分享:
分享到微信朋友圈
X
综述
基于MRI的人工智能在子宫内膜癌和宫颈癌中的应用进展
王先虹 毕秋 毕国力

WANG X H, BI Q, BI G L. Application progress of MRI-based artificial intelligence in endometrial and cervical cancers[J]. Chin J Magn Reson Imaging, 2023, 14(8): 182-186, 202.引用本文:王先虹, 毕秋, 毕国力. 基于MRI的人工智能在子宫内膜癌和宫颈癌中的应用进展[J]. 磁共振成像, 2023, 14(8): 182-186, 202. DOI:10.12015/issn.1674-8034.2023.08.032.


[摘要] 传统影像学观测在子宫内膜癌(endometrial carcinoma, EC)和宫颈癌(cervical cancer, CC)的诊断、分期及预后评估等方面的效能尚待提高。近年来人工智能(artificial intelligence, AI)在超声、CT、MRI等医学影像领域取得了较大进展,其通过高通量提取数据特征,可观测到肉眼无法识别的病灶内部异质性。目前,AI分析广泛运用于EC和CC的诊疗研究中,但尚缺乏基于MRI的AI分析在EC和CC中应用的系统性评价。本文就AI的定义及医学应用、基于MRI的AI分析在EC和CC中的术前诊断、分期、病理组织学评估和预后预测等方面进行综述,以期进一步实现对EC和CC患者的早期诊断、个体化治疗及精准预后。期待未来基于MRI的AI技术能够深入到病理、分子甚至基因水平,为推动个体化精准医疗提供新思路。
[Abstract] The evaluation efficiency of traditional imaging observation in the diagnosis, staging, and prognosis of endometrial carcinoma (EC) and cervical cancer (CC) remains to be improved. In recent years, artificial intelligence (AI) has made significant advances in medical imaging fields such as ultrasound, CT, MRI, etc. With the advantage of high throughput extraction of data features, AI can observe the internal heterogeneity of lesions that cannot be recognized by the naked eye. At present, AI analysis is widely used in the diagnosis and treatment of EC and CC, but there is still a lack of systematic review of the application of MRI-based AI analysis in EC and CC. In this article, we review the definition and medical applications of AI, as well as preoperative diagnosis, staging, pathological histological assessment, and prognosis prediction of MRI-based AI analysis in EC and CC, in order to further achieve early diagnosis, individualized treatment, and accurate prognosis for patients with EC and CC. It is expected that MRI-based AI technology can penetrate to the pathological, molecular, and even genetic levels in the future, providing new ideas for promoting personalized precision medicine.
[关键词] 子宫内膜癌;宫颈癌;磁共振成像;影像组学;人工智能;诊断;分期;预后
[Keywords] endometrial cancer;cervical cancer;magnetic resonance imaging;radiomics;artificial intelligence;diagnosis;staging;prognosis

王先虹 1, 2   毕秋 2   毕国力 2*  

1 昆明理工大学医学院,昆明 650000

2 昆明理工大学附属医院/云南省第一人民医院磁共振科,昆明 650032

通信作者:毕国力,E-mail:guolibi76@163.com

作者贡献声明:毕国力设计本研究的方案,对文章进行指导,获得了云南省科技厅-昆明医科大学应用基础研究联合专项面上项目基金支持;王先虹起草和撰写稿件,获取、分析或解释文章数据,对稿件内容进行修改;毕秋对稿件内容进行了修改,对文章知识性内容做批评性审阅,获得昆明理工大学医学专项联合面上项目基金支持;全体作者都同意发表最后的修改稿,同意对本研究的所有方面负责,确保本研究的准确性和诚信。


基金项目: 云南省科技厅-昆明医科大学应用基础研究联合专项面上项目 202001AY070001-110 昆明理工大学-云南省第一人民医院医学专项联合面上项目 KUST-KH2022027Y
收稿日期:2023-01-06
接受日期:2023-06-25
中图分类号:R445.2  R737.33 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2023.08.032
引用本文:王先虹, 毕秋, 毕国力. 基于MRI的人工智能在子宫内膜癌和宫颈癌中的应用进展[J]. 磁共振成像, 2023, 14(8): 182-186, 202. DOI:10.12015/issn.1674-8034.2023.08.032.

0 前言

       常见子宫恶性肿瘤包括子宫内膜癌(endometrial carcinoma, EC)和宫颈癌(cervical cancer, CC),属于全球妇女发病率较高的肿瘤[1]。EC是西欧等发达国家最常见的妇科恶性肿瘤,属于我国第二常见的子宫恶性肿瘤,且有年轻化趋势[2]。CC是我国妇女最常见的子宫恶性肿瘤,发病率在全球女性生殖系统恶性肿瘤中排第三,也是年轻妇女癌症致死的主要疾病[3]。不同分期EC和CC治疗方式及手术切除范围各不相同,国际妇产科联合会(International Federation of Obstetrics and Gynecologic, FIGO)指出评估EC和CC术前分期的首选检查方式为MRI[4, 5]。近年来,人工智能(artificial intelligence, AI)已经逐渐应用于临床医学领域,影像组学在EC和CC中的应用经历了从诊断、鉴别诊断、组织病理学分级到肌层浸润、宫旁浸润、淋巴结转移(lymph node metastasis, LNM)、淋巴血管间隙侵犯(lymphovascular space invasion, LVSI)评估再到疗效评估及预后预测的发展过程,基于MRI的影像组学研究也逐渐从单一模态发展为多模态[6]。目前大多数基于MRI的AI分析对EC研究仅侧重于对良性与恶性疾病进行分类,对CC研究侧重于对LNM进行分类[7],但AI可解释性和数据集分组不均衡的问题仍然是最大的挑战之一,且多数训练组样本量较小,以及缺乏外部验组,这些都是目前存在的挑战[7]。本文主要介绍基于MRI的AI分析在EC和CC研究领域的当前应用现状、存在的问题及发展前景,旨在对EC和CC的诊断、分期及预后预测等方面提供文献参考,为推动个体化精准医疗提供新思路。

