综述
MRI新技术在胎盘植入性病变评估的研究进展
磁共振成像, 2023,14(1) : 194-197,202. DOI: 10.12015/issn.1674-8034.2023.01.036
摘要

胎盘植入性病变(placenta accrete spectrum disorders, PAS)是全球孕产妇严重并发症之一,可导致产后出血,增加围手术期子宫切除风险,造成不良妊娠结局。随着剖宫产率的增长,胎盘植入的发生率及产妇死亡率也不断增高,故早期产前诊断对于指导孕产妇和新生儿护理和降低并发症风险至关重要。MRI是评估PAS的一种优秀工具,可对超声怀疑的PAS患者提供额外的信息,如侵犯的范围和程度,是否有宫外受累,以及预测手术过程中的紧急情况(如失血、输血和子宫切除)。MRI新技术扩散加权成像(diffusion-weighted imaging, DWI)、体素内不相干运动(intravoxel incoherent motion, IVIM)DWI、血氧水平依赖(blood oxygen level-dependent, BOLD)成像在准确诊断PAS的基础上,还可以提供更多的胎盘功能信息。另外,基于MRI人工智能技术的应用进一步提高了PAS的诊断准确性,减少了对放射科医生的经验依赖,还能对患者预后、并发症风险等进行评估和预测,从而提升临床对PAS的诊疗决策水平。本文旨在探讨MRI新技术和基于MRI的人工智能技术在PAS评估方面的研究进展,以期为临床提升孕妇产前胎盘植入诊断准确性以及降低术中风险提供一种新的思路和方法。

引用本文: 王颖超, 黄刚. MRI新技术在胎盘植入性病变评估的研究进展 [J] . 磁共振成像, 2023, 14(1) : 194-197,202. DOI: 10.12015/issn.1674-8034.2023.01.036.
参考文献导出:   Endnote    NoteExpress    RefWorks    NoteFirst    医学文献王
扫  描  看  全  文

正文
作者信息
基金 0  关键词  0
English Abstract
评论
阅读 0  评论  0
相关资源
引用 | 论文 | 视频

本刊刊出的所有论文不代表本刊编委会的观点,除非特别声明

0 前言

胎盘植入性病变(placenta accrete spectrum disorders, PAS)是指胎盘绒毛不同程度侵入子宫肌层。依据胎盘植入子宫肌层的深度以及是否侵入子宫毗邻器官分为胎盘粘连(placenta accreta)、胎盘植入(placenta increta)以及穿透性胎盘植入(placenta percreta)[1]。PAS的发病机制通常认为是胎盘滋养细胞经子宫内膜-肌层界面的缺损发生异常深部侵入[2]。随着剖宫产率的不断增长,胎盘植入的发生率也不断增高到24%~67%不等,产妇死亡率估计高达7%[3]。此外,PAS的并发症在未诊断的情况下发生率较高,包括弥括15%的膀胱切开术风险和2%的输尿管损伤风险[4]。因此在分娩前准确地诊断胎盘植入能有效减少产妇的失血和不良结局的发生[5]

MRI的优势是允许更大的视野全面评估胎盘,对描述胎盘的确切位置、粘连或向肌层的植入程度具有很高的可靠性[6, 7]。MRI已被证明是胎盘植入诊断的有用工具,如国际指南[8]中所述,如存在危险因素、胎盘后位或可疑胎盘植入时应行MRI检查。常规的T1WI、T2WI对PAS的诊断敏感度及特异度为82%、88%[9],而随着MRI新技术的发展,如:扩散加权成像(diffusion-weighted imaging, DWI)、体素内不相干运动(intravoxel incoherent motion, IVIM)DWI、血氧水平依赖(blood oxygen level-dependent, BOLD)成像、影像组学及深度学习(deep learning, DL)等技术,在提高诊断PAS准确性的基础上,还可以提供更多的功能信息,以及对患者预后、并发症风险等进行评估和预测,有助于妊娠管理和分娩计划的临床决策,最大限度地减少胎儿和孕产妇围产期并发症。故本文就MRI新技术对PAS评估的研究进展进行综述。

