分享:
分享到微信朋友圈
X
综述
磁共振成像在乳腺癌诊断及预后评估中的应用现状及研究进展
吴俊锋 刘文亚

Cite this article as: WU J F, LIU W Y. Application and research progress of MRI in diagnosis and prognosis evaluation of breast cancer[J]. Chin J Magn Reson Imaging, 2023, 14(4): 171-175.本文引用格式:吴俊锋, 刘文亚. 磁共振成像在乳腺癌诊断及预后评估中的应用现状及研究进展[J]. 磁共振成像, 2023, 14(4): 171-175. DOI:10.12015/issn.1674-8034.2023.04.030.


[摘要] 乳腺癌是全球女性最常见的癌症。近年来,随着磁共振成像(magnetic resonance imaging, MRI)在乳腺疾病诊断中的应用越来越广泛,乳腺良恶性病变的诊断准确率逐步提高。同时,MRI可以评估乳腺癌患者的预后并指导临床选择治疗方案。本文就术前多模态MRI及基于MRI的影像组学与人工智能在乳腺癌诊断及预后评估中的应用现状及研究进展展开综述,从而加强影像医师对乳腺癌的认识,提高对女性人群乳腺癌的早期诊断及预后评估水平。
[Abstract] Breast cancer is the most common cancer in women worldwide. In recent years, magnetic resonance imaging (MRI) has been widely applied in the diagnosis of breast diseases, which has improved the diagnostic accuracy of benign and malignant breast lesions. Meanwhile, MRI can be used to predict the prognosis of patients with breast cancer and guide the clinical selection of treatment plans. This article reviews the application status and research advances of preoperative multi-model MRI, MRI radiomics and artificial intelligence (AI) in the diagnosis and prognosis of breast cancer, aiming to strengthen the understanding of radiologist to breast cancer and to improve the early diagnosis and prognosis evaluation of breast cancer.
[关键词] 乳腺癌;诊断;预后;磁共振成像;扩散加权成像;动态对比增强磁共振成像;影像组学;人工智能
[Keywords] breast cancer;diagnosis;prognosis;magnetic resonance imaging;diffusion‐weighted imaging;dynamic contrast-enhanced magnetic resonance imaging;radiomics;artificial intelligence

吴俊锋    刘文亚 *  

新疆医科大学第一附属医院放射科,乌鲁木齐 830000

通信作者:刘文亚,E-mail:13999202977@163.com

作者贡献声明:刘文亚设计本研究的方案,对稿件的重要内容进行了修改;吴俊锋起草和撰写稿件,获取、分析或解释本研究的文献。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。


收稿日期:2022-11-27
接受日期:2023-04-05
中图分类号:R445.2  R737.9 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2023.04.030
本文引用格式:吴俊锋, 刘文亚. 磁共振成像在乳腺癌诊断及预后评估中的应用现状及研究进展[J]. 磁共振成像, 2023, 14(4): 171-175. DOI:10.12015/issn.1674-8034.2023.04.030.

0 前言

       2020年,乳腺癌已超过肺癌成为全球女性最常见的癌症,约占女性癌症病例的1/4和女性癌症死亡病例的1/6[1]。预计在2022年,中国和美国女性乳腺癌发病例数分别约40万、25万,均占女性人群常见肿瘤的第一位,死亡例数分别约为12万、4万[2]。在目前的临床实践中,乳腺癌预后评估及治疗方案制订主要依赖雌激素受体(estrogen receptor, ER)、孕激素受体(progestogen receptor, PR)及人表皮生长因子受体2(human epidermal growth factor receptor, HER-2)、Ki-67的表达及组织学分级等分子生物学及组织学信息[3, 4, 5, 6]。然而,这些信息需要活检或术后病理才能获得,具有侵入性、滞后性及不可重复性等缺点。多模态MRI具有高软组织分辨率、无创性及可重复性等优点,已常规应用于乳腺癌的临床诊疗及预后评估,为临床决策提供影像依据[7]。乳腺癌MRI影像组学通过高通量提取和定量分析肿瘤信息,有望能提高乳腺癌MRI诊断的准确性,并为乳腺癌患者制订科学、精准、个性化治疗方案提供更多依据[8]。本文就术前多模态MRI及基于MRI的影像组学在乳腺癌的诊疗研究进展予以综述,有利于提高乳腺癌的早期诊断及预后评估,改善乳腺癌患者预后及生存质量。