1 AI的定义及医学应用

       广义的AI是指计算机执行任务的能力,是通过发展理论、方法和应用系统来模拟、延伸和拓展人类智能的一门新兴技术科学,属于计算机科学的一个分支[8]。AI技术可以从大量数据中提取出人类观察容易遗漏的关键信息[9]。机器学习属于AI的一个分支[10]。目前,基于MRI的AI分析技术主要分为影像组学和深度学习(deep Learning, DL)两种形式,二者均属于机器学习的范畴[11]

1.1 影像组学在EC和CC中的应用

       LAMBIN等[12]在2012年首次提出影像组学概念,影像组学可从图像中高通量挖掘定量图像特征,并深入分析影像数据来构建精细化模型,以提高疾病诊疗及预后的准确性,属于AI在医学图像中应用最多的技术,目前已逐步应用到妇科恶性肿瘤诊断、分期、病理分级、风险评估及预后预测中[13]。最近的研究表明[14],基于MRI的影像组学模型术前预测EC分级、深肌层浸润、LVSI和LNM方面具有较高的敏感性和特异性。SONG等[15]从T2加权图像(T2-weighted imaging, T2WI)、扩散加权图像(diffusion weighted imaging, DWI)和对比增强T1加权图像(contrast-enhanced T1-weighted imaging, CE-T1WI)中提取特征构建影像组学模型预测EC中的微卫星不稳定性,结果显示支持向量机(support vector machine, SVM)模型在训练、内部验证和外部验证组的受试者工作特征曲线下面积(area under the curve, AUC)分别为0.921、0.903和0.937。一项荟萃分析[16]对39篇CC预测模型文章进行了系统评价,尽管各项研究采用不同机器学习算法开发预测模型,但所有模型均展现出良好的性能,该研究表明以影像组学为代表的AI对CC预测具有较高的价值,有助于临床治疗决策的制订。

1.2 DL在EC和CC中的应用

       DL是机器学习的一种形式,目前,基于DL的模型在支持疾病检测和诊断的各种医学成像任务中取得了显著成功,在医学图像重建中具有较广阔的应用前景[11,17]。DL模型在辅助EC分析方面表现出良好的性能。HODNELAND等[18]利用卷积神经网络自动提取肿瘤体积和全体积的肿瘤纹理特征,该方法达到了人类专家级表现。MAO等[19]开发了基于U-net网络的语义分割模型对早期EC患者MRI图像自动分割,该模型对ⅠA期和ⅠB期EC具有较好的诊断性能。KALBHOR等[20]结合不同机器学习分类器和模糊最小-最大神经网络的DL模型分别用于CC特征提取和宫颈细胞图像分类,构建的多种DL训练模型对CC图像分类具有良好的准确性,其中RestNet-50模型性能最佳,其分类准确率达到95.33%。XU等[21]开发了一种新型DL模型能够对八种类型的宫颈细胞进行分类,该模型性能表现优于其他模型,其分类准确率高达99.81%。LI等[22]开发了四种用于分类的卷积神经算法构建DL模型以区分CC图像和良性组织图像,其中Xception模型在外部验证组中区分宫颈鳞状细胞癌和良性组织的AUC值达0.974、区分宫颈腺癌和良性组织的AUC值达0.958。以上研究表明,以DL为代表的AI正在迅速崛起,并运用于EC和CC的诊断、分类和预后评估中,未来有巨大的发展潜力。

2 基于MRI的AI分析在EC中的应用

2.1 诊断及鉴别诊断

       EC、子宫内膜增生和息肉均较常见,为了避免治疗不足或过度治疗,保护患者的生育能力,术前必须准确鉴别子宫内膜良恶性病变[23]。虽然现阶段可以通过诊刮、活检和宫腔镜检等检查在术前初步识别子宫内膜良恶性病变,但由于有创性和各种限制因素无法准确诊断局灶性EC[24]。而功能MRI能对原发肿瘤和转移灶进行定量测量,在EC诊断和治疗中发挥重要作用[25]。JIANG等[26]开发和验证基于T2WI和CE-T1WI的影像组学模型评估MRI不可见的EC,训练组和验证组AUC分别为0.873和0.918,表明影像组学模型可以为检测MRI不可见的EC提供良好的性能。ZHANG等[27]从T2WI、DWI和表观扩散系数(apparent diffusion coefficient, ADC)图提取特征并结合临床信息构建了多模态影像组学-临床模型,该模型对术前鉴别并发EC与非典型子宫内膜增生具有较高的效能,其AUC值高达0.932。BI等[28]回顾性多中心收集了术前影像难以鉴别的早期EC和子宫内膜良性病变,构建出临床结合MRI组学的综合模型,且综合模型均具有较高的诊断效能。以上研究初步证实了以影像组学为代表的AI图像分析对于影像难以发现的早期EC具有较高诊断价值,未来有潜力进一步探索影像组学对镜下EC和不典型增生内膜的鉴别诊断价值。

