分享:
分享到微信朋友圈
X
临床研究
基于T2WI影像组学鉴别穿透性胎盘植入的价值研究
冯刘娟 张灵洁 程美英 张小安 李思柯 鲁钰 刘世鹏 杨金泽 赵鑫

Cite this article as: FENG L J, ZHANG L J, CHENG M Y, et al. The value of T2WI imaging-based histology in the ability to identify penetrating placenta implantation[J]. Chin J Magn Reson Imaging, 2025, 16(3): 83-89.本文引用格式:冯刘娟, 张灵洁, 程美英, 等. 基于T2WI影像组学鉴别穿透性胎盘植入的价值研究[J]. 磁共振成像, 2025, 16(3): 83-89. DOI:10.12015/issn.1674-8034.2025.03.013.


[摘要] 目的 探讨基于MRI的影像组学模型对穿透性胎盘植入(placenta percreta, PP)的鉴别能力。材料与方法 回顾性分析2021年1月至2023年12月在郑州大学第三附属医院放射科行胎盘MRI平扫且MRI提示为PP的80例孕产妇的临床及影像资料,其中以术中所见为标准诊断为PP 48例,非PP 32例。所有患者按7∶3的比例随机划分为训练集与测试集。在轴位、冠状位及矢状位的T2WI序列上手动勾画感兴趣区,并提取影像组学特征。对提取的影像组学特征首先进行Z-score正则化操作,再通过t检验进行特征筛选,随后计算Pearson相关系数,最后采用最小绝对收缩和选择算子算法对组学的特征进行筛选及降维,并计算影像组学分数。从7种不同的机器学习算法中选取最优算法进行影像组学模型的构建。对临床信息和影像组学评分分别做单因素逻辑回归分析,将差异有统计学意义的因素纳入到多因素分析,得到独立危险因素(临床信息同时用于构建临床模型),并将其可视化(列线图)建立联合预测模型(影像组学-临床模型)。绘制受试者工作特征曲线,并通过曲线下面积(area under the curve, AUC)、敏感度、特异度和准确率等指标比较模型的效能,运用校准曲线评价模型的校准程度,决策曲线分析评估模型的临床实用价值。结果 筛选出的独立危险因素为孕次和影像组学评分,其优势比分别为0.272 [95%置信区间(confidence interval, CI):0.151~0.492]和1 934.105(95% CI:118.985~31 445.149)。影像组学模型与临床模型在训练集中的AUC值分别为0.948(95% CI:0.884~1.000)和0.723(95% CI:0.596~0.850),在测试集中的AUC值分别为0.828(95% CI:0.601~1.000)和0.676(95% CI:0.474~0.878)。在训练集中影像组学-临床模型的AUC值为0.962(95% CI:0.906~1.000)。DeLong检验结果表明在训练集中临床模型与影像组学模型间及临床模型与影像组学-临床模型间差异均具有统计学意义(P<0.05),但影像组学模型与影像组学-临床模型间差异无统计学意义(P>0.05)。影像组学模型与影像组学-临床模型均具有较好的校准度及临床应用价值。结论 影像组学-临床模型具有较好的诊断效能,可作为对PP鉴别的方式,有助于临床医师对妊娠终止时机和方式的制订提供可靠的依据。
[Abstract] Objective To explore the ability of a MRI based imaging histologic model to identify placenta percreta (PP).Materials and Methods A retrospective study was conducted to collecting data from 80 cases of pregnant women who underwent placental MRI scanning and MRI indications pointing to PP in the Department of Radiology of the Third Affiliated Hospital of Zhengzhou University from January 2021 to December 2023, with surgical findings as the standard, including 48 cases of PP and 32 cases of non-PP. The region of interest was manually outlined on the axial, coronal and sagittal T2WI sequences, and the features of imaging histology were extracted. All patients were randomly divided into training and test sets in the ratio of 7∶3. The extracted imaging histology features were firstly subjected to Z-score regularization, then feature screening by t test, followed by calculation of Pearson correlation coefficients, and finally the least absolute shrinkage and selection operator algorithm was used. Selection operator algorithm for screening and dimensionality reduction of the features of the histology, and calculate the radiomics score. The optimal algorithm was selected from 7 different machine learning algorithms and used to construct an radiomics model. Univariate logistic regression analysis was performed on both clinical data and radiomics scores, revealing statistically significant differences. Subsequently, factors demonstrating significant differences were incorporated into multivariate analysis to identify independent risk factors (clinical information was used to construct clinical models). These factors were then visualized to construct a predictive combined model (nomogram). The receiver operating characteristic curve was plotted, and the efficacy of the model was compared by the indicators of area under the curve (AUC), sensitivity, specificity, and accuracy, and the calibration curve was used to evaluate the calibration degree of the model, and the decision curve analysis was used to assess the effectiveness of the model. The calibration curve was used to evaluate the calibration degree of the model, and the decision curve analysis was used to assess the clinical utility value of the model.Results The multivariate analysis identified two independent risk factors: parity and radiomics score. Parity demonstrated a protective effect with an odds ratio of 0.272 [95% confidence interval (CI): 0.151 to 0.492], while the radiomics score showed a strong positive association with an exceptionally high odds ratio of 1 934.105 (95% CI: 118.985 to 31 445.149). The AUC values for the imaging histology model and the clinical model in the training set were 0.948 (95% CI: 0.884 to 1.000) and 0.723 (95% CI: 0.596 to 0.850), respectively, and in the test set were 0.828 (95% CI: 0.601 to 1.000) and 0.676 (95% CI: 0.474 to 0.878). The AUC value of the imaging histology-clinical model in the training set was 0.962 (95% CI: 0.906 to 1.000). The results of DeLong test showed that there were significant differences in the training set, both between the clinical model and the imaging histology model as well as between the clinical model and the imaging histology-clinical model (P < 0.05), but the differences between the imaging histology model and the imaging histology-clinical model were not statistically significant (P > 0.05). Both the radiomics model and the radiomics-clinical model had good calibration and clinical application value in the test set.Conclusions Imaging histology-clinical modeling has better diagnostic efficacy and can be used as a modality for the identification of PP. It provides a reliable foundation for clinicians in determining the timing and method of pregnancy termination, thereby aiding in the formulation of informed clinical decisions.
[关键词] 胎盘植入性疾病;穿透性胎盘植入;影像组学;磁共振成像;鉴别
[Keywords] placenta accreta spectrum disorders;placenta percreta;radiomics;magnetic resonance imaging;differentiate