1 多模态MRI对乳腺癌的诊断及预后评估

1.1 MRI形态学特征

       乳腺癌MRI形态学特征包括肿瘤大小、边缘、强化表现、瘤周水肿、背景强化等。乳腺癌常表现为不规则形状(83.59%)、边缘不清楚(85.50%)、肿块样强化(52.31%)、内部不均匀强化(71.72%)。在动态对比增强MRI(dynamic contrast enhanced MRI, DCE-MRI),明显血管化的肿瘤常趋向表现为早期明显强化,延迟期廓清明显,其时间-信号强度曲线多为Ⅱ型或Ⅲ型(91.17%),与良性病灶有明显区别[9, 10]。在肿瘤信号特点上,T1WI呈低信号,T2WI表现为等或低信号,提示肿瘤纤维化。相比之下,三阴性乳腺癌常因肿瘤内坏死表现为高信号[11, 12]。乳腺癌MRI形态学特征对肿瘤分子分型及预后评估具有较大价值[13]。SEYFETTIN等[14]发现Luminal A型乳腺癌形态多不规则,而HER-2阳性及三阴性乳腺癌多为椭圆形或圆形,轮廓光滑;腋窝淋巴结转移在三阴性乳腺癌最常见。据BORIA等[15]报道,对于肿块型乳腺癌,较小的肿瘤(直径<2 cm)与低组织学分级Ki-67阴性表达、Luminal A型相关,肿瘤边缘毛刺与其低组织学分级ER及PR阳性及HER-2阴性相关。瘤周水肿与肿瘤的复发具有独立的相关性[16],可作为预后不良的指标之一。MBERU等[17]发现患侧乳腺皮肤增厚并强化与更差的预后有关。CHEN等[18]的研究提示三阴性乳腺癌多为环形强化,而边缘不规则、强化均匀以及T2WI呈中或低的信号强度与Luminal型乳腺癌有相关性。乳腺癌内部强化与Luminal B型存在关联,研究中的28例Luminal B型病灶中没有一例表现为均匀强化[19]。病灶直径<2 cm提示Luminal A型,环形强化、病灶内坏死、瘤周水肿提示三阴性乳腺[20]

       除了癌灶及瘤周特征,乳腺实质背景强化(background parenchymal enhancement, BPE)与乳腺癌的诊断、预后及治疗反应亦有相关性。BPE可以分低BPE及高BPE。高BPE与致密型腺体以及乳腺癌假阳性率有相关性[21]。此类人群患癌风险更大,应经常随访及筛查癌灶[22, 23, 24]。乳腺癌新辅助化疗后BPE受抑制是乳腺增强MRI的常见表现。对于激素受体阳性乳腺癌,新辅助化疗后没有BPE可能提示其治疗反应差[25]。LIM等[26]研究认为在绝经后乳腺癌术前MRI增高的BPE可能是较差的无复发率的预测因素。而另有学者认为BPE与乳腺癌患者无复发率及总体生存率无显著相关性[27, 28]

       由此可见,乳腺MRI形态学特征评估乳腺癌预后及分子分型有提示意义,病灶直径<2 cm、无瘤周水肿、无患侧乳腺皮肤增厚及强化可能提示预后良好。乳腺癌灶边缘规则,T2WI呈高信号并伴瘤周水肿,增强后呈环形强化常提示三阴性乳腺癌。但这些定性特征一般为影像医师的主观判断,容易受临床经验影响,其临床价值相对受限,通过人工智能(artificial intelligencek, AI)定量传统形态学特征,提高可重复性,将成为今后MRI的发展之路。

1.2 乳腺MRI功能学参数

       乳腺MRI功能成像是目前研究的一大热点,主要包括通过扩散技术及动态对比增强产生的相关参数。研究证实,MRI功能成像在乳腺癌的诊断及预后评估中存在重要价值。

       扩散加权成像(diffusion-weighted imaging, DWI)可评估人体水分子的生理特征及功能状态,b值表示扩散加权梯度的强度。表观扩散系数(apparent diffusion coefficient, ADC)反映组织的生物学特征,比如细胞的性质及含水量。乳腺癌中位ADC值较良性肿瘤低。当ADC值低于1.0×10-3 mm2/s时考虑乳腺癌相对可靠,但高于此值不能排除乳腺癌[29, 30, 31, 32]。ADC值除了应用于乳腺癌诊断,也可应用于评估乳腺癌的预后。乳腺癌中位ADC值在ER阴性组比阳性组高(1.110×10-3 mm2/s vs. 1.050×10-3 mm2/s,P=0.015);相较于Luminal A、Luminal B/HER-2阴性的乳腺癌,HER-2阳性乳腺癌中位ADC值最高(1.190×10-3 mm2/s)[33]。CHEN等[18]的研究表明三阴性乳腺癌ADC值最低[(0.910±0.184)×10-3 mm2/s,P<0.001],但其研究病例样本含量少,可能存在误差。然而,DWI中ADC值测量可能会有误差,且其图像信噪比高,空间分辨率低,容易受设备条件干扰损失重要信息。近年来,通过DCE-MRI药代动力学计算的定量参数与乳腺癌的生物预后因子及分析分型的相关性成为研究热点。定量参数能较准确地反映肿瘤的微循环,常用的有容量转移常数(Ktrans)、速率常数(kep)、血管外细胞外间隙容积比(Ve)。Ktrans反映对比剂从血管到细胞外液间隙速率,kep反映流入细胞外液间隙的对比剂反流回血浆快慢。Ktrans及Kep中位值在恶性病灶比良性更高[34]。ER、PR阳性的乳腺癌Ktrans、Kep均低于ER、PR阴性,且ER、PR表达与Ktrans、Kep及ADC值均呈负相关,三阴性乳腺癌Ktrans、Kep高于其他各型,Ve低于其他各型[35],提示三阴性乳腺癌的新生毛细血管密度高,微血管灌注更明显。另有研究指出Ktrans、Kep与Ki-67呈显著正相关,主要是由于未成熟血管的增加导致了对比剂在血管间及周围组织的流出及回流加速,导致了更高的Ktrans、Kep[36]。定量参数具有较好的客观性,需要良好的动态对比增强图像质量并受不同设备及扫描参数的影响,计算模型不同,数值可能也存在差异。总之随着乳腺波谱成像等技术的应用,MRI功能成像对乳腺癌的应用前景将更加广阔。