       此外,WANG等[29]构建SVM机器学习模型预测T2高信号间质来源子宫肿瘤恶性程度,临床-影像组合模型的AUC值高达0.91,该研究表明以AI为代表的影像组学方法可有效区分良性平滑肌瘤和恶性间质来源子宫肿瘤。DAI等[30]证明,结合临床参数构建多种机器学习模型可以有效区分子宫肉瘤和非典型子宫平滑肌瘤,并且迁移学习模型性能优于影像组学模型。综上表明,AI广泛运用于子宫肿瘤的诊断与鉴别诊断中,为临床提供实用价值。

2.2 FIGO分期

       根据EC的FIGO分期标准,肌层浸润深度、有无宫颈间质浸润及有无LNM均属于EC手术病理分期的主要内容[23]。EC不同分期的手术切除范围及术后治疗方案不同,术前通过影像学手段评估EC分期有利于临床医师个体化手术方案的制订[2]。在常规MRI的AI研究方面,STANZIONE等[31]开发了一种基于MRI影像组学驱动的机器学习模型对深部肌层侵犯检测具有较高的可行性,使用机器学习时,影像科医师的阅片诊断准确率从82%提高到100%。在功能MRI的AI研究方面,GHOSH等[32]发现基于扩散张量成像(diffusion tensor imaging, DTI)的直方图分析可以精确评估深肌层浸润,其原因可能为DTI后处理参数图可间接反映病灶密度和纤维走行方向,进而将EC从正常子宫组织中区分开来。在多模态MRI的AI研究方面,YAN等[33]从T2WI、DWI、ADC和CE-TIWI图像提取影像特征,并使用随机森林(random forest, RF)分类器建立影像组学模型评估盆腔LNM,结果影像组学模型辅助影像科医师诊断盆腔LNM的效能明显高于单独的影像科医生。方如旗等[34]用多模态MRI影像组学诺模图初步预测EC宫颈间质浸润,发现诺模图预测宫颈间质浸润的效能高于常规MRI模型,训练组和验证组的AUC分别为0.91和0.76,表明影像组学在EC宫颈间质浸润方面有很大的探索价值。

2.3 病理分级及分型

       不同病理类型及分化的EC患者的治疗方式及预后差异很大[2]。有学者[35]用1.5 T MRI扫描ADC图预测EC的病理分级,结果表明ADC直方图与EC分级不存在相关性。然而,另有学者[36]用3.0 T MRI扫描ADC图对EC的病理分级进行评估,结果显示肿瘤分化级别越高其ADC值越低,提示ADC值可能是预测EC病理分级的有用指标,但目前尚缺乏不同场强的ADC直方图分析预测EC病理分级的对比研究,ADC直方图对于EC分级的价值也尚待进一步探索。此外,有研究[37]表明,结合ADC值和T2WI影像组学特征的诺模图对EC病理分级有较高的准确性。ZHENG等[38]发现,对比ADC模型和影像组学模型,基于MRI影像组学诺模图在术前预测EC的病理组织学分型表现出更高的预测能力,该模型在训练和测试队列的AUC分别为0.925和0.915。以上研究结果提示,基于MRI的AI分析方法可以表现EC影像异质性,进而区分病理分型及分化。

2.4 LVSI状态

       EC的LVSI状态与患者复发风险及预后密切相关,术前判断LVSI是EC管理中面临的一个大问题[2]。有学者[39]构建出全肿瘤影像组学模型预测LVSI,该模型在训练组和验证组的AUC分别为0.92和0.81,具有较高的准确性。LONG等[40]基于MRI的传统影像组学和计算机视觉开发出用于预测EC患者的LVSI状态的影像组学诺模图,与影像组学模型和在影像组学模型中加入计算机视觉组学特征的模型相比,诺模图的效能、敏感度和特异度均较高。LIU等[41]回顾性多中心收集了早期伴或不伴LVSI的EC患者,将放射组学特征与年龄和癌抗原125相结合,构建个体化预测诺模图,该诺模图在训练组和验证组的AUC分别为0.89和0.85,该研究表明影像组学诺模图可单独预测早期EC患者的LVSI。以上研究结果表明,基于MRI的影像组学诺模图可以术前预测EC患者的LVSI状态,有助于临床医生进行更好的临床决策。

2.5 术前风险评估及预后预测

       欧洲肿瘤学会临床实践指南规定EC分为4个风险类别:低风险、中风险、中高风险和高风险,不同风险EC患者的治疗方案及预后差异较大[2]。CELLI等[42]通过基于T2WI和ADC纹理特征影像基因组学模型较好地预测低风险EC患者,该模型预测低风险EC患者的AUC为0.74。LEFEBVRE等[43]基于多参数MRI三维影像组学的机器学习模型可以对EC患者进行较好的术前风险分层。MICCÒ等[39]基于T2WI构建出全肿瘤影像组学模型术前预测高危EC,结果该模型在区分低风险与其他风险类别EC展现出较好的效能。JACOB等[44]基于MRI纹理特征结合基因组数据构建出预测EC患者预后的模型,该模型对于预测患者术后五年生存率的AUC为0.72。LIU等[45]建立的基于多参数MRI影像组学诺模图模型预测EC五年无进展生存期表现出良好的性能,在训练和测试队列的AUC分别为0.840和0.958。综上,基于MRI的影像组学有助于鉴别EC患者术前进行不同层次风险评估及术后生存预测,进而辅助临床治疗,实现个体化精准治疗。