冯刘娟 1   张灵洁 2   程美英 1   张小安 1   李思柯 1   鲁钰 1   刘世鹏 1   杨金泽 1   赵鑫 1*  

1 郑州大学第三附属医院影像科,郑州 450052

2 郑州大学第三附属医院超声医学科,郑州 450052

通信作者:赵鑫,E-mail: zdsfyzx@zzu.edu.cn

作者贡献声明:赵鑫设计本研究的具体方案,并对稿件重要内容进行了修改;冯刘娟收集、分析、解释本研究数据并起草和撰写稿件;张灵洁、程美英、张小安、李思柯、鲁钰、刘世鹏、杨金泽在实验数据采集和统计分析等方面做了贡献,并对稿件的重要内容进行了修改;张小安和赵鑫分别获得了2020年河南省重大公益专项和郑州市科技局协同创新重大专项的资金支持;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: 2020年河南省重大公益专项 201300310800 郑州市科技局协同创新重大专项 18XTZX12009
收稿日期:2025-01-13
接受日期:2025-03-10
中图分类号:R445.2  R714.2 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2025.03.013
本文引用格式:冯刘娟, 张灵洁, 程美英, 等. 基于T2WI影像组学鉴别穿透性胎盘植入的价值研究[J]. 磁共振成像, 2025, 16(3): 83-89. DOI:10.12015/issn.1674-8034.2025.03.013.

0 引言

       目前普遍认为胎盘植入性疾病(placenta accreta spectrum, PAS)的发生是因胎盘绒毛的滋养层细胞发生过度增殖、子宫底蜕膜的发育不良及其结构缺陷所致,当具有高度侵袭能力的胎盘绒毛通过缺陷的底蜕膜侵入子宫肌层甚至穿透子宫浆膜层时,常会增加胎盘剥离的困难程度,影响子宫收缩,延长生产时长[1, 2],是妊娠中晚期发生子宫破裂、产后出血,甚至急诊子宫切除的重要原因之一[3, 4]。近年来,随着我国三孩政策开放后晚婚晚育人群的增加,以及人工流产率和剖宫产率的不断增长,PAS的发生率也逐年升高[5, 6, 7]。根据胎盘侵入子宫肌层的深度不同可将PAS分为三种不同的类型,即粘连性胎盘植入、植入性胎盘植入和穿透性胎盘植入(placenta percreta, PP),当胎盘组织侵入子宫肌层达浆膜层甚至累及周围脏器时称为PP[3, 8]。有研究显示,随着胎盘植入的程度不断加深,其可能引起如产后出血、子宫切除、急性肾功能损伤、弥散性血管内凝血等不良妊娠结局的风险也相对升高[9, 10]。因此,产前对PAS的程度进行准确分型尤为重要,特别是对PP类型的识别,是临床医生决定分娩时机及分娩方式的关键。