2 基于MRI的影像组学与AI在乳腺癌的研究及应用

2.1 基于MRI的影像组学与AI理论

       影像组学是一种新兴的医学转化领域,从放射图像中提取高维数据,旨在建立可应用于临床实践的可靠模型,以期用于诊断疾病、判断预后及评估疾病对治疗的反应。AI的概念是在20世纪被提出的,可以理解为机器从大量代表性样本中识别并学习某种特定关系或模式的能力[37]。AI的模式识别算法分为机器学习算法和深度学习算法;机器学习是指利用某些算法指导计算机利用影像组学提取的图像特征数据得出适当的模型,并利用该模型对新数据给出判断的过程。它不能直接从图像中学习,需要人工手动提取特征[38]。深度学习是机器学习的一种极大修饰,从图像中自动提取特征并分类,实现了机器的自动学习[39, 40]。在乳腺图像领域,AI正成为一个关键组成部分,包括乳腺癌诊断、检测新辅助治疗以及预测治疗结果[41]。AI可以帮助影像医师识别病灶,提高乳腺癌的检出率及准确性。

2.2 对乳腺癌的自动检出

       ZHANG等[42]从MRI各序列及其参数提取影像组学特征构建模型,将207位患者分为训练组(159人)和验证组(48人),结果显示最佳影像组学模型受试者工作特征曲线下面积为0.921,准确率为83.3%,提示影像组学特征建立的模型在鉴别乳腺病灶良恶性方面有很强的能力。LIU等[43]基于DCE-MRI的影像组学研究,使用133位患者的99个纹理和直方图参数来区分病变的良恶性,通过影像组学提取特征建立模型,诊断准确率达到84%,用机器学习算法构建卷积神经网络(convolutional neural network, CNN)的模型,其诊断的准确率达到了91%,表明机器学习有作为临床诊断工具的发展潜力。DALMIŞ等[44]利用机器学习开发了一个计算机辅助检测系统,该系统利用早期增强扫描获得的空间信息寻找病灶,平均敏感度为64.29%±5.37%,可用于简易MRI检查模式的病灶筛选。ZHOU等[45]通过对1537位女性乳腺MRI图像回顾性分析,构建CNN模型以期检测出病灶,准确率为83.7%,敏感度为90.8%,特异度为69.3%。但是,AI对乳腺癌的诊断也存在一定局限性,其识别出的多为较明显且孤立的肿块型病灶,而BPE可能被认为是非肿块型病灶。

2.3 对乳腺癌病理类型及分子分型的预测

       在临床中获得乳腺癌的病理类型及分子分型通常依赖有创性的检查或者手术。近年来,随着AI技术在乳腺MRI的不断发展,学者开始将AI用于预测乳腺癌分子分型。例如,XU等[46]发现使用DCE-MRI联合瘤内及瘤周特征的影像组学模型对浸润性乳腺癌有良好的术前预测能力,有助于乳腺癌患者在保乳术前制订个体化手术计划。ZHANG等[47]分析128名浸润性导管癌患者ADC图并提取特征,建立出基于ADC的放射组学模型。该模型能无创地预测浸润性导管癌Ki-67指数,AUC为0.75±0.08,准确率约70%。YIN等[48]对136例浸润性乳腺癌基于MRI(对比增强图像、ADC及T2WI)构建三种CNN模型。结果发现,基于对比增强图像构建的CNN模型在预测乳腺癌分子分型效果最好(AUC:0.762~0.920),其在鉴别三阴性及HER-2过表达乳腺癌的准确率分别为85.7%及84.7%。HA等[49]开发一种新型CNN模型预测216例乳腺癌的分子分型,提高了Luminal A、B型的鉴别能力(准确率为70%,AUC为0.853)。SHENG等[50]使用图像处理软件提取分析190例中国女性浸润性导管癌DCE-MRI的定量影像组学特征,在鉴别三阴性及非三阴性浸润性导管癌的有较强的能力(准确率为85.71%,AUC为0.903)。