1 扩散成像

扩散成像主要通过检测水分子的扩散方向和规模,间接反映组织微观结构的变化[10]目前已用于多个器官,包括胎盘[11]。而在PAS诊断方面研究比较多的主要是DWI和IVIM。DWI是一种功能MRI技术,能够提供定量的表观扩散系数(apparent diffusion coefficient, ADC),在胎盘成熟过程中ADC值能够反映胎盘细微结构的变化,为比较胎盘正常或病理过程的任何变化提供了一种定量方法[12]。胎盘细胞密度比子宫肌层高,在高b值(>400 s/mm2)的DWI图像更有利于识别胎盘-肌层界面异常,胎盘植入层面主要表面为高信号,同样ADC值也比肌层的要低[12],而常规T2加权成像上仅比正常肌层稍高,因此常规序列联合DWI能够对胎盘植入的分级做出更准确的判断[11]。IVIM是利用多个扩散敏感因子(b值)构建双指数的模型将组织扩散(慢扩散)和毛细血管灌注(快速扩散)分离开[13]。IVIM主要用于肝脏[14, 15]、胰腺[16, 17]、和肾脏[18, 19]分析,由于胎盘也属于高度血管化的器官,含有高血分数和大灌注成分,因此在此方面的研究也越来越多。在PAS的发生机制中,由于子宫壁慢性损伤[20]导致局部疤痕区域的血液循环受到损害,而胎盘绒毛侵入子宫肌层可引起子宫桡状和弓状血管扩张和植入部位新生血管形成,IVIM可观察这种局部的灌注变化[21]。因此,PAS的胎盘-子宫肌层交界处因具有异常侵袭性胎盘床的高血管性质以及伴随着子宫-胎盘界面的血管重构,致大部分植入区在胎盘基板内或基板下均表现出血运过度现象[22]。IVIM中的灌注分数(f)和伪扩散系数(D*)能够反映PAS中的高灌注区域,对胎盘植入有较好的预测效果[23, 24]。有研究计算了粘连性胎盘、植入性胎盘及穿透性胎盘的f和D*值,并发现f值较高的患者,更容易在手术中发生严重出血,也更容易需要输血[22,25]。这也表明IVIM可能在预测危及生命的失血方面发挥有益作用,从而为临床提供个性化诊疗措施。

DWI和IVIM与常规T2WI相比,图像分辨率较低,对呼吸运动、肠运动、磁场不均匀性、磁化率等伪影非常敏感,很少单独用于PAS的诊断[26]。但随着MRI软硬件技术的发展,扩散成像的图像质量和扫描速度都在不断提升,对PAS诊断价值会越来越大。

2 BOLD

BOLD是一种能反映组织血氧水平的功能MRI技术,主要通过观察血液中脱氧血红蛋白的相对数量改变引起的T2*加权图像信号强度变化[27]。这种技术在绘制脑功能图方面有很大的用途,其中空间模式被用来解释功能网络[28]。目前在大脑之外的应用也越来越多[29, 30, 31],其中就包括胎盘。胎盘的BLOD信号是复杂的,信号受组织内血红蛋白浓度、血容量和血流的影响[32]。高氧控制下的BOLD信号和胎儿生长受限的变化已在胎盘和其他胎儿器官中得到证实[33, 34]。而利用BOLD技术观察孕妇吸氧前后胎盘的信号强度变化,发现PAS孕妇胎盘的氧合潜力较正常孕妇胎盘更高,可能的原因是PAS中子宫面的血管重塑使邻近子宫侧胎盘处于一个相对较高的灌注状态[35]。吸氧后重塑血管的氧分压和子宫面胎盘氧分压迅速升高,随之胎儿面胎盘绒毛毛细血管的氧分压升高,这使得PAS胎盘保持了相对正常的功能[36]。但这些相关研究一致的问题是缺乏标准化的MRI协议和检查时间较长,容易造成结果的不一致。此外,没有提及可能影响胎盘功能的母体背景,如目标患者的孕周和母体体型。综合考虑母体背景对胎盘BOLD信号的影响,才能准确评估PAS的实时氧合,以此来对患者进行更准确的风险分层。