       在PAS的诊断中没有特异性的检验指标和临床症状。超声因其无创、便携、可重复、对血流敏感等优点仍是目前诊断PAS的首选方法,但容易受到孕妇体型及羊水量等的影响从而干扰诊断的准确性[11, 12, 13]。MRI检查对软组织的分辨率较高,特别是对后壁胎盘及宫旁浸润等方面,但其诊断目前尚未有统一的标准[14, 15],仍需放射科医师凭借经验肉眼进行诊断,无法避免主观意识影响。影像组学是一种新兴的研究领域,通过对影像图像特征的高通量提取及挖掘目前已广泛应用于疾病的诊断和其预后分析[16, 17]。影像组学通过客观的特征提取或可弥补MRI检查在PAS诊断中的不足。国内外对有无PAS发生的影像组学研究较多,但针对PP的研究鲜有报道,有研究表明,PP是发生产后出血的重要原因之一,产前对其正确识别尤为重要[18]。因此本研究旨在通过影像组学技术建立预测模型从而提高对产前PP的鉴别诊断能力。

1 材料与方法

1.1 一般资料

       本文回顾性分析2021年1月至2023年12月在郑州大学第三附属医院(河南省妇幼保健院)放射科行胎盘MRI平扫且MRI提示为PP的80例孕产妇的临床及影像资料。纳入标准:(1)在本院行剖宫产分娩,有完整的临床病史;(2)胎龄≥24周;(3)MRI检查均在产前1个月内完成。排除标准:(1)多胎妊娠;(2)影像图像不完整或有伪影影响观察;(3)孕妇存在其他子宫病变(如子宫肌瘤、恶性肿瘤等)。本研究遵循《赫尔辛基宣言》,经本院伦理委员会批准,免除患者知情同意,批准文号:2024-108-01。

1.2 仪器与方法

       本研究采用美国GE 1.5 T MRI扫描仪和自带的相控阵体部线圈完成MRI扫描。检查前叮嘱患者少量憋尿,采取仰卧位,双膝部适度垫高,先对盆腔部位精准定位后展开初次扫描,随后依次完成轴位、矢状位以及冠状位的扫描。首先采用单次激发快速自旋回波(single-shot fast spin echo, SSFSE)序列采集T2WI图像,参数:矩阵288×288,视野40.0 mm×40.0 mm,层厚5.0 mm,层间隔1.0 mm,回波时间100 ms,重复时间2000 ms;继而进行稳态进动平衡(fast-imaging employing steady-state acquisition, FIESTA)序列扫描,参数:矩阵224×256,视野40.0 mm×40.0 mm,层厚5.0 mm,层间隔1.0 mm,回波时间Min Full,重复时间4.3 ms。

1.3 诊断标准

       以孕产妇剖宫产时术中的情况进行确诊[19]。术中对PP的诊断为胎盘不能自动剥离,往往需手动剥离或钳夹,常发生不可控制的大出血,于子宫浆膜层可见密集且显著扩张的血管,或于周围脏器见胎盘组织。

1.4 影像组学分析

1.4.1 图像分割和特征提取

       将轴位、冠状位及矢状位T2WI图像导入ITK-SNAP 4.0软件(www.itksnap.org),对胎盘边缘感兴趣区(region of interest, ROI)的勾画先由一名具有3年盆腔MRI影像诊断的主治医师(医师1)进行,再由另一名具有7年盆腔MRI诊断经验的副主任医师(医师2)进行检查、修改及保存,如图1。将原始图像及ROI导入Python 3.6(http://www.python.org)软件中,使用Pyradiomics库提供的Scikit-learn模块对三个不同切面的特征进行提取,共提取3591个影像特征(轴位、冠状位、矢状位各1197个)。由第三名具有12年工作经验的主任医师(医师3)在训练集中随机挑选20例单独完成对胎盘ROI的勾画,同样对特征进行提取,用组内相关系数(intra-class correlation coefficient, ICC)对医师2和医师3的勾画结果进行一致性评价,ICC>0.75认为一致性较好。

图1  女,32岁,术后诊断为完全性前置胎盘,穿透性胎盘植入。冠状位(1A)、矢状位(1B)、轴位(1C)感兴趣区勾画示意图。
Fig. 1  Female, 32 years old, postoperative diagnosis of complete placenta praevia with placenta percreta. Schematic diagram of the region of interest delineation on coronal (1A), sagittal (1B), and axial (1C) images.