       不同研究模型对预测乳腺癌分子分型均有一定价值,但准确率及AUC也有所差别。这可能与MRI图像质量不一致、特征提取方法差异及患者个体临床及病理特点等因素有关。在未来,通过国际化、多中心合作取得标准图像,扩大样本量,分种族研究并结合临床及病理特点构建模型,联合放射科医师的判断,或能提高其精准率,以便AI技术能更好应用于临床。

2.4 对乳腺癌新辅助化疗疗效的预测

       新辅助化疗可缩小肿瘤大小,降低腋窝淋巴结清扫的必要性,甚至病理完全缓解[51]。在治疗过程的早期预测哪些患者会对新辅助化疗产生应答很重要,因为这有助于减少不必要的毒性。LI等[52]对448例浸润性导管癌患者提取并过滤肿瘤及瘤周的MRI特征并构建影像组学模型,训练组及验证组AUC分别为0.98及0.92。EUN等[53]对136例乳腺癌患者在新辅助化疗前和3~4个周期后(治疗中期)行MRI检查,采用随机森林法构建预测模型,结合纹理参数对病理完全缓解患者分类并检测模型的效能,结果发现在新辅助化疗中期时,采用随机森林法对比增强图像的纹理参数是有价值的(AUC为0.82),并与病理完全缓解相关。因此,基于乳腺MRI的影像组学对预测经新辅助化疗的乳腺癌患者的疗效有较高价值,提供了治疗结果的关键信息,无反应者可选择其他治疗办法,促进新型靶向治疗的前期使用,有利于完善乳腺癌的精准诊疗。

2.5 对乳腺癌复发风险的预测

       系统评估乳腺癌复发风险有利于乳腺癌精准化治疗,改善患者预后。MA等[54]利用深度学习算法对137例三阴性乳腺癌患者新辅助化疗前后DCE-MRI图像的肿瘤区域自动分割建立影像组学模型,对预测三阴性乳腺癌患者新辅助化疗后3年内是否复发有较好的预测性能。KOH等[55]对231例三阴性乳腺癌患者基于DCE-MRI提取三维影像组学特征构建预测患者全身复发的模型,使用不同MRI设备分测试集及验证集,分析结合了临床及病理数据,结果显示此模型能预测三阴性乳腺癌患者系统型复发。CHITALIA等[56]基于10余年前95例乳腺癌术前DCE-MRI提取病灶异质性组学特征构建模型,并用同时期163例患者的图像验证,结果为肿瘤异质性的组学模型可以预测其10年复发风险。总之,影像组学模型预测乳腺癌复发风险可以实现,AI的准确性也得到很多研究的证实,但仍需要更多的外部验证。

2.6 术前预测腋窝淋巴结转移

       临床评估腋窝淋巴结状态主要基于侵入性操作,效率较低,可能导致假阴性率。术后并发症的发生也会影响患者预后。无创且精准地预测腋窝淋巴结转移可避免不必要的腋窝淋巴结活检或清扫术,故成为目前研究热点。SONG等[57]回顾性分析432例乳腺癌患者,提取其第二期DCE-MRI组学特征,使用多因素logistic回归分析建立基于影像组学特征和临床因素(如组织学分级、多灶性等)的影像组学列线图模型,训练组及验证组AUC分别为0.907,0.874,特异度在80%以上。YU等[58]使用机器学习随机森林及支持向量机算法提取MRI特征,结合临床病理信息及腋窝淋巴结状态构建多组学标签,对预测腋窝淋巴结转移有较高价值。该研究为多中心研究,具有全面性及创新性,但存在不同中心图像异质性等问题。上述研究显示影像组学无创性预测腋窝淋巴结转移具备可行性,有助于改善患者预后,但结果仍不够精确,重复性仍存在问题,无法直接应用于临床。

3 小结与展望

       综上所述,MRI对乳腺的诊断不仅局限于定性及病灶范围的确定,通过对多模态MRI的形态学、功能学分析,术前无创性预测乳腺癌患者预后成为可能。近年来基于MRI的影像组学及AI在乳腺癌诊疗的应用价值已成为国内外学者研究的热点。但仍存在一些共性问题:(1)单中心、回顾性、小样本量研究较多,缺乏多中心的合作及外部验证;(2)多模态MRI使用不同设备、不同扫描参数、不同图像处理方法,图像的差异影响精确性和一致性;(3)特征提取软件缺乏及数据处理方法多样,使得基于实验室的AI模型可能不能直接应用于临床实践[59]。因此,未来该领域需要多中心、大样本、前瞻性及重复性的研究验证并发展现有研究成果。此外,使用大且多样化的数据集训练深度学习算法模型将有助于精准自动分割,提高一致性。将乳腺癌临床病理特征与影像组学模型结合构建综合模型可能也是一大研究方向。今后,随着计算机及影像技术的不断发展和应用,乳腺癌精准治疗将取得更大突破。