3 MRI影像组学及DL
3.1 影像组学及DL概述

图像是一种数据(包含肉眼不能辨识的信息),高通量地挖掘图像特征并通过机器学习(machine learning, ML)可针对临床问题建立各种生物学标签[37]。挖掘图像特征的方式基本可以分为影像组学(radiomics)和DL。影像组学主要通过对数据里隐藏的定量特征(如:形状特征、一阶直方图特征、二阶纹理特征等)的提取并通过ML建模,来实现分类或预测等目的[38]。影像组学特征是人为给定的,故其可解释性强,其工作流程[39]主要有:图像采集与重建、病灶分割、特征提取和特征分析(特征筛选和模型建立)。DL是一种基于多层人工神经网络(artificial neural networks, ANNs)的ML方法,通过自动提取图像特征,最大限度地提高问题任务的模型性能,其流程主要包括输入、卷积、驰化、连接、输出[40]。在医学影像领域中,DL最常被应用于目标检测(如病变位置)、目标分割(如病变轮廓)和目标分类(如恶性与良性病变)[41],其中在病变检测方面,DL模型的诊断效能可与医疗专业人员相当[42]。虽然DL挖掘特征可解释性较差,但建模的效果更好。通常DL对样本数量的要求比影像组学多,与影像组学相比DL的计算成本要高得多[43]

3.2 影像组学及DL在PAS评估中的研究

纹理特征是影像组学超出肉眼可见范围反映异质性的一组特征方法,基于影像组学的纹理分析和ML可以帮助放射科医生识别PAS下的胎盘组织异常,在临床实践中发挥辅助工具的作用,并扩展现有人工智能的进一步应用研究,有研究报道胎龄与妊娠子宫DWI和ADC图像的纹理特征之间存在相关性[44]。与正常胎盘相比,PAS胎盘像素强度标准差和分形分析有显著差异[45]。基于MRI图像纹理影像组学特征构建的ML模型对准确预测PAS取得了良好的效果,其敏感度、特异度、准确率和受试者工作特征(receiver operating characteristic, ROC)曲线下面积(area under the curve, AUC)分别为100.0%、88.5%、95.2%和0.98[46]。采用支持向量机算法对T2WI图像的轴位、冠状位和矢状位进行视觉和纹理分析对比结果中,冠状位和矢状位模型效能(AUC:0.98)显著高于放射科医生的视觉评估(AUC:0.75)。而影像组学特征联合临床特征共同构建的ML模型对诊断PAS具有更高的效能(AUC:0.995)[47]。此外,有研究利用子宫瘢痕区域的胎盘MRI图像纹理特征构建影像组学模型对临床怀疑PAS的孕妇在手术中是否需要切除子宫进行预测,结果表明这些客观的放射组学特征具有很强的预测能力[9]。而有研究通过影像组学分析,发现有35个影像组学特征与估计出血量有很强的相关性,并在多个中心保持稳定,因此结合产前临床因素和T2WI胎盘影像学特征的临床放射学模型对产后出血有很好的预测作用[48, 49]。尽管影像组学已发展为克服视觉判读的主观性和提高诊断准确性的一个重要领域,但目前PAS研究都局限于小样本的回顾性研究,且样本中阴性病例比阳性病例多,不可避免地会存在一些结果偏差,使PAS的影像组学研究在评估胎盘异质性方面低于理想水平[50]。而基于不同的MRI设备和扫描参数构建的组学模型,限制了结果的推广,亟待需要更多的外部数据集来进一步验证。参照影像组学模型在肿瘤性疾病的研究结果,联合一些重要的临床背景因素(如分娩时的孕龄和剖宫产史)也许能进一步提升模型的评估PAS效能[48]。此外,病灶手动和半自动分割是影像组学研究最常用的方法,研发高精准的自动化分割方法可以大大提升影像组学的临床易用性和可重复性。