1.4.2 特征筛选及模型构建

       对提取的影像特征首先实施Z-score正则化操作,旨在抵消因不同特征数值波动而引发的干扰,接着运用t检验来筛选特征,仅留存满足P<0.05条件的部分,继而计算Pearson相关系数,对|r|>0.9的特征予以剔除,最后使用最小绝对收缩和选择算子(least absolute shrinkage and selection operator, LASSO)对特征展开进一步的筛选及降维处理,通过十折交叉验证选取最优λ值,并依据λ值的不同将回归系数向0趋近,选定非0系数特征,同时对保留下来的特征赋予权重并完成特征权重的计算。

       在Python软件的Scikit-learn库中采用7种不同的机器学习算法进行模型的构建,并采用受试者工作特征(receiver operating characteristic, ROC)曲线下面积(area under the curve, AUC)、敏感度、特异度、准确度、阳性预估值(positive predictive value, PPV)、阴性预估值(negative predictive value, NPV)等指标评价其学习能力,从而选取最优的算法进行影像组学模型的构建。

       对收集到的临床信息进行单因素的逻辑回归分析,筛选出对PP具有预测价值的因素,随后将单因素分析中具有统计学差异(P<0.05)的因素纳入到多因素的逻辑回归分析,筛选出对PP具有独立预测的因素,从而构建临床模型。

       将临床信息与影像组学评分(radiomics score, Rad score)纳入到单多因素逻辑回归分析中,筛选出对PP具有独立预测的因素,并将其可视化(列线图)构建联合模型(影像组学-临床模型)。

1.5 统计学分析

       运用Python软件(版本3.7.3,http://www.python.org)完成统计学分析。对于服从正态分布的定量资料以(x¯±s)表示,不符合正态分布的定量资料以MQ1,Q3)表示,分别采用独立样本t检验和Mann-Whitney U检验对组间差异进行对比分析,对于计数数据用n(%)表示,采用χ2检验或Fisher确切概率法进行组间差异比较。本研究通过单因素逻辑回归分析筛选对PP有预测价值的因素,随后将单因素分析中差异有统计学意义的因素纳入多因素逻辑回归分析,筛选独立预测因素。采用逐步向前的变量选择方法构建逻辑回归模型。采用ROC曲线对构建的各模型其诊断效能进行评估,并计算AUC、敏感度、特异度、准确度、PPV、NPV等指标。采用DeLong检验比较AUC的差异。校准曲线评价模型的校准程度,决策曲线分析(decision curve analysis, DCA)评估模型的临床实用价值。P<0.05为差异具有统计学意义。

2 结果

2.1 患者的临床资料

       根据纳排标准最终共纳入80例患者。经手术证实为PP的有48例,非PP的有32例。将所有患者按照7∶3的比例随机划分为训练集(56例)与测试集(24例),其中训练集中诊断为PP的有32例,测试集中诊断为PP的有16例,见表1。在训练集中PP与非PP组间产后出血量、此前产次和此前剖宫产次的差异具有统计学意义(P<0.05)。

表1  患者的一般临床资料
Tab. 1  General clinical data of patients

2.2 影像组学特征的筛选

       T2WI图像ROI的3591个影像组学特征经过t检验后剩余388个特征,选取Pearson相关系数|r|<0.9的特征后剩余121个,最后经LASSO算法交叉验证降维后剩余14个特征进行影像组学模型的构建。筛选及收缩系数的收缩路径曲线见图2

图2  采用LASSO交叉验证选取的最佳参数及模型构建的影像组学特征。2A:回归系数图;2B:均方误差图;2C:最佳特征系数。
Fig. 2  LASSO was used to cross-verify the optimal parameters and the image omics features of the model construction. 2A: Plot of regression coefficients; 2B: Plot of mean square error; 2C: Plot of best characteristic and coefficients.

2.3 影像组学模型的构建及性能评价

       对14种影像组学特征通过7种不同的机器学习方式进行模型构建,其性能比较详见表2。舍弃掉过拟合及有过拟合趋势的学习方式后可以得知逻辑回归模型的效能最好,在训练组中其AUC值、敏感度、特异度、准确度、PPV、NPV分别为0.948 [95%置信区间(confidence interval, CI):0.884~1.000]、90.6%、91.7%、91.1%、93.5%、88.0%。

表2  不同分类器对影像组学模型构建的性能比较
Tab. 2  Comparison of the performance of different classifiers for image omics model construction

2.4 影像组学-临床模型的构建与验证

       将收集到的临床信息和Rad score进行单因素及多因素的逻辑回归分析,在进行单因素分析时孕次、此前产次、此前剖宫产次和Rad score差异均具有统计学意义(P<0.05)。多因素分析时孕次的优势比(odds ratio, OR)值为0.272(95% CI:0.151~0.492),P<0.001。Rad score的OR值为1 934.105(95% CI:118.985~31 445.149),P<0.001,详细见表3。将筛选出来的对PP的独立预测因素可视化构建列线图(图3),即构建影像组学和临床联合模型(影像组学-临床模型),联合模型对PP型胎盘植入具有较好的鉴别能力,在训练集中的AUC、敏感度、特异度、准确度分别为0.962(95% CI:0.906~1.000)、90.6%、91.7%、91.1%,详见表4图4