3 基于MRI的AI分析在CC中的应用

3.1 诊断及FIGO分期

       根据CC的FIGO分期标准,肿瘤大小、有无宫颈浸润及有无LNM均属于CC手术病理分期的主要内容[46]。当CC病灶较小时,在传统MRI观测上难以与正常宫颈组织鉴别。REN等[47]构建出对早期CC患者间质浸润深度有较高术前诊断性能的影像组学诺模图,且其术前诊断方面优于影像科医生,从而为临床早期诊断间质浸润深度提供依据。ZHAO等[48]从T1WI、T2WI、DWI和动态对比增强序列中提取一阶统计量和纹理构建模型,结果SVM分类器建立的T2WI模型对早期CC的术前分期展现出最优效能,其训练集和测试集的AUC分别为0.915和0.907。WANG等[49]通过构建T2WI的RF模型、扩散峰度成像(diffusion kurtosis imaging, DKI)模型和T2WI加DKI的组合模型来鉴别CC临床病理特征,结果表明组合模型对CC组织学亚型有出色诊断鉴别能力,且DKI模型在区分FIGO分期方面表现出最佳性能,为CC患者分期提供科学依据。此外,有无LNM也是CC FIGO分期的主要内容[5]。最近研究表明[50, 51],基于MRI影像组学的诺模图对预测早期CC盆腔LNM具有良好的性能,可以在术前有效预测早期CC患者LNM。以上研究提示,基于MRI的影像组学对CC术前诊断及分期展现出传统MRI不具有的优势,从而为临床早期诊断和分期提供依据。

3.2 病理分级及分型

       CC最常见的病理类型为鳞状细胞癌,其次是腺癌,腺癌患者对放化疗的敏感性较低,预后较差,病理分级及分型是影响CC患者预后的重要因素[52]。WANG等[53]成功构建出了区分宫颈腺癌和鳞状细胞癌的多参数MRI影像组学模型,且矢状面T2WI组学模型具有较好的鉴别能力,说明在矢状位上提取的特征更能反映腺癌和鳞状细胞癌的异质性。LIU等[54]回顾性多中心收集235名CC患者,构建出基于MRI的影像组学模型对鉴别宫颈腺癌和鳞状细胞癌有较好的效能,且该模型对鉴别CC的分级也具有一定的价值,该研究表明MRI影像组学特征可作为一种无创评价CC临床病理指标的方法。以上研究表明,以影像组学为代表的AI图像分析技术重新定义了影像学检查在疾病诊断中的模式,对CC病理分级及分型的预测具有潜在价值。

3.3 LVSI及分子分型

       深部肌层浸润是LVSI的独立危险因素,而有LVSI的患者具有较高的淋巴结和远处转移的风险,常提示预后不良[55]。CUI等[56]从T2WI和CE-T1WI图像的肿瘤和肿瘤周围提取影像特征,构建出结合影像组学特征和细胞分化程度的影像组学诺模图预测早期CC患者LVSI具有最佳的预测性,该研究提示诺模图可用作预测CC患者LVSI的非侵入性工具。有学者认为基于MRI的影像组学结合DL也可以在术前早期预测CC患者的LVSI情况[57]。XIAO等[58]回顾性研究233名ⅠB~ⅡB期CC患者,构建出基于多参数MRI影像组学诺模图可以独立地预测早期CC患者的LVSI,为临床治疗决策提供了依据。国内学者研究发现,基于MRI的机器学习算法模型能预测CC组织中P53的表达,且部分影像组学特征在P53阳性与阴性患者中存在显著差异[59],说明MRI影像组学在检测肿瘤分子基因表型中有巨大潜力,对CC的临床决策有一定价值。

3.4 疗效评估及预后预测

       最近,LIU等[60]通过T2WI提取特征,构建出RF模型在预测晚期CC放化疗效果评估方面有潜在的应用价值。多项研究[61, 62]将临床结合MRI影像组学特征构建了影像组学诺模图,用于评估局部晚期CC同步放化疗患者的预后,该模型对患者的无进展生存期和总生存期均有良好的预测价值。ZHOU等[63]构建结合组学评分与临床危险因素的组合模型,并与单独的临床模型进行比较,结果发现组合模型在预测无病生存期中的表现明显优于单独的临床模型,为临床决策提供了依据。目前对于CC疗效评估及预后预测尚缺乏公认的生物标志物,影像组学作为一种无创和低成本的方法,可为临床提供参考意见。