[1]
SUNG H, FERLAY J, SIEGEL R L, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249. DOI: 10.3322/caac.21660.
[2]
XIA C F, DONG X S, LI H, et al. Cancer statistics in China and United States, 2022: profiles, trends, and determinants[J]. Chin Med J (Engl), 2022, 135(5): 584-590. DOI: 10.1097/CM9.0000000000002108.
[3]
CURIGLIANO G, BURSTEIN H J, WINER E P, et al. De-escalating and escalating treatments for early-stage breast cancer: the St. Gallen International Expert Consensus Conference on the Primary Therapy of Early Breast Cancer 2017[J/OL]. Ann Oncol, 2019, 30(7): 1181 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/30624592/. DOI: 10.1093/annonc/mdy537.
[4]
GAO J J, SWAIN S M. Luminal A breast cancer and molecular assays: a review[J]. Oncologist, 2018, 23(5): 556-565. DOI: 10.1634/theoncologist.2017-0535.
[5]
CESCA M G, VIAN L, CRISTÓVÃO-FERREIRA S, et al. HER2-positive advanced breast cancer treatment in 2020[J/OL]. Cancer Treat Rev, 2020, 88: 102033 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/32534233/. DOI: 10.1016/j.ctrv.2020.102033.
[6]
KORDE L A, SOMERFIELD M R, CAREY L A, et al. Neoadjuvant chemotherapy, endocrine therapy, and targeted therapy for breast cancer: ASCO guideline[J]. J Clin Oncol, 2021, 39(13): 1485-1505. DOI: 10.1200/JCO.20.03399.
[7]
RAHMAT K, MUMIN N A, HAMID M T R, et al. MRI breast: current imaging trends, clinical applications, and future research directions[J]. Curr Med Imaging, 2022, 18(13): 1347-1361. DOI: 10.2174/1573405618666220415130131.
[8]
KIM S Y, CHO N. Breast magnetic resonance imaging for patients with newly diagnosed breast cancer: a review[J]. J Breast Cancer, 2022, 25(4): 263-277. DOI: 10.4048/jbc.2022.25.e35.
[9]
KAZAMA T, TAKAHARA T, HASHIMOTO J. Breast cancer subtypes and quantitative magnetic resonance imaging: a systemic review[J/OL]. Life (Basel), 2022, 12(4): 490 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/35454981. DOI: 10.3390/life12040490.
[10]
ZHANG J, WANG L, LIU H F. Imaging features derived from dynamic contrast-enhanced magnetic resonance imaging to differentiate malignant from benign breast lesions: a systematic review and meta-analysis[J]. J Comput Assist Tomogr, 2022, 46(3): 383-391. DOI: 10.1097/RCT.0000000000001289.
[11]
YUEN S, MONZAWA S, YANAI S, et al. The association between MRI findings and breast cancer subtypes: focused on the combination patterns on diffusion-weighted and T2-weighted images[J]. Breast Cancer, 2020, 27(5): 1029-1037. DOI: 10.1007/s12282-020-01105-z.
[12]
LI Q, XIAO Q, YANG M, et al. Histogram analysis of quantitative parameters from synthetic MRI: correlations with prognostic factors and molecular subtypes in invasive ductal breast cancer[J/OL]. Eur J Radiol, 2021, 139: 109697 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/33857828/. DOI: 10.1016/j.ejrad.2021.109697.
[13]
ÖZTÜRK V S, POLAT Y D, SOYDER A, et al. The relationship between MRI findings and molecular subtypes in women with breast cancer[J]. Curr Probl Diagn Radiol, 2020, 49(6): 417-421. DOI: 10.1067/j.cpradiol.2019.07.003.
[14]
SEYFETTIN A, DEDE I, HAKVERDI S, et al. MR imaging properties of breast cancer molecular subtypes[J]. Eur Rev Med Pharmacol Sci, 2022, 26(11): 3840-3848. DOI: 10.26355/eurrev_202206_28951.
[15]
BORIA F, TAGLIATI C, BALDASSARRE S, et al. Morphological MR features and quantitative ADC evaluation in invasive breast cancer: correlation with prognostic factors[J]. Clin Imaging, 2018, 50: 141-146. DOI: 10.1016/j.clinimag.2018.02.011.
[16]
CHEON H, KIM H J, KIM T H, et al. Invasive breast cancer: prognostic value of peritumoral edema identified at preoperative MR imaging[J]. Radiology, 2018, 287(1): 68-75. DOI: 10.1148/radiol.2017171157.
[17]