       DeLong检验发现,无论是在训练集还是测试集中,临床模型与影像组学-临床模型之间的差异均具有统计学意义(P<0.05),详见表5。通过校准曲线及DCA比较可以得知影像组学-临床模型具有较好的校准度及临床应用价值(图5)。

图3  影像组学和临床因素联合构建的列线图。Rad score为影像组学评分;Gravidity为孕次。
Fig. 3  Nomogram constructed by combining radiomics and clinical factors. Rad score refers to the radiomics score; Gravidity refers to the number of pregnancies.
图4  临床(Clinic)模型、影像组学(Radiomics)模型、影像组学-临床(Radiomics-Clinic)模型的受试者工作特征(ROC)曲线。4A:训练集,4B:测试集。AUC:曲线下面积,CI:置信区间。
Fig. 4  Receiver operating characteristic (ROC) curves of the clinical model, radiomics model, and radiomics-clinical model. 4A: Training set; 4B: Test set. AUC: area under the curve; CI: confidence interval.
图5  临床(Clinic)模型、影像组学(Radiomics)模型、影像组学-临床(Radiomics-Clinic)模型的校准曲线(5A、5B)和决策曲线分析(5C、5D)。5A和5C为训练集,5B和5D为测试集。
Fig. 5  Calibration curves (5A, 5B) and decision curve analysis (5C, 5D) of the clinical model, radiomics model, and radiomics-clinical model. 5A and 5C represent the training set, 5B and 5D represent the test set.
表3  临床信息和影像组学评分的单因素及多因素分析
Tab. 3  Univariate and multivariate analyses of clinical information and radiomics score
表4  临床模型、影像组学模型和影像组学-临床模型之间的效能比较
Tab. 4  Comparison of the efficacy among the clinical model, radiomics model, and radiomics-clinical model
表5  临床模型、影像组学模型和影像组学-临床模型AUC的DeLong检验结果(P值)
Tab. 5  DeLong test results of the AUC of the clinical model, radiomics model, and radiomics-clinical modelmodel (P value)

3 讨论

       本研究通过高通量的影像组学特征提取,构建不同的预测模型,寻找对PP更好的鉴别诊断方式,这是目前国内少有的仅针对PP的研究。结果显示,孕次和Rad score是鉴别PP的独立预测因素,构建的影像组学-临床模型对产前正确识别PP具有更好的诊断效能,有助于辅助临床医师充分完善术前准确,从而改善母婴结局[20, 21]

3.1 临床特征在鉴别PP中的价值分析

       近些年来,PAS的发病率不断提高,但是对PAS的产前诊断仍缺乏较为全面的方法。白秀丽等[22]等的研究表明年龄、孕次、人工流产史、孕早期保胎史、剖宫产手术史、辅助生殖、前置胎盘是影响PAS的独立危险因素。JAUNIAUX等[23]的研究同样表明PAS的发生原因与高孕产发病率相关,主要为子宫手术史,尤其是剖宫产。本研究发现仅孕次是导致PP发生的临床独立危险因素,但在训练集的临床资料组间比较中发现PP组和非PP组的此前产次和此前剖宫产次的组间比较差异均是有统计学意义的,是因为PAS的发生与子宫内膜-肌层界面的损伤密切相关,而高孕、高产及剖宫产等是不可忽视的因素,且随着次数的增多,肌层受损的程度相对越重,PP的发生风险也越高。

3.2 影像组学在PP中的应用价值分析

       基于MRI对PAS的诊断主要凭借放射科医师的经验对其宏观特征进行分析,因此无法避免主观因素产生的影响。熊星等[24]基于临床特征及常规胎盘MRI征象经逻辑回归构建的列线图可以作为PAS产前预测的辅助方法,但是特征的评估仍需肉眼识别,缺乏客观性。而影像组学则是通过图像采集、ROI勾画、特征提取、特征筛选从而构建模型[25, 26, 27],可以提取更多肉眼无法识别的特征,减少一定的主观性影响,现已应用于多种疾病的诊断,如肿瘤、心血管等[28, 29, 30]。SUN等[31]通过纹理分析及机器学习的方式对PAS的产前预测进行研究,结果显示该方式可以更准确地诊断是否发生PAS。ROMEO等[32]也采用同样的方式对前置胎盘患者是否发生PAS进行了研究。而REN等[33]则在该基础上不仅对矢状位进行了分析,冠状位及轴位也进行了研究,结果表明多平面的组合可以使MRI的纹理特征得到更充分的体现。因此,本研究基于当前影像组学在PAS中的应用采用了多平面结合的方式对PAS分型中的PP型进行鉴别诊断。