在解决同一个目标问题时,DL往往具有比影像组学更高的效能和更好的鲁棒性,但是DL对数据量的要求更高,要获得高效能的模型往往需要大量的数据。DL在PAS的评估方面研究甚少,可能与临床PAS数据不够丰富有关。已有的研究利用卷积神经网络(convolutional neural network, CNN)实现对女性子宫的自动分割;全三维全卷积DL多类分割技术能够在单个CNN模型下对三维MRI图像的宫腔和胎盘进行高精度分割,通过检测子宫和胎盘的形状和位置,对子宫腔和胎盘的体积进行准确估算[51]。DL和T2WI图像的影像组学特征与临床因素相结合构建的诺谟图可更准确地预测胎盘植入,其预测精度为94.1%,明显优于单纯依靠放射组学特征和DL特征的传统ML方法[47,52]。对U-net进行训练后,在MRI图像进行胎盘区域的分割,并提取其影像组学特征,通过构建一个深层动态模型CNN来提取深层特征,可实现产前诊断胎盘植入并确定其亚型,从而为胎盘植入的产前诊断及分型提供一种新的方法[53]。基于DL的模型,还可用于预测凶险性前置胎盘患者剖宫产术中的出血量,可作为降低凶险性前置胎盘患者术中出血风险的临床辅助工具,从而辅助术前止血计划[54]。但基于DL的分割技术会由于不同怀孕阶段胎儿的大小、外观变化影响分割性能,未来还需要在更大的数据集上进行不断迭代和测试[55]

4 小结

随着MRI新技术的不断发展,DWI进一步提高了胎盘-肌层界面异常的识别能力,在提供胎盘植入深度、位置的基础上,使复杂妊娠的胎盘功能诊断成为可能。IVIM可定量和无创地表征胎盘在体内的灌注,是诊断PAS的一种有用的功能性诊断技术。此外,IVIM在预测PAS患者危及生命的失血量方面可能发挥有益作用,但这是一个值得进一步研究的领域。BOLD作为一种直接估计胎盘组织氧合变化的无创方法,在PAS中可评估胎儿血液在胎盘中的实时氧合,这有助于未来的宫内治疗,并确定治疗效果和风险分类。影像组学及DL为我们提供了一种新的方法来反映组织或肿瘤的异质性,并且比主观评价更趋于稳定,因此是诊断PAS甚至评估PAS严重程度的可行工具,未来有望对患者预后、术中风险及相关并发症等进行评估和预测,从而为临床提供更多可靠的影像学依据,对孕产妇预后及治疗具有更高的指导价值。

本文引用格式:

王颖超, 黄刚. MRI新技术在胎盘植入性病变评估的研究进展[J]. 磁共振成像, 2023, 14(1): 194-197, 202.

Cite this article as:

WANG Y C, HUANG G. Research progress of MRI in placenta accrete spectrum disorders[J]. Chin J Magn Reson Imaging, 2023, 14(1): 194-197, 202.

ACKNOWLEDGMENTS

University Innovation Fund Project of Education Department of Gansu Province (No. 2021B-232); Research Fund Project for Young Teachers of Hexi University (No. QN2020005).