4 小结与展望

       目前,基于MRI的AI分析在EC和CC的研究仍处于探索阶段,仍有很多领域需要发掘。越来越多的研究表明,MRI影像组学在EC和CC的智能诊断、评估及预测等方面具有广阔的应用前景,但也存在一定的局限性。首先,人工手动勾画病灶非常耗时,病灶的自动及半自动勾画分割准确性不高,需要未来进一步研究;其次,大多数研究缺乏多中心验证,样本量少且泛化能力弱,而且大多是回顾性研究,存在一定的选择偏倚;最后,影像组学标准不统一,如MRI设备的型号、场强和扫描参数不一导致获取的影像图像存在差异性。另外影像组学研究中的特征提取、建模方法和软件选择等存在差异,导致研究结果可重复性低,针对以上问题,有学者提出了图像生物标记物标准化倡议,即通过标准化影像组学特征得到参考值,从而能够验证和校准不同的影像组学软件,促进了影像组学的临床应用[64]。当下,基于MRI的DL在子宫肿瘤图像分割、诊断、良恶性鉴别方面展现出卓越的性能,但存在训练样本量过少、开发的DL模型过度拟合和欠拟合情况,未来的关键问题是增加训练样本量、多中心验证、前瞻性临床试验的验证[7]。影像组学和DL都属于机器学习领域,是AI的一个分支,未来也有望将影像组学与DL进一步结合,实现双剑合璧的效果。在未来的发展中,有望在医学影像领域中引入联邦学习技术以解决数据孤岛、数据隐私等问题,进而最大程度发挥AI技术在子宫疾病中的诊疗优势。期待未来基于MRI的AI技术能够深入到病理、分子甚至基因水平,为推动个体化精准医疗提供新思路。