MBERU V, MCFARLANE J, MACASKILL E J, et al. A retrospective review of MRI features associated with metastasis-free survival in women with breast cancer: focusing on skin thickening and skin enhancement[J/OL]. Br J Radiol, 2021, 94(1128): 20210472 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/34591686/. DOI: 10.1259/bjr.20210472.
[18]
CHEN H, LI W, WAN C, et al. Correlation of dynamic contrast-enhanced MRI and diffusion-weighted MR imaging with prognostic factors and subtypes of breast cancers[J/OL]. Front Oncol, 2022, 12: 942943 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/35992872/. DOI: 10.3389/fonc.2022.942943.
[19]
GRIMM L J, ZHANG J, BAKER J A, et al. Relationships between MRI breast imaging-reporting and data system (BI-RADS) lexicon descriptors and breast cancer molecular subtypes: internal enhancement is associated with luminal B subtype[J]. Breast J, 2017, 23(5): 579-582. DOI: 10.1111/tbj.12799.
[20]
GALATI F, RIZZO V, MOFFA G, et al. Radiologic-pathologic correlation in breast cancer: do MRI biomarkers correlate with pathologic features and molecular subtypes?[J/OL]. Eur Radiol Exp, 2022, 6(1): 39 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/35934721/. DOI: 10.1186/s41747-022-00289-7.
[21]
ELMI A, CONANT E F, KOZLOV A, et al. Preoperative breast MR imaging in newly diagnosed breast cancer: comparison of outcomes based on mammographic modality, breast density and breast parenchymal enhancement[J]. Clin Imaging, 2021, 70: 18-24. DOI: 10.1016/j.clinimag.2020.10.021.
[22]
HU N, ZHAO J H, LI Y, et al. Breast cancer and background parenchymal enhancement at breast magnetic resonance imaging: a meta-analysis[J/OL]. BMC Med Imaging, 2021, 21(1): 32 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/33607959/. DOI: 10.1186/s12880-021-00566-8.
[23]
THOMPSON C M, MALLAWAARACHCHI I, DWIVEDI D K, et al. The association of background parenchymal enhancement at breast MRI with breast cancer: a systematic review and meta-analysis[J]. Radiology, 2019, 292(3): 552-561. DOI: 10.1148/radiol.2019182441.
[24]
MANN R M, ATHANASIOU A, BALTZER P A T, et al. Breast cancer screening in women with extremely dense breasts recommendations of the European Society of Breast Imaging (EUSOBI)[J]. Eur Radiol, 2022, 32(6): 4036-4045. DOI: 10.1007/s00330-022-08617-6.
[25]
ONISHI N, LI W, NEWITT D C, et al. Breast MRI during neoadjuvant chemotherapy: lack of background parenchymal enhancement suppression and inferior treatment response[J]. Radiology, 2021, 301(2): 295-308. DOI: 10.1148/radiol.2021203645.
[26]
LIM Y, KO E S, HAN B K, et al. Background parenchymal enhancement on breast MRI: association with recurrence-free survival in patients with newly diagnosed invasive breast cancer[J]. Breast Cancer Res Treat, 2017, 163(3): 573-586. DOI: 10.1007/s10549-017-4217-5.
[27]
GULLO R LO, DAIMIEL I, ROSSI SACCARELLI C, et al. MRI background parenchymal enhancement, fibroglandular tissue, and mammographic breast density in patients with invasive lobular breast cancer on adjuvant endocrine hormonal treatment: associations with survival[J/OL]. Breast Cancer Res, 2020, 22(1): 93 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/32819432/. DOI: 10.1186/s13058-020-01329-z.
[28]
RELLA R, BUFI E, BELLI P, et al. Association between contralateral background parenchymal enhancement on MRI and outcome in patients with unilateral invasive breast cancer receiving neoadjuvant chemotherapy[J]. Diagn Interv Imaging, 2022, 103(10): 486-494. DOI: 10.1016/j.diii.2022.04.004.
[29]
GOTO M, LE BIHAN D, YOSHIDA M, et al. Adding a model-free diffusion MRI marker to BI-RADS assessment improves specificity for diagnosing breast lesions[J]. Radiology, 2019, 292(1): 84-93. DOI: 10.1148/radiol.2019181780.
[30]
ZHANG L, TANG M, MIN Z Q, et al. Accuracy of combined dynamic contrast-enhanced magnetic resonance imaging and diffusion-weighted imaging for breast cancer detection: a meta-analysis[J]. Acta Radiol, 2016, 57(6): 651-660. DOI: 10.1177/0284185115597265.