利益冲突
作者利益冲突声明:

全体作者均声明无利益冲突。

参考文献References
[1]
DONOVANB M, SHAINKERS A. Placenta Accreta Spectrum[J/OL]. Neoreviews, 2021, 22(11): e722-e733 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/34725137/. DOI: 10.1542/neo.22-11-e722.
[2]
TANTBIROJNP, CRUMC P, PARASTM M. Pathophysiology of placenta creta: the role of decidua and extravillous trophoblast[J]. Placenta, 2008, 29(7): 639-645. DOI: 10.1016/j.placenta.2008.04.008.
[3]
ZUCKERWISEL C, CRAIGA M, NEWTONJ M, et al. Outcomes following a clinical algorithm allowing for delayed hysterectomy in the management of severe placenta accreta spectrum[J/OL]. Am J Obstet Gynecol, 2020, 222(2): 179.e171-179.e179 [2022-12-05]. https://pubmed.ncbi.nlm.nih.gov/31469990/. DOI: 10.1016/j.ajog.2019.08.035.
[4]
SILVERR M, LANDONM B, ROUSED J, et al. Maternal morbidity associated with multiple repeat cesarean deliveries[J]. Obstet Gynecol, 2006, 107(6): 1226-1232. DOI: 10.1097/01.AOG.0000219750.79480.84.
[5]
SCHWICKERTA, VAN BEEKHUIZENH J, BERTHOLDTC, 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.
[6]
FIOCCHIF, MONELLIF, BESUTTIG, et al. MRI of placenta accreta: diagnostic accuracy and impact of interventional radiology on foetal-maternal delivery outcomes in high-risk women[J/OL]. Br J Radiol, 2020, 93(1114): 20200267 [2022-12-05]. https://pubmed.ncbi.nlm.nih.gov/32706979/. DOI: 10.1259/bjr.20200267.
[7]
FINAZZOF, D'ANTONIOF, MASSELLIG, et al. Interobserver agreement in MRI assessment of severity of placenta accreta spectrum disorders[J]. Ultrasound Obstet Gynecol, 2020, 55(4): 467-473. DOI: 10.1002/uog.20381.
[8]
CANTWELLR, CLUTTON-BROCKT, COOPERG, et al. Saving Mothers' Lives: reviewing maternal deaths to make motherhood safer: 2006-2008. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom[J]. BJOG, 2011, 118(Suppl 1): 1-203. DOI: 10.1111/j.1471-0528.2010.02847.x.
[9]
DOQ N, LEWISM A, XIY, et al. MRI of the placenta accreta spectrum (PAS) disorder: radiomics analysis correlates with surgical and pathological outcome[J]. J Magn Reson Imaging, 2020, 51(3): 936-946. DOI: 10.1002/jmri.26883.
[10]
TAVAKOLIA, KRAMMERJ, ATTENBERGERU I, et al. Simultaneous multislice diffusion-weighted imaging of the kidneys at 3 T[J]. Invest Radiol, 2020, 55(4): 233-238. DOI: 10.1097/RLI.0000000000000637.
[11]
SANNANANJAB, ELLERMEIERA, HIPPED S, et al. Utility of diffusion-weighted MR imaging in the diagnosis of placenta accreta spectrum abnormality[J]. Abdom Radiol (NY), 2018, 43(11): 3147-3156. DOI: 10.1007/s00261-018-1599-8.
[12]
HOROWITZJ M, HOTALENI M, MILLERE S, et al. How can pelvic MRI with diffusion-weighted imaging help my pregnant patient?[J]. Am J Perinatol, 2020, 37(6): 577-588. DOI: 10.1055/s-0039-1685492.
[13]
FEDERAUC. Intravoxel incoherent motion MRI as a means to measure in vivo perfusion: A review of the evidence[J/OL]. NMR Biomed, 2017, 30: e3780 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/28885745/. DOI: 10.1002/nbm.3780.
[14]
LEFEBVRET, HÉBERTM, BILODEAUL, et al. Intravoxel incoherent motion diffusion-weighted MRI for the characterization of inflammation in chronic liver disease[J]. Eur Radiol, 2021, 31(3): 1347-1358. DOI: 10.1007/s00330-020-07203-y.