3.3 影像组学在PP中的创新性分析

       现有的影像组学研究大多集中于对产前PAS的预测,而对PAS中某种类型的研究较少[34]。邹锦莉等[10]的研究表明影像组学-临床模型可作为产前预测是否存在PAS的方法,且Rad score对PAS亚型具有较好的鉴别能力,尤其是对PP。因此本研究针对PP进行多平面的特征提取和筛选,通过7种不同的学习方式针对产前PP进行识别,舍弃掉过拟合及表现出过拟合趋势的模型,虽然SVM在训练集中的诊断效能更好,但在测试集中逻辑回归模型相对更稳定。在对临床及影像组学模型的构建中分类器的稳定性是极其重要的[35]。因此,经综合分析确定逻辑回归模型更好。本研究结果显示基于逻辑回归模型构建的影像组学-临床模型能够对产前PP进行更准确地识别。

3.4 本研究的局限性

       (1)为回顾性研究,不可避免存在偏倚,后续有必要对该模型继续开展前瞻性多中心验证工作,从而提升诊断效能;(2)对ROI的勾画为手动完成,不可避免地存在一定的主观影响,应尽可能开发全自动ROI勾画工具;(3)样本量较小,需扩大样本量进一步对模型的构建进行完善。

4 结论

       综上所述,本研究基于MRI多平面T2WI的影像组学特征提取和筛选结合临床数据所建立的预测模型能够对PP进行更好地识别,有助于制订更完备的手术方案,从而降低产后不良结局的发生,提高母婴产后生活质量。