[1]
MILLER K D, NOGUEIRA L, DEVASIA T, et al. Cancer treatment and survivorship statistics, 2022[J]. CA Cancer J Clin, 2022, 72(5): 409-436. DOI: 10.3322/caac.21731.
[2]
OAKNIN A, BOSSE T J, CREUTZBERG C L, et al. Endometrial cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up[J]. Ann Oncol, 2022, 33(9): 860-877. DOI: 10.1016/j.annonc.2022.05.009.
[3]
SIEGEL R L, MILLER K D, FUCHS H E, et al. Cancer statistics, 2022[J]. CA A Cancer J Clinicians, 2022, 72(1): 7-33. DOI: 10.3322/caac.21708.
[4]
KASIUS J C, PIJNENBORG J M A, LINDEMANN K, et al. Risk stratification of endometrial cancer patients: FIGO stage, biomarkers and molecular classification[J/OL]. Cancers, 2021, 13(22): 5848 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/34831000/. DOI: 10.3390/cancers13225848.
[5]
WRIGHT J D, MATSUO K, HUANG Y M, et al. Prognostic performance of the 2018 international federation of gynecology and obstetrics cervical cancer staging guidelines[J]. Obstet Gynecol, 2019, 134(1): 49-57. DOI: 10.1097/AOG.0000000000003311.
[6]
MANGANARO L, NICOLINO G M, DOLCIAMI M, et al. Radiomics in cervical and endometrial cancer[J/OL]. Br J Radiol, 2021, 94(1125): 20201314 [2023-05-01]. https://pubmed.ncbi.nlm.nih.gov/34233456/. DOI: 10.1259/bjr.20201314.
[7]
SHRESTHA P, POUDYAL B, YADOLLAHI S, et al. A systematic review on the use of artificial intelligence in gynecologic imaging - Background, state of the art, and future directions[J]. Gynecol Oncol, 2022, 166(3): 596-605. DOI: 10.1016/j.ygyno.2022.07.024.
[8]
ZHOU J, ZENG ZHI YING, LI L. Progress of artificial intelligence in gynecological malignant tumors[J]. Cancer Manag Res, 2020, 12: 12823-12840. DOI: 10.2147/cmar.s279990.
[9]
ZHANG C, GU J H, ZHU Y Y, et al. AI in spotting high-risk characteristics of medical imaging and molecular pathology[J]. Prec Clin Med, 2021, 4(4): 271-286. DOI: 10.1093/pcmedi/pbab026.
[10]
LECOINTRE L, DANA J, LODI M, et al. Artificial intelligence-based radiomics models in endometrial cancer: a systematic review[J]. Eur J Surg Oncol, 2021, 47(11): 2734-2741. DOI: 10.1016/j.ejso.2021.06.023.
[11]
EMMERT-STREIB F, YANG Z, FENG H, et al. An introductory review of deep learning for prediction models with big data[J/OL]. Front Artif Intell, 2020, 3: 4 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/33733124/. DOI: 10.3389/frai.2020.00004.
[12]
LAMBIN P, RIOS-VELAZQUEZ E, LEIJENAAR R, et al. Radiomics: extracting more information from medical images using advanced feature analysis[J]. Eur J Cancer, 2012, 48(4): 441-446. DOI: 10.1016/j.ejca.2011.11.036.
[13]
DAOUD T, SARDANA S, STANIETZKY N, et al. Recent imaging updates and advances in gynecologic malignancies[J/OL]. Cancers, 2022, 14(22): 5528 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/36428624/. DOI: 10.3390/cancers14225528.
[14]
DI DONATO V, KONTOPANTELIS E, CUCCU I, et al. Magnetic resonance imaging-radiomics in endometrial cancer: a systematic review and meta-analysis[J/OL]. Int J Gynecol Cancer, 2023: ijgc-2023-004313 [2023-05-01]. https://pubmed.ncbi.nlm.nih.gov/37094971/. DOI: 10.1136/ijgc-2023-004313.
[15]
SONG X L, LUO H J, REN J L, et al. Multisequence magnetic resonance imaging-based radiomics models for the prediction of microsatellite instability in endometrial cancer[J]. Radiol Med, 2023, 128(2): 242-251. DOI: 10.1007/s11547-023-01590-0.
[16]
JHA A K, MITHUN S, SHERKHANE U B, et al. Systematic review and meta-analysis of prediction models used in cervical cancer[J/OL]. Artif Intell Med, 2023, 139: 102549 [2023-05-01]. https://pubmed.ncbi.nlm.nih.gov/3710051/. DOI: 10.1016/j.artmed.2023.102549.
[17]
YAQUB M, JINCHAO J C, ARSHID K, et al. Deep learning-based image reconstruction for different medical imaging modalities[J/OL]. Comput Math Methods Med, 2022, 2022: 8750648 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/35756423/. DOI: 10.1155/2022/8750648.
[18]
HODNELAND E, DYBVIK J A, WAGNER-LARSEN K S, et al. Automated segmentation of endometrial cancer on MR images using deep learning[J/OL]. Sci Rep, 2021, 11(1): 179 [2023-05-01]. https://pubmed.ncbi.nlm.nih.gov/33420205/. DOI: 10.1038/s41598-020-80068-9.
[19]
MAO W, CHEN C X, GAO H C, et al. A deep learning-based automatic staging method for early endometrial cancer on MRI images[J/OL]. Front Physiol, 2022, 13: 974245 [2023-05-01]. https://pubmed.ncbi.nlm.nih.gov/36111158/. DOI: 10.3389/fphys.2022.974245.
[20]
KALBHOR M, SHINDE S, POPESCU D E, et al. Hybridization of deep learning pre-trained models with machine learning classifiers and fuzzy Min-max neural network for cervical cancer diagnosis[J/OL]. Diagnostics (Basel), 2023, 13(7): 1363 [2023-05-01]. https://pubmed.ncbi.nlm.nih.gov/37046581/. DOI: 10.3390/diagnostics13071363.
[21]
XU L X, CAI F H, FU Y H, et al. Cervical cell classification with deep-learning algorithms[J]. Med Biol Eng Comput, 2023, 61(3): 821-833. DOI: 10.1007/s11517-022-02745-3.
[22]
LI Y X, CHEN F, SHI J J, et al. Convolutional neural networks for classifying cervical cancer types using histological images[J]. J Digit Imaging, 2023, 36(2): 441-449. DOI: 10.1007/s10278-022-00722-8.
[23]
KOH W J, ABU-RUSTUM N R, BEAN S, et al. Uterine neoplasms, version 1.2018, NCCN clinical practice guidelines in oncology[J]. J Natl Compr Canc Netw, 2018, 16(2): 170-199. DOI: 10.6004/jnccn.2018.0006.
[24]
FERRARI F, FORTE S, ARRIGONI G, et al. Impact of endometrial sampling technique and biopsy volume on the diagnostic accuracy of endometrial cancer[J]. Transl Cancer Res, 2020, 9(12): 7697-7705. DOI: 10.21037/tcr-20-2074.