[15]
CHENGZ, YANGQ, HEH, et al. Intravoxel incoherent motion diffusion-weighted imaging and shear wave elastography for evaluating peritumoral liver fibrosis after transarterial chemoembolization in a VX2 rabbit liver tumor model[J/OL]. Front Physiol, 2022, 13: 893925 [2022-12-05]. https://pubmed.ncbi.nlm.nih.gov/36311244/. DOI: 10.3389/fphys.2022.893925.
[16]
VAN V DPHI, BECKERA S, CIRITSISA, et al. Intravoxel incoherent motion analysis of abdominal organs: application of simultaneous multislice acquisition[J]. Invest Radiol, 2018, 53(3): 179-185. DOI: 10.1097/RLI.0000000000000426.
[17]
GURNEY-CHAMPIONO J, FROELINGM, KLAASSENR, et al. Minimizing the acquisition time for intravoxel incoherent motion magnetic resonance imaging acquisitions in the liver and pancreas[J]. Invest Radiol, 2016, 51(4): 211-220. DOI: 10.1097/RLI.0000000000000225.
[18]
BANEO, HECTORSS J, GORDICS, et al. Multiparametric magnetic resonance imaging shows promising results to assess renal transplant dysfunction with fibrosis[J]. Kidney Int, 2020, 97(2): 414-420. DOI: 10.1016/j.kint.2019.09.030.
[19]
ZHANGJ L, LEEV S. Renal perfusion imaging by MRI[J]. J Magn Reson Imaging, 2020, 52(2): 369-379. DOI: 10.1002/jmri.26911.
[20]
JAUNIAUXE, COLLINSS, BURTONG 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.
[21]
BAOY W, PANGY, SUNZ Y, et al. Functional diagnosis of placenta accreta by intravoxel incoherent motion model diffusion-weighted imaging[J]. Eur Radiol, 2021, 31(2): 740-748. DOI: 10.1007/s00330-020-07200-1.
[22]
LUT, PUH, LIK D, et al. Can introvoxel incoherent motion MRI be used to differentiate patients with placenta accreta spectrum disorders?[J/OL]. BMC Pregnancy Childbirth, 2019, 19(1): 531 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/31888572/. DOI: 10.1186/s12884-019-2676-x.
[23]
SIAUVEN, HAYOTP H, DELOISONB, et al. Assessment of human placental perfusion by intravoxel incoherent motion MR imaging[J]. J Matern Fetal Neonatal Med, 2019, 32(2): 293-300. DOI: 10.1080/14767058.2017.1378334.
[24]
LUT, WANGY, GUOA, et al. Monoexponential, biexponential and diffusion kurtosis MR imaging models: quantitative biomarkers in the diagnosis of placenta accreta spectrum disorders[J/OL]. BMC Pregnancy Childbirth, 2022, 22(1): 349 [2022-12-05]. https://pubmed.ncbi.nlm.nih.gov/35459146/. DOI: 10.1186/s12884-022-04644-9.
[25]
LUT, SONGB, PUH, et al. Prognosticators of intravoxel incoherent motion (IVIM) MRI for adverse maternal and neonatal clinical outcomes in patients with placenta accreta spectrum disorders[J]. Transl Androl Urol, 2020, 9(2): 258-266. DOI: 10.21037/tau.2019.12.27.
[26]
KRISTIB A, DITTEN H, CAROLINEH, et al. Placental diffusion-weighted MRI in normal pregnancies and those complicated by placental dysfunction due to vascular malperfusion[J]. Placenta, 2020, 91: 52-58. DOI: 10.1016/j.placenta.2020.01.009.
[27]
SCHABELM C, ROBERTSV H J, LOJ O, et al. Functional imaging of the nonhuman primate Placenta with endogenous blood oxygen level-dependent contrast[J]. Magn Reson Med, 2016, 76(5): 1551-1562. DOI: 10.1002/mrm.26052.