[31]
KANG B J, LIPSON J A, PLANEY K R, et al. Rim sign in breast lesions on diffusion-weighted magnetic resonance imaging: diagnostic accuracy and clinical usefulness[J]. J Magn Reson Imaging, 2015, 41(3): 616-623. DOI: 10.1002/jmri.24617.
[32]
SUROV A, MEYER H J, WIENKE A. Can apparent diffusion coefficient (ADC) distinguish breast cancer from benign breast findings? A meta-analysis based on 13 847 lesions[J/OL]. BMC Cancer, 2019, 19(1): 955 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/31615463/. DOI: 10.1186/s12885-019-6201-4.
[33]
MARTINCICH L, DEANTONI V, BERTOTTO I, et al. Correlations between diffusion-weighted imaging and breast cancer biomarkers[J]. Eur Radiol, 2012, 22(7): 1519-1528. DOI: 10.1007/s00330-012-2403-8.
[34]
KHOULI R H EL, MACURA K J, KAMEL I R, et al. 3-T dynamic contrast-enhanced MRI of the breast: pharmacokinetic parameters versus conventional kinetic curve analysis[J]. AJR Am J Roentgenol, 2011, 197(6): 1498-1505. DOI: 10.2214/AJR.10.4665.
[35]
王倩, 刘万花, 王瑞, 等. 3.0T动态增强MRI定量参数、表观扩散系数与乳腺癌预后因子及分子分型的相关性[J]. 中国医学影像学杂志, 2019, 27(7): 517-521. DOI: 10.3969/j.issn.1005-5185.2019.07.009.
WANG Q, LIU W H, WANG R, et al. Correlation between quantitative parameters, apparent diffusion coefficient of 3.0T dynamic enhanced MRI and prognostic factors as well as molecular types of breast cancer[J]. Chin J Med Imaging, 2019, 27(7): 517-521. DOI: 10.3969/j.issn.1005-5185.2019.07.009.
[36]
LIU L, MEI N, YIN B, et al. Correlation of DCE-MRI perfusion parameters and molecular biology of breast infiltrating ductal carcinoma[J/OL]. Front Oncol, 2021, 11: 561735 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/34722229/. DOI: 10.3389/fonc.2021.561735.
[37]
肖文铉, 江一舟, 邵志敏. 人工智能在乳腺癌精准诊疗中的应用[J]. 实用肿瘤杂志, 2022, 37(2): 112-116. DOI: 10.13267/j.cnki.syzlzz.2022.019.
XIAO W X, JIANG Y Z, SHAO Z M. Application of artif cial intelligence in precision medicine for breast cancer[J]. J Pract Oncol, 2022, 37(2): 112-116. DOI: 10.13267/j.cnki.syzlzz.2022.019.
[38]
TAGLIAFICO A S, PIANA M, SCHENONE D, et al. Overview of radiomics in breast cancer diagnosis and prognostication[J]. Breast, 2020, 49: 74-80. DOI: 10.1016/j.breast.2019.10.018.
[39]
MEYER-BASE A, MORRA L, TAHMASSEBI A, et al. AI-enhanced diagnosis of challenging lesions in breast MRI: a methodology and application primer[J]. J Magn Reson Imaging, 2021, 54(3): 686-702. DOI: 10.1002/jmri.27332.
[40]
BITENCOURT A, DAIMIEL NARANJO I, GULLO R LO, et al. AI-enhanced breast imaging: where are we and where are we heading?[J/OL]. Eur J Radiol, 2021, 142: 109882 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/34392105/. DOI: 10.1016/j.ejrad.2021.109882.
[41]
MEYER-BÄSE A, MORRA L, MEYER-BÄSE U, et al. Current status and future perspectives of artificial intelligence in magnetic resonance breast imaging[J/OL]. Contrast Media Mol Imaging, 2020, 2020: 6805710 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/32934610/. DOI: 10.1155/2020/6805710.
[42]
ZHANG Q, PENG Y S, LIU W, et al. Radiomics based on multimodal MRI for the differential diagnosis of benign and malignant breast lesions[J]. J Magn Reson Imaging, 2020, 52(2): 596-607. DOI: 10.1002/jmri.27098.
[43]
LIU Z S, FENG B, LI C L, et al. Preoperative prediction of lymphovascular invasion in invasive breast cancer with dynamic contrast-enhanced-MRI-based radiomics[J]. J Magn Reson Imaging, 2019, 50(3): 847-857. DOI: 10.1002/jmri.26688.
[44]
DALMIŞ M U, VREEMANN S, KOOI T, et al. Fully automated detection of breast cancer in screening MRI using convolutional neural networks[J/OL]. J Med Imaging (Bellingham), 2018, 5(1): 014502 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/29340287/. DOI: 10.1117/1.JMI.5.1.014502.
[45]
ZHOU J, LUO L Y, DOU Q, et al. Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in MR images[J]. J Magn Reson Imaging, 2019, 50(4): 1144-1151. DOI: 10.1002/jmri.26721.