[25]
SERTIC M, KILCOYNE A, CATALANO O A, et al. Quantitative imaging of uterine cancers with diffusion-weighted MRI and 18-fluorodeoxyglucose PET/CT[J]. Abdom Radiol (NY), 2022, 47(9): 3174-3188. DOI: 10.1007/s00261-021-03218-1.
[26]
JIANG X T, SONG J C, ZHANG A N, et al. Preoperative assessment of MRI-invisible early-stage endometrial cancer with MRI-based radiomics analysis[J]. J Magn Reson Imaging, 2023, 58(1): 247-255. DOI: 10.1002/jmri.28492.
[27]
ZHANG J Y, ZHANG Q, WANG T T, et al. Multimodal MRI-based radiomics-clinical model for preoperatively differentiating concurrent endometrial carcinoma from atypical endometrial hyperplasia[J/OL]. Front Oncol, 2022, 12: 887546 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/35692806/. DOI: 10.3389/fonc.2022.887546.
[28]
BI Q, WANG Y X, DENG Y C, et al. Different multiparametric MRI-based radiomics models for differentiating stage IA endometrial cancer from benign endometrial lesions: a multicenter study[J/OL]. Front Oncol, 2022, 12: 939930 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/35992858/. DOI: 10.3389/fonc.2022.939930.
[29]
WANG T T, GONG J, LI Q, et al. A combined radiomics and clinical variables model for prediction of malignancy in T2 hyperintense uterine mesenchymal tumors on MRI[J]. Eur Radiol, 2021, 31(8): 6125-6135. DOI: 10.1007/s00330-020-07678-9.
[30]
DAI M Y, LIU Y, HU Y, et al. Combining multiparametric MRI features-based transfer learning and clinical parameters: application of machine learning for the differentiation of uterine sarcomas from atypical leiomyomas[J]. Eur Radiol, 2022, 32(11): 7988-7997. DOI: 10.1007/s00330-022-08783-7.
[31]
STANZIONE A, CUOCOLO R, DEL GROSSO R, et al. Deep myometrial infiltration of endometrial cancer on MRI: a radiomics-powered machine learning pilot study[J]. Acad Radiol, 2021, 28(5): 737-744. DOI: 10.1016/j.acra.2020.02.028.
[32]
GHOSH A, SINGH T, SINGLA V, et al. DTI histogram parameters correlate with the extent of myoinvasion and tumor type in endometrial carcinoma: a preliminary analysis[J]. Acta Radiol, 2020, 61(5): 675-684. DOI: 10.1177/0284185119875019.
[33]
YAN B C, LI Y, MA F H, et al. Radiologists with MRI-based radiomics aids to predict the pelvic lymph node metastasis in endometrial cancer: a multicenter study[J]. Eur Radiol, 2021, 31(1): 411-422. DOI: 10.1007/s00330-020-07099-8.
[34]
方如旗, 陈霞平, 章国伟, 等. MR影像组学诺模图预测子宫内膜癌宫颈间质浸润的价值[J]. 临床放射学杂志, 2022, 41(6): 1105-1109. DOI: 10.19401/j.cnki.1007-3639.2021.06.08.
FANG R Q, CHEN X P, ZHANG G W, et al. Radiomics nomogram for prediction of cervical stromal invasion in endometrial cancer[J]. J Clin Radiol, 2022, 41(6): 1105-1109. DOI: 10.19401/j.cnki.1007-3639.2021.06.08.
[35]
BONATTI M, PEDRINOLLA B, CYBULSKI A J, et al. Prediction of histological grade of endometrial cancer by means of MRI[J]. Eur J Radiol, 2018, 103: 44-50. DOI: 10.1016/j.ejrad.2018.04.008.
[36]
REYES-PÉREZ J A, VILLASEÑOR-NAVARRO Y, JIMÉNEZ DE LOS SANTOS M E, et al. The apparent diffusion coefficient (ADC) on 3-T MRI differentiates myometrial invasion depth and histological grade in patients with endometrial cancer[J]. Acta Radiol, 2020, 61(9): 1277-1286. DOI: 10.1177/0284185119898658.
[37]
ZHANG K Y, ZHANG Y, FANG X, et al. Nomograms of combining apparent diffusion coefficient value and radiomics for preoperative risk evaluation in endometrial carcinoma[J/OL]. Front Oncol, 2021, 11: 705456 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/34386425/. DOI: 10.3389/fonc.2021.705456.
[38]
ZHENG T, YANG L S, DU J, et al. Combination analysis of a radiomics-based predictive model with clinical indicators for the preoperative assessment of histological grade in endometrial carcinoma[J/OL]. Front Oncol, 2021, 11: 582495 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/34235069/. DOI: 10.3389/fonc.2021.582495.
[39]
MICCÒ M, GUI B, RUSSO L, et al. Preoperative tumor texture analysis on MRI for high-risk disease prediction in endometrial cancer: a hypothesis-generating study[J/OL]. J Pers Med, 2022, 12(11): 1854 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/36579601/. DOI: 10.3390/jpm12111854.
[40]
LONG L, SUN J Q, JIANG L L, et al. MRI-based traditional radiomics and computer-vision nomogram for predicting lymphovascular space invasion in endometrial carcinoma[J]. Diagn Interv Imaging, 2021, 102(7/8): 455-462. DOI: 10.1016/j.diii.2021.02.008.
[41]
LIU X F, YAN B C, LI Y, et al. Radiomics feature as a preoperative predictive of lymphovascular invasion in early-stage endometrial cancer: a multicenter study[J/OL]. Front Oncol, 2022, 12: 966529 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/36059674/. DOI: 10.3389/fonc.2022.966529.
[42]
CELLI V, GUERRERI M, PERNAZZA A, et al. MRI- and histologic-molecular-based radio-genomics nomogram for preoperative assessment of risk classes in endometrial cancer[J/OL]. Cancers, 2022, 14(23): 5881 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/36497362/. DOI: 10.3390/cancers14235881.
[43]
LEFEBVRE T L, UENO Y, DOHAN A, et al. Development and validation of multiparametric MRI-based radiomics models for preoperative risk stratification of endometrial cancer[J]. Radiology, 2022, 305(2: 375-386. DOI: 10.1148/radiol.212873.
[44]