[28]
GRUTZENDLERJ, NEDERGAARDM. Cellular control of brain capillary blood flow: in vivo imaging veritas[J]. Trends Neurosci, 2019, 42(8): 528-536. DOI: 10.1016/j.tins.2019.05.009.
[29]
YOUW, ANDESCAVAGEN N, KAPSEK, et al. Hemodynamic responses of the placenta and brain to maternal hyperoxia in fetuses with congenital heart disease by using blood oxygen-level dependent MRI[J]. Radiology, 2020, 294(1): 141-148. DOI: 10.1148/radiol.2019190751.
[30]
VAN DEN BOOMENM, MANHARDM K, SNELG J H, et al. Blood oxygen level-dependent MRI of the myocardium with multiecho gradient-echo spin-echo imaging[J]. Radiology, 2020, 294(3): 538-545. DOI: 10.1148/radiol.2020191845.
[31]
JIANGK, FERGUSONC M, LERMANL O. Noninvasive assessment of renal fibrosis by magnetic resonance imaging and ultrasound techniques[J]. Transl Res, 2019, 209: 105-120. DOI: 10.1016/j.trsl.2019.02.009.
[32]
SIAUVEN, CHALOUHIG E, DELOISONB, et al. Functional imaging of the human placenta with magnetic resonance[J]. Am J Obstet Gynecol, 2015, 213(4): S103-S114. DOI: 10.1016/j.ajog.2015.06.045.
[33]
INGRAME, MORRISD, NAISHJ, et al. MR imaging measurements of altered placental oxygenation in pregnancies complicated by fetal growth restriction[J]. Radiology, 2017, 285(3): 953-960. DOI: 10.1148/radiol.2017162385.
[34]
MAGAWAS, NIIM, ENOMOTON, et al. Evaluation of placental oxygenation in fetal growth restriction using blood oxygen level-dependent magnetic resonance imaging[J]. Placenta, 2022, 126: 40-45. DOI: 10.1016/j.placenta.2022.06.005.
[35]
MAGAWAS, NIIM, ISHIDAM, et al. Evaluation of placental oxygenation index using blood oxygen level-dependent magnetic resonance imaging (BOLD-MRI) during normal late pregnancy[J]. J Matern Fetal Neonatal Med, 2022, 35(25): 5274-5281. DOI: 10.1080/14767058.2021.1878140.
[36]
SINDINGM, PETERSD A, POULSENS S, et al. Placental baseline conditions modulate the hyperoxic BOLD-MRI response[J]. Placenta, 2018, 61: 17-23. DOI: 10.1016/j.placenta.2017.11.002.
[37]
GILLIESR J, KINAHANP E, HRICAKH. Radiomics: images are more than pictures, they are data[J]. Radiology, 2016, 278(2): 563-577. DOI: 10.1148/radiol.2015151169.
[38]
ATHERS, KADIRT, GLEESONF. Artificial intelligence and radiomics in pulmonary nodule management: current status and future applications[J]. Clin Radiol, 2020, 75(1): 13-19. DOI: 10.1016/j.crad.2019.04.017.
[39]
LAMBINP, LEIJENAARR T H, DEISTT M, et al. Radiomics: the bridge between medical imaging and personalized medicine[J]. Nat Rev Clin Oncol, 2017, 14(12): 749-762. DOI: 10.1038/nrclinonc.2017.141.
[40]
HOSNYA, PARMARC, QUACKENBUSHJ, et al. Artificial intelligence in radiology[J]. Nat Rev Cancer, 2018, 18(8): 500-510. DOI: 10.1038/s41568-018-0016-5.
[41]
CURRIEG M. Intelligent imaging: artificial intelligence augmented nuclear medicine[J]. J Nucl Med Technol, 2019, 47(3): 217-222. DOI: 10.2967/jnmt.119.232462.
[42]
GHAFOORIANM, KARSSEMEIJERN, HESKEST, et al. Location Sensitive Deep Convolutional Neural Networks for Segmentation of White Matter Hyperintensities[J/OL]. Sci Rep, 2017, 7(1): 5110 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/28698556/. DOI: 10.1038/s41598-017-05300-5.