[1]
BROWN L A, MENENDEZ-BOBSEINE M. Placenta accreta spectrum[J]. J Midwifery Womens Health, 2021, 66(2): 265-269. DOI: 10.1111/jmwh.13182.
[2]
HORGAN R, ABUHAMAD A. Placenta accreta spectrum: prenatal diagnosis and management[J]. Obstet Gynecol Clin North Am, 2022, 49(3): 423-438. DOI: 10.1016/j.ogc.2022.02.004.
[3]
JAUNIAUX E, AYRES-DE-CAMPOS D, LANGHOFF-ROOS J, et al. FIGO classification for the clinical diagnosis of placenta accreta spectrum disorders[J]. Int J Gynaecol Obstet, 2019, 146(1): 20-24. DOI: 10.1002/ijgo.12761.
[4]
MA Y D, HU Y Y, MA J M. Animal models of the placenta accreta spectrum: current status and further perspectives[J/OL]. Front Endocrinol, 2023, 14: 1118168 [2025-01-08]. https://pubmed.ncbi.nlm.nih.gov/37223034/. DOI: 10.3389/fendo.2023.1118168.
[5]
EINERSON B D, GILNER J B, ZUCKERWISE L C. Placenta accreta spectrum[J]. Obstet Gynecol, 2023, 142(1): 31-50. DOI: 10.1097/AOG.0000000000005229.
[6]
MATSUO K, YOUSSEFZADEH A, MANDELBAUM R, et al. Hospital surgical volume-outcome relationship in Caesarean hysterectomy for placenta accreta spectrum[J]. BJOG, 2022, 129(6): 986-993. DOI: 10.1111/1471-0528.16993.
[7]
SILVER R M. Abnormal placentation: placenta previa, Vasa previa, and placenta accreta[J]. Obstet Gynecol, 2015, 126(3): 654-668. DOI: 10.1097/AOG.0000000000001005.
[8]
JAUNIAUX E, BHIDE A, KENNEDY A, et al. FIGO consensus guidelines on placenta accreta spectrum disorders: Prenatal diagnosis and screening[J]. Int J Gynaecol Obstet, 2018, 140(3): 274-280. DOI: 10.1002/ijgo.12408.
[9]
ZHANG H J, DOU R C, YANG H X, et al. Maternal and neonatal outcomes of placenta increta and percreta from a multicenter study in China[J]. J Matern Fetal Neonatal Med, 2019, 32(16): 2622-2627. DOI: 10.1080/14767058.2018.1442429.
[10]
邹锦莉, 胡振远, 王新莲, 等. 基于磁共振T2WI影像组学模型对胎盘植入性疾病进行产前诊断及分型[J]. 磁共振成像, 2024, 15(1): 137-144. DOI: 10.12015/issn.1674-8034.2024.01.022.
ZOU J L, HU Z Y, WANG X L, et al. Radiomics model based on MR T2WI for prenatal diagnosis and classification of placenta accreta spectrum disorders[J]. Chin J Magn Reson Imag, 2024, 15(1): 137-144. DOI: 10.12015/issn.1674-8034.2024.01.022.
[11]
FRATELLI N, FICHERA A, PREFUMO F. An update of diagnostic efficacy of ultrasound and magnetic resonance imaging in the diagnosis of clinically significant placenta accreta spectrum disorders[J]. Curr Opin Obstet Gynecol, 2022, 34(5): 287-291. DOI: 10.1097/GCO.0000000000000811.
[12]
FRATELLI N, PREFUMO F, MAGGI C, et al. Third-trimester ultrasound for antenatal diagnosis of placenta accreta spectrum in women with placenta previa: results from the ADoPAD study[J]. Ultrasound Obstet Gynecol, 2022, 60(3): 381-389. DOI: 10.1002/uog.24889.
[13]
王霞, 赵福敏, 李雅倩, 等. 《腹部放射学会和欧洲泌尿生殖放射学会关于胎盘植入性疾病MRI检查的联合共识声明》要点解读[J]. 中华妇幼临床医学杂志(电子版), 2020, 16(2): 161-170. DOI: 10.3877/cma.j.issn.1673-5250.2020.02.007.
WANG X, ZHAO F M, LI Y Q, et al. Interpretation of society of abdominal radiology and European society of urogenital radiology joint consensus statement for MR imaging of placenta accreta spectrum disorders[J]. Chin J Obstet Gynecol Pediatr Electron Ed, 2020, 16(2): 161-170. DOI: 10.3877/cma.j.issn.1673-5250.2020.02.007.
[14]
KAPOOR H, HANAOKA M, DAWKINS A, et al. Review of MRI imaging for placenta accreta spectrum: Pathophysiologic insights, imaging signs, and recent developments[J]. Placenta, 2021, 104: 31-39. DOI: 10.1016/j.placenta.2020.11.004.
[15]
潘婷, 夏蕾, 石容容, 等. 凶险性前置胎盘伴胎盘植入患者孕晚期胎盘位置、MRI信号征象特点和产前诊断价值[J]. 中国CT和MRI杂志, 2023, 21(2): 131-133. DOI: 10.3969/j.issn.1672-5131.2023.02.044.
PAN T, XIA L, SHI R R, et al. Placental sites and the characteristics of MRI signs in patients with pernicious placenta previa and placenta accreta during the third trimester and their prenatal diagnosis value[J]. Chin J CT MRI, 2023, 21(2): 131-133. DOI: 10.3969/j.issn.1672-5131.2023.02.044.
[16]
MAYERHOEFER M E, MATERKA A, LANGS G, et al. Introduction to radiomics[J]. J Nucl Med, 2020, 61(4): 488-495. DOI: 10.2967/jnumed.118.222893.
[17]
MCCAGUE C, RAMLEE S, REINIUS M, et al. Introduction to radiomics for a clinical audience[J]. Clin Radiol, 2023, 78(2): 83-98. DOI: 10.1016/j.crad.2022.08.149.
[18]
郑蔚然, 杨馨蕊, 闫婕, 等. 结合国际指南,探究胎盘植入性疾病诊治进展[J]. 中华围产医学杂志, 2020, 23(12): 843-848. DOI: 10.3760/cma.j.cn113903-20200726-00700.
ZHENG W R, YANG X R, YAN J, et al. Review on placenta accreta spectrum based on international guidelines[J]. Chin J Perinat Med, 2020, 23(12): 843-848. DOI: 10.3760/cma.j.cn113903-20200726-00700.
[19]
魏芸, 赵青, 杨冬, 等. 改良超声评分量表诊断胎盘植入性疾病的价值[J]. 中国医学影像学杂志, 2023, 31(4): 385-389. DOI: 10.3969/j.issn.1005-5185.2023.04.017.