JACOB H, DYBVIK J A, YTRE-HAUGE S, et al. An MRI-based radiomic prognostic index predicts poor outcome and specific genetic alterations in endometrial cancer[J/OL]. J Clin Med, 2021, 10(3: 538 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/33540589/. DOI: 10.3390/jcm10030538.
[45]
LIU D F, YANG L S, DU D, et al. Multi-parameter MR radiomics based model to predict 5-year progression-free survival in endometrial cancer[J/OL]. Front Oncol, 2022, 12: 813069 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/35433486/. DOI: 10.3389/fonc.2022.813069.
[46]
BHATLA N, AOKI D, SHARMA D N, et al. Cancer of the cervix uteri: 2021 update[J]. Int J Gynaecol Obstet, 2021, 155(Suppl 1): 28-44. DOI: 10.1002/ijgo.13865.
[47]
REN J, LI Y, YANG J J, et al. MRI-based radiomics analysis improves preoperative diagnostic performance for the depth of stromal invasion in patients with early stage cervical cancer[J/OL]. Insights Imaging, 2022, 13(1: 17 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/35092505/. DOI: 10.1186/s13244-022-01156-0.
[48]
ZHAO X, WANG X Y, ZHANG B H, et al. Classifying early stages of cervical cancer with MRI-based radiomics[J]. Magn Reson Imaging, 2022, 89: 70-76. DOI: 10.1016/j.mri.2022.03.002.
[49]
WANG M D, PERUCHO J A U, VARDHANABHUTI V, et al. Radiomic features of T2-weighted imaging and diffusion kurtosis imaging in differentiating clinicopathological characteristics of cervical carcinoma[J]. Acad Radiol, 2022, 29(8: 1133-1140. DOI: 10.1016/j.acra.2021.08.018.
[50]
XIA X M, LI D D, DU W, et al. Radiomics based on nomogram predict pelvic lymphnode metastasis in early-stage cervical cancer[J/OL]. Diagnostics (Basel), 2022, 12(10: 2446 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/36292135/. DOI: 10.3390/diagnostics12102446.
[51]
ZHANG Z H, WAN X J, LEI X Y, et al. Intra- and peri-tumoral MRI radiomics features for preoperative lymph node metastasis prediction in early-stage cervical cancer[J/OL]. Insights Imaging, 2023, 14(1: 65 [2023-05-01]. https://pubmed.ncbi.nlm.nih.gov/37060378/. DOI: 10.1186/s13244-023-01405-w.
[52]
RAO Q X, HAN X, WEI Y, et al. Novel prognostic nomograms in cervical cancer based on analysis of 1075 patients[J]. Cancer Med, 2023, 12(5: 6092-6104. DOI: 10.1002/cam4.5335.
[53]
WANG W, JIAO Y N, ZHANG L C, et al. Multiparametric MRI-based radiomics analysis: differentiation of subtypes of cervical cancer in the early stage[J]. Acta Radiol, 2022, 63(6: 847-856. DOI: 10.1177/02841851211014188.
[54]
LIU Y, SONG T, DONG T F, et al. MRI-based radiomics analysis to evaluate the clinicopathological characteristics of cervical carcinoma: a multicenter study[J]. Acta Radiol, 2023, 64(1: 395-403. DOI: 10.1177/02841851211065142.
[55]
ZHANG H, YU R, ZHANG L, et al. Chemotherapy versus chemoradiotherapy for FIGO stages IB1 and IIA1 cervical squamous cancer patients with lymphovascular space invasion: a retrospective study[J/OL]. BMC Cancer, 2022, 22(1: 202 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/35193525/. DOI: 10.1186/s12885-022-09309-6.
[56]
CUI L P, YU T, KAN Y Y, et al. Multi-parametric MRI-based peritumoral radiomics on prediction of lymph-vascular space invasion in early-stage cervical cancer[J]. Diagn Interv Radiol, 2022, 28(4: 312-321. DOI: 10.5152/dir.2022.20657.
[57]
JIANG X R, LI J X, KAN Y Y, et al. MRI based radiomics approach with deep learning for prediction of vessel invasion in early-stage cervical cancer[J]. IEEE/ACM Trans Comput Biol Bioinform, 2021, 18(3: 995-1002. DOI: 10.1109/TCBB.2019.2963867.
[58]
XIAO M L, LI Y, MA F H, et al. Multiparametric MRI radiomics nomogram for predicting lymph-vascular space invasion in early-stage cervical cancer[J/OL]. Br J Radiol, 2022, 95(1134: 20211076 [2022-12-27]. https://pubmed.ncbi.nlm.nih.gov/35312379/. DOI: 10.1259/bjr.20211076.
[59]
韦明珠, 赵振华, 胡红杰, 等. 宫颈癌MRI影像组学参数预测宫颈鳞癌p53的价值[J]. 中国医学影像学杂志, 2019, 27(12: 934-937, 947. DOI: 10.3969/j.issn.1005-5185.2019.12.013.
WEI M Z, ZHAO Z H, HU H J, et al. MRI radiomics parameters in predicting p53 of cervical squamous cell carcinoma[J]. Chin J Med Imaging, 2019, 27(12: 934-937, 947. DOI: 10.3969/j.issn.1005-5185.2019.12.013.
[60]
LIU D F, ZHANG X H, ZHENG T, et al. Optimisation and evaluation of the random forest model in the efficacy prediction of chemoradiotherapy for advanced cervical cancer based on radiomics signature from high-resolution T2 weighted images[J]. Arch Gynecol Obstet, 2021, 303(3: 811-820. DOI: 10.1007/s00404-020-05908-5.
[61]
LALISCIA C, GADDUCCI A, MATTIONI R, et al. MRI-based radiomics: promise for locally advanced cervical cancer treated with a tailored integrated therapeutic approach[J]. Tumori, 2022, 108(4: 376-385. DOI: 10.1177/03008916211014274.
[62]
ZHANG Y, LIU L, ZHANG K Y, et al. Nomograms combining clinical and imaging parameters to predict recurrence and disease-free survival after concurrent chemoradiotherapy in patients with locally advanced cervical cancer[J]. Acad Radiol, 2023, 30(3: 499-508. DOI: 10.1016/j.acra.2022.08.002.
[63]
ZHOU Y, GU H L, ZHANG X L, et al. Multiparametric magnetic resonance imaging-derived radiomics for the prediction of disease-free survival in early-stage squamous cervical cancer[J]. Eur Radiol, 2022, 32(4: 2540-2551. DOI: 10.1007/s00330-021-08326-6.
[64]
ZWANENBURG A, VALLIÈRES M, ABDALAH M A, et al. The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping[J]. Radiology, 2020, 295(2: 328-338. DOI: 10.1148/radiol.2020191145.

上一篇 磁共振成像VI-RADS评分在膀胱癌中的应用及进展
下一篇 MRI影像组学在早期宫颈癌不良病理因素评估的研究进展
  
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