[43]
ZHOUY, YENG G, YIZ. Evolutionary compression of deep neural networks for biomedical image segmentation[J]. IEEE Trans Neural Netw Learn Syst, 2020, 31(8): 2916-2929. DOI: 10.1109/TNNLS.2019.2933879.
[44]
ROMEOV, RICCIARDIC, CUOCOLOR, 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.
[45]
CHENE, MARW A, HOROWITZJ M, et al. Texture analysis of placental MRI: can it aid in the prenatal diagnosis of placenta accreta spectrum?[J]. Abdom Radiol (NY), 2019, 44(9): 3175-3184. DOI: 10.1007/s00261-019-02104-1.
[46]
SUNH Q, QUH B, CHENL, 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.
[47]
路涛, 张天悦, 李缃琦, . 基于T2WI影像组学模型产前预测胎盘植入性病变[J]. 中国医学影像技术, 2021, 37(12): 1854-1859. DOI: 10.13929/j.issn.1003-3289.2021.12.022.
LUT, ZHANGT Y, LIX Q, et al. Radiomics model based on T2WI for prenatal predicting placenta accreta spectrum disorders[J]. Chin J Med Imaging Technol, 2021, 37(12): 1854-1859. DOI: 10.13929/j.issn.1003-3289.2021.12.022.
[48]
WUQ X, YAOK, LIUZ Y, et al. Radiomics analysis of placenta on T2WI facilitates prediction of postpartum haemorrhage: a multicentre study[J]. EBioMedicine, 2019, 50: 355-365. DOI: 10.1016/j.ebiom.2019.11.010.
[49]
WUQ, YAOK, LIUZ, et al. Erratum to 'Radiomics analysis of placenta on T2WI facilitates prediction of postpartum haemorrhage: A multicentre study'[J/OL]. EBioMedicine, 2020, 55: 102773 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/32335372/. DOI: 10.1016/j.ebiom.2020.102773.
[50]
STANZIONEA, VERDEF, CUOCOLOR, et al. Placenta Accreta Spectrum Disorders and Radiomics: Systematic review and quality appraisal[J/OL]. Eur J Radiol, 2022, 155: 110497 [2022-12-05]. https://pubmed.ncbi.nlm.nih.gov/36030661/. DOI: 10.1016/j.ejrad.2022.110497.
[51]
KURATAY, NISHIOM, KIDOA, et al. Automatic segmentation of the uterus on MRI using a convolutional neural network[J/OL]. Comput Biol Med, 2019, 114: 103438 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/31521902/. DOI: 10.1016/j.compbiomed.2019.103438.
[52]
SHAOQ, XUANR R, WANGY T, et al. Deep learning and radiomics analysis for prediction of placenta invasion based on T2WI[J]. Math Biosci Eng, 2021, 18(5): 6198-6215. DOI: 10.3934/mbe.2021310.
[53]
XUANR, LIT, WANGY, et al. Prenatal prediction and typing of placental invasion using MRI deep and radiomic features[J/OL]. Biomed Eng Online, 2021, 20(1): 56 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/34090428/. DOI: 10.1186/s12938-021-00893-5.
[54]
LIUJ, WUT, PENGY, et al. Grade Prediction of Bleeding Volume in Cesarean Section of Patients With Pernicious Placenta Previa Based on Deep Learning[J/OL]. Front Bioeng Biotechnol, 2020, 8: 343 [2022-04-06]. https://pubmed.ncbi.nlm.nih.gov/32426340/. DOI: 10.3389/fbioe.2020.00343.
[55]
RAVANBAKHSHM, TSCHERNEZKIV, LASTF, et al. Human-machine collaboration for medical image segmentation[J]. Proc IEEE Int Conf Acoust Speech Signal Process, 2020, 2020: 1040-1044. DOI: 10.1109/ICASSP40776.2020.9053555.
 
 
展开/关闭提纲
查看图表详情
回到顶部
放大字体
缩小字体
标签
关键词