WEI Y, ZHAO Q, YANG D, et al. Application of modified ultrasound scoring scale in diagnosis of placenta accreta spectrum disorders[J]. Chin J Med Imag, 2023, 31(4): 385-389. DOI: 10.3969/j.issn.1005-5185.2023.04.017.
[20]
赵扬玉, 王妍, 陈练. 胎盘植入的围手术期管理[J]. 中华妇产科杂志, 2018, 53(11): 787-789. DOI: 10.3760/cma.j.issn.0529-567x.2018.11.014.
ZHAO Y Y, WANG Y, CHEN L. Chin J Obstet Gynecol, 2018, 53(11): 787-789. DOI: 10.3760/cma.j.issn.0529-567x.2018.11.014.
[21]
SCHWICKERT A, VAN BEEKHUIZEN H J, BERTHOLDT C, et al. Association of peripartum management and high maternal blood loss at cesarean delivery for placenta accreta spectrum (PAS): A multinational database study[J]. Acta Obstet Gynecol Scand, 2021, 100(Suppl 1): 29-40. DOI: 10.1111/aogs.14103.
[22]
白秀丽, 张素英, 陈杨萍. 胎盘植入性疾病的相关影响因素分析及预测[J]. 中国现代医生, 2024, 62(6): 40-44. DOI: 10.3969/j.issn.1673-9701.2024.06.010.
BAI X L, ZHANG S Y, CHEN Y P. Analysis and prediction of related influencing factors of placental implantation disease[J]. China Mod Dr, 2024, 62(6): 40-44. DOI: 10.3969/j.issn.1673-9701.2024.06.010.
[23]
JAUNIAUX E, COLLINS S, BURTON G J. Placenta accreta spectrum: pathophysiology and evidence-based anatomy for prenatal ultrasound imaging[J]. Am J Obstet Gynecol, 2018, 218(1): 75-87. DOI: 10.1016/j.ajog.2017.05.067.
[24]
熊星, 王佳, 张妤, 等. 基于临床及常规MRI征象Logistic回归模型列线图诊断胎盘植入[J]. 中国医学影像技术, 2021, 37(7): 1049-1053. DOI: 10.13929/j.issn.1003-3289.2021.07.020.
XIONG X, WANG J, ZHANG Y, et al. Logistic regression model nomogram based on clinical and conventional MRI characteristics for diagnosis of placental accreta[J]. Chin J Med Imag Technol, 2021, 37(7): 1049-1053. DOI: 10.13929/j.issn.1003-3289.2021.07.020.
[25]
CELLINA M, PIROVANO M, CIOCCA M, et al. Radiomic analysis of the optic nerve at the first episode of acute optic neuritis: an indicator of optic nerve pathology and a predictor of visual recovery?[J]. Radiol Med, 2021, 126(5): 698-706. DOI: 10.1007/s11547-020-01318-4.
[26]
KOLOSSVÁRY M, GERSTENBLITH G, BLUEMKE D A, et al. Contribution of risk factors to the development of coronary atherosclerosis as confirmed via coronary CT angiography: A longitudinal radiomics-based study[J]. Radiology, 2021, 299(1): 97-106. DOI: 10.1148/radiol.2021203179.
[27]
SUAREZ-IBARROLA R, HEIN S, REIS G, et al. Current and future applications of machine and deep learning in urology: a review of the literature on urolithiasis, renal cell carcinoma, and bladder and prostate cancer[J]. World J Urol, 2020, 38(10): 2329-2347. DOI: 10.1007/s00345-019-03000-5.
[28]
GUIOT J, VAIDYANATHAN A, DEPREZ L, et al. A review in radiomics: Making personalized medicine a reality via routine imaging[J]. Med Res Rev, 2022, 42(1): 426-440. DOI: 10.1002/med.21846.
[29]
NAM D, CHAPIRO J, PARADIS V, et al. Artificial intelligence in liver diseases: improving diagnostics, prognostics and response prediction[J/OL]. JHEP Rep, 2022, 4(4): 100443 [2025-01-08]. https://pubmed.ncbi.nlm.nih.gov/35243281/. DOI: 10.1016/j.jhepr.2022.100443.
[30]
NAZARI M, SHIRI I, ZAIDI H. Radiomics-based machine learning model to predict risk of death within 5-years in clear cell renal cell carcinoma patients[J/OL]. Comput Biol Med, 2021, 129: 104135 [2025-01-08]. https://pubmed.ncbi.nlm.nih.gov/33254045/. DOI: 10.1016/j.compbiomed.2020.104135.
[31]
SUN H Q, QU H B, CHEN L, et al. Identification of suspicious invasive placentation based on clinical MRI data using textural features and automated machine learning[J]. Eur Radiol, 2019, 29(11): 6152-6162. DOI: 10.1007/s00330-019-06372-9.
[32]
ROMEO V, RICCIARDI C, CUOCOLO R, et al. Machine learning analysis of MRI-derived texture features to predict placenta accreta spectrum in patients with placenta previa[J]. Magn Reson Imaging, 2019, 64: 71-76. DOI: 10.1016/j.mri.2019.05.017.
[33]
REN H N, MORI N, MUGIKURA S, et al. Prediction of placenta accreta spectrum using texture analysis on coronal and sagittal T2-weighted imaging[J]. Abdom Radiol, 2021, 46(11): 5344-5352. DOI: 10.1007/s00261-021-03226-1.
[34]
DO Q N, LEWIS M A, MADHURANTHAKAM A J, et al. Texture analysis of magnetic resonance images of the human placenta throughout gestation: A feasibility study[J/OL]. PLoS One, 2019, 14(1): e0211060 [2025-01-08]. https://pubmed.ncbi.nlm.nih.gov/30668581/. DOI: 10.1371/journal.pone.0211060.
[35]
SHU Z Y, MAO D W, SONG Q W, et al. Multiparameter MRI-based radiomics for preoperative prediction of extramural venous invasion in rectal cancer[J]. Eur Radiol, 2022, 32(2): 1002-1013. DOI: 10.1007/s00330-021-08242-9.

上一篇 基于ADC影像组学的机器学习模型预测子宫内膜癌肌层浸润深度的价值
下一篇 R2*、PDFF评估中老年女性腰椎骨髓铁沉积、脂肪含量及对骨质疏松的诊断价值
  
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