[46]
XU H, LIU J K, CHEN Z, et al. Intratumoral and peritumoral radiomics based on dynamic contrast-enhanced MRI for preoperative prediction of intraductal component in invasive breast cancer[J]. Eur Radiol, 2022, 32(7): 4845-4856. DOI: 10.1007/s00330-022-08539-3.
[47]
ZHANG Y, ZHU Y F, ZHANG K, et al. Invasive ductal breast cancer: preoperative predict Ki-67 index based on radiomics of ADC maps[J]. Radiol Med, 2020, 125(2): 109-116. DOI: 10.1007/s11547-019-01100-1.
[48]
YIN H L, BAI L T, JIA H H, et al. Noninvasive assessment of breast cancer molecular subtypes on multiparametric MRI using convolutional neural network with transfer learning[J]. Thorac Cancer, 2022, 13(22): 3183-3191. DOI: 10.1111/1759-7714.14673.
[49]
HA R, MUTASA S, KARCICH J, et al. Predicting breast cancer molecular subtype with MRI dataset utilizing convolutional neural network algorithm[J]. J Digit Imaging, 2019, 32(2): 276-282. DOI: 10.1007/s10278-019-00179-2.
[50]
SHENG W Y, XIA S L, WANG Y R, et al. Invasive ductal breast cancer molecular subtype prediction by MRI radiomic and clinical features based on machine learning[J/OL]. Front Oncol, 2022, 12: 964605 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/36172153/. DOI: 10.3389/fonc.2022.964605.
[51]
FISHER C S. Neoadjuvant chemotherapy for breast cancer: the ultimate "spy"[J]. Ann Surg Oncol, 2022, 29(11): 6508-6510. DOI: 10.1245/s10434-022-12153-4.
[52]
LI C C, LU N, HE Z F, et al. A noninvasive tool based on magnetic resonance imaging radiomics for the preoperative prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer[J]. Ann Surg Oncol, 2022, 29(12): 7685-7693. DOI: 10.1245/s10434-022-12034-w.
[53]
EUN N L, KANG D, SON E J, et al. Texture analysis with 3.0-T MRI for association of response to neoadjuvant chemotherapy in breast cancer[J]. Radiology, 2020, 294(1): 31-41. DOI: 10.1148/radiol.2019182718.
[54]
MA M M, GAN L Y, LIU Y H, et al. Radiomics features based on automatic segmented MRI images: prognostic biomarkers for triple-negative breast cancer treated with neoadjuvant chemotherapy[J/OL]. Eur J Radiol, 2022, 146: 110095 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/34890936/. DOI: 10.1016/j.ejrad.2021.110095.
[55]
KOH J, LEE E, HAN K, et al. Three-dimensional radiomics of triple-negative breast cancer: prediction of systemic recurrence[J/OL]. Sci Rep, 2020, 10(1): 2976 [2022-11-26]. https://pubmed.ncbi.nlm.nih.gov/32076078/. DOI: 10.1038/s41598-020-59923-2.
[56]
CHITALIA R D, ROWLAND J, MCDONALD E S, et al. Imaging phenotypes of breast cancer heterogeneity in preoperative breast dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) scans predict 10-year recurrence[J]. Clin Cancer Res, 2020, 26(4): 862-869. DOI: 10.1158/1078-0432.CCR-18-4067.
[57]
SONG D L, YANG F, ZHANG Y J, et al. Dynamic contrast-enhanced MRI radiomics nomogram for predicting axillary lymph node metastasis in breast cancer[J/OL]. Cancer Imaging, 2022, 22(1): 17 [2022-22-26]. https://pubmed.ncbi.nlm.nih.gov/35379339/. DOI: 10.1186/s40644-022-00450-w.
[58]
YU Y F, HE Z F, OUYANG J, et al. Magnetic resonance imaging radiomics predicts preoperative axillary lymph node metastasis to support surgical decisions and is associated with tumor microenvironment in invasive breast cancer: a machine learning, multicenter study[J/OL]. EBioMedicine, 2021, 69: 103460 [2022-22-26]. https://pubmed.ncbi.nlm.nih.gov/34233259/. DOI: 10.1016/j.ebiom.2021.103460.
[59]
JONES M A, ISLAM W, FAIZ R, et al. Applying artificial intelligence technology to assist with breast cancer diagnosis and prognosis prediction[J/OL]. Front Oncol, 2022, 12: 980793 [2022-22-26]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9471147. DOI: 10.3389/fonc.2022.980793.

上一篇 MRI影像组学在乳腺癌诊疗中的研究进展
下一篇 磁共振弹性成像在肝脏占位性病变中的研究进展
  
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