: 758622 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/35251957/. DOI: 10.3389/fonc.2022.758622.
[41]
LI G, LI L, LI Y, et al. An MRI radiomics approach to predict survival and tumour-infiltrating macrophages in gliomas[J]. Brain, 2022, 145(3): 1151-1161. DOI: 10.1093/brain/awab340.
[42]
XU C, PENG Y, ZHU W, et al. An automated approach for predicting glioma grade and survival of LGG patients using CNN and radiomics[J/OL]. Front Oncol, 2022, 12: 969907 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/36033433/. DOI: 10.3389/fonc.2022.969907.
[43]
WANG J, ZHENG X, ZHANG J, et al. An MRI-based radiomics signature as a pretreatment noninvasive predictor of overall survival and chemotherapeutic benefits in lower-grade gliomas[J]. Eur Radiol, 2021, 31(4): 1785-1794. DOI: 10.1007/s00330-020-07581-3.
[44]
CHOI Y, NAM Y, JANG J, et al. Radiomics may increase the prognostic value for survival in glioblastoma patients when combined with conventional clinical and genetic prognostic models[J]. Eur Radiol, 2020, 31(4): 2084-2093. DOI: 10.1007/s00330-020-07335-1.
[45]
XU Y, HE X, LI Y, et al. The nomogram of MRI-based radiomics with complementary visual features by machine learning improves stratification of glioblastoma patients: A multicenter study[J]. J Magn Reson Imaging, 2021, 54(2): 571-583. DOI: 10.1002/jmri.27536.
[46]
LUO C, YANG J, LIU Z, et al. Predicting the recurrence and overall survival of patients with glioma based on histopathological images using deep learning[J/OL]. Front Neurol, 2023, 14: 1100933 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/37064206/. DOI: 10.3389/fneur.2023.1100933.
[47]
QIU J, TAO Z C, DENG K X, et al. Diagnostic accuracy of dynamic contrast-enhanced magnetic resonance imaging for distinguishing pseudoprogression from glioma recurrence: a meta-analysis[J]. Chin Med J (Engl), 2021, 134(21): 2535-2543. DOI: 10.1097/CM9.0000000000001445.
[48]
SHI W, QU C, WANG X, et al. Diffusion kurtosis imaging combined with dynamic susceptibility contrast-enhanced MRI in differentiating high-grade glioma recurrence from pseudoprogression[J/OL]. Eur J Radiol, 2021, 144: 109941 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34735828/. DOI: 10.1016/j.ejrad.2021.109941.
[49]
KIM J Y, PARK J E, JO Y, et al. Incorporating diffusion- and perfusion-weighted MRI into a radiomics model improves diagnostic performance for pseudoprogression in glioblastoma patients[J]. Neuro Oncol, 2019, 21(3): 404-414. DOI: 10.1093/neuonc/noy133.
[50]
PARK Y W, CHOI D, PARK J E, et al. Differentiation of recurrent glioblastoma from radiation necrosis using diffusion radiomics with machine learning model development and external validation[J/OL]. Sci Rep, 2021, 11(1): 2913 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/33536499/. DOI: 10.1038/s41598-021-82467-y.
[51]
WANG K, QIAO Z, ZHAO X, et al. Individualized discrimination of tumor recurrence from radiation necrosis in glioma patients using an integrated radiomics-based model[J]. Eur J Nucl Med Mol Imaging, 2020, 47(6): 1400-1411. DOI: 10.1007/s00259-019-04604-0.
[52]
LIN Z, WEN M, YU E, et al. ANXA1 as a prognostic and immune microenvironmental marker for gliomas based on transcriptomic analysis and experimental validation[J/OL]. Front Cell Dev Biol, 2021, 9: 659080 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34422796/. DOI: 10.3389/fcell.2021.659080.
[53]
CHEN A X, GARTRELL R D, ZHAO J, et al. Single-cell characterization of macrophages in glioblastoma reveals MARCO as a mesenchymal pro-tumor marker[J/OL]. Genome Med, 2021, 13(1): 88 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/34011400/. DOI: 10.1186/s13073-021-00906-x.
[54]
ZHANG X, LIU S, ZHAO X, et al. Magnetic resonance imaging-based radiomic features for extrapolating infiltration levels of immune cells in lower-grade gliomas[J]. Strahlenther Onkol, 2020, 196(10): 913-921. DOI: 10.1007/s00066-020-01584-1.
[55]
FAN X, LI J, HUANG B, et al. Noninvasive radiomics model reveals macrophage infiltration in glioma[J/OL]. Cancer Lett, 2023, 573: 216380 [2023-11-01]. https://pubmed.ncbi.nlm.nih.gov/37660885/. DOI: 10.1016/j.canlet.2023.216380.
[56]
KIM A R, CHOI K S, KIM M S, et al. Absolute quantification of tumor-infiltrating immune cells in high-grade glioma identifies prognostic and radiomics values[J]. Cancer Immunol Immunother, 2021, 70(7): 1995-2008. DOI: 10.1007/s00262-020-02836-w.
[57]
REHFELD M, MATSCHKE J, HAGEL C, et al. Differential expression of stem cell markers in proliferating cells in glioma[J]. J Cancer Res Clin Oncol, 2021, 147(10): 2969-2982. DOI: 10.1007/s00432-021-03704-5.
[58]
WANG Z H, TANG X P, WU J, et al. Radiomics features based on T2-weighted fluid-attenuated inversion recovery MRI predict the expression levels of CD44 and CD133 in lower-grade gliomas[J]. Future Oncol, 2022, 18(7): 807-819. DOI: 10.2217/fon-2021-1173.
[59]
HATT M, LE REST C C, TIXIER F, et al. Radiomics: Data Are Also Images[J]. J Nucl Med, 2019, 60(Supplement 2): 38S-44S. DOI: 10.2967/jnumed.118.220582.

上一篇 脊髓损伤后神经性疼痛的神经影像研究进展
下一篇 磁共振成像人工智能技术在垂体神经内分泌肿瘤治疗中的研究进展
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2