分享:
分享到微信朋友圈
X
综述
术前MRI在预测乳腺癌预后中的应用及展望
边小倩 杜思瑶 张立娜

Cite this article as: Bian XQ, Du SY, Zhang LN. Application and prospect of preoperative MRI in predicting the prognosis of breast cancer[J]. Chin J Magn Reson Imaging, 2022, 13(6): 147-150.本文引用格式:边小倩, 杜思瑶, 张立娜. 术前MRI在预测乳腺癌预后中的应用及展望[J]. 磁共振成像, 2022, 13(6): 147-150. DOI:10.12015/issn.1674-8034.2022.06.031.


[摘要] 乳腺癌是女性发病率排名首位且致死率排名前五的肿瘤,其预后因素复杂。近年来,MRI积极探索与乳腺癌预后相关的影像学标志物,包括形态学、血流动力学、功能成像、影像组学等诸多参数。研究证实了肿瘤大小及边缘、非肿块样强化、边缘强化、瘤周水肿、背景强化是预后相关的形态学参数;血流动力学时间—强度曲线(time-intensity curve,TIC)及定量、半定量参数在不同程度上与预后相关;尽管还存在一些争议,弥散加权成像(diffusion-weighted imaging,DWI)及其衍生技术在乳腺癌预后中显示出巨大的预测潜力;基于MRI的影像组学进一步揭示了更多与预后相关的高维度参数,计算机引导的人工智能正在兴起。本文就术前MRI预测乳腺癌预后的研究进展进行综述,为本领域下一步的相关研究提供参考。
[Abstract] Breast cancer is the most common cancer in women and the top 5 in terms of mortality, and its prognostic factors are complex. In recent years, MRI has actively explored imaging markers related to breast cancer prognosis, including morphology, hemodynamics, functional imaging, radiomics and many other parameters. The study confirmed that tumor size and edge, non-mass-like enhancement, rim enhancement, peritumoral edema, and background enhancement are morphological parameters related to prognosis; hemodynamic time-intensity curves (TIC) and quantitative and semi-quantitative parameters are associated with prognosis to varying degrees; although there are still some controversies, diffusion-weighted imaging (DWI) and its derived techniques have shown great potential in prognosis prediction; MRI-based radiomics has further revealed more high-dimensional parameters related to prognosis, and computer-guided artificial intelligence is emerging. This article reviews the research progress of preoperative MRI in predicting the prognosis of breast cancer, and provides a reference for the next related research in this field.
[关键词] 乳腺癌;磁共振成像;动态增强磁共振成像;弥散加权成像;影像组学;人工智能;预后;综述
[Keywords] breast cancer;magnetic resonance imaging;dynamic contrast-enhanced magnetic resonance imaging;diffusion‐weighted imaging;radiomics;artificial intelligence;prognosis;review

边小倩    杜思瑶    张立娜 *  

中国医科大学附属第一医院放射科,沈阳 110001

张立娜,E-mail:zhanglnda@163.com

作者利益冲突声明:全体作者均声明无利益冲突。


基金项目: 科技部重大慢性非传染性疾病防控研究重点专项 2017YFC1309100
收稿日期:2022-01-05
接受日期:2022-06-06
中图分类号:R445.2  R737.9 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2022.06.031
本文引用格式:边小倩, 杜思瑶, 张立娜. 术前MRI在预测乳腺癌预后中的应用及展望[J]. 磁共振成像, 2022, 13(6): 147-150. DOI:10.12015/issn.1674-8034.2022.06.031

       2022年美国癌症协会的统计数据显示,乳腺癌是女性发病率排名首位且致死率排名前5的肿瘤[1],其前5年的复发风险为10.4%,后5年的复发风险为4.5%[2]。目前,临床已证实腋窝淋巴结转移、脉管侵犯、激素受体表达、组织学分级、分子亚型等是乳腺癌复发风险及预后的重要影响因素[3, 4, 5, 6]。但以上指标只能通过活检或术后病理获得,存在侵入性和不可重复性。MRI具有无创、易获得和可重复的特点,近年来已被广泛应用于乳腺癌预后相关研究,其部分影像特征及参数包括形态学、血流动力学、功能成像、影像组学等被证实为预测乳腺癌患者预后的成像标志物[7, 8],为患者制定更优化的治疗策略提供重要支持。

1 基于MRI的形态学特征

       MRI形态学特征是反映病灶基本属性的常见指标。其中肿瘤大小、边缘、非肿块样强化、边缘强化、瘤周水肿、背景强化等是目前已知的预后相关影像参数。

       临床已经证实大于2 cm的肿瘤更易复发,基于MRI测量的肿瘤大小与术后病理大小一致性良好,复发组明显大于非复发组(平均大小3.9 cm vs. 2.7 cm,P=0.02)[9]。肿块不规则伴毛刺边缘更多见于三阴性亚型、存在脉管侵犯的侵袭性病例中,往往预后较差[10, 11, 12]。相对肿块,非肿块样强化表现出更差的预后[13, 14],这可能与非肿块样病变难以早期诊断且阳性切缘有关[15]。边缘强化是肿瘤外围对比于中心的增强[13],被认为是由肿瘤内微血管密度降低引起的,而非肿瘤周围微血管密度的增加[16]。在三阴性亚组的多变量分析中,发现仅边缘增强与复发之间存在显著相关性,结果显示边缘增强的肿瘤复发率是非边缘增强的14倍(HR=14.019;95% CI:1.773~110.864)[13,17]。瘤周水肿的存在是乳腺癌复发的独立危险因素(HR=2.77,P=0.022)[18],在T2加权成像(T2-weighted images,T2WI)中,不同的水肿类型(包括瘤周水肿、胸前水肿、皮下水肿和弥漫性水肿)与肿瘤侵袭性呈现不同程度的相关性[19],而三阴性乳腺癌出现瘤周水肿的几率更大[20],说明瘤周水肿与高侵袭性肿瘤相关,是瘤周脉管侵犯和炎症的间接指标。在目前的指南中,瘤周水肿并不是乳腺癌的关键预后因素,大数据高质量的相关研究可能会推动指南的未来更新。

       除了原发灶及瘤周的特征,乳腺背景实质强化(background parenchymal enhancement,BPE)对乳腺癌预后也具有一定的预测价值。同侧乳腺的BPE可能因乳腺癌的存在而受到血管化增加的影响,导致BPE的MRI评估假性抬高[21, 22],因此已提出对侧BPE的MRI评估作为改进乳腺癌决策的工具。根据乳腺成像报告和数据系统的BPE模式分类,在绝经后乳腺癌患者中,高BPE等级表现出更差的无复发生存结果(HR=3.086,P=0.003)[23]。但BPE的评估受主观因素、年龄和激素水平影响较大,因此,需要更多影响因素控制完善的相关数据以及量化技术的开发进一步证实。

       综上,传统MRI形态学特征受临床医生的主观判断及临床经验的影响较大,且大多为定性特征,因此对乳腺癌预后的价值相对受限,通过人工智能,如深度学习(deep learning,DL)全自动病灶分割,提高其可重复性,将传统的形态学指标进行定量化将是乳腺MRI的主流和趋势。

2 基于MRI的功能学参数

       MRI功能学参数是乳腺影像研究的热点之一。目前常用的功能学指标包括动态增强、弥散及衍生序列所产生的相关参数。尽管还存在着一些争议,系列研究已经证实了功能学参数在预测乳腺癌预后中的重要价值。

2.1 血流动力学

       动态对比增强磁共振成像(dynamic contrast-enhanced magnetic resonance imaging,DCE-MRI)是评估血管功能最常用的方法,包括基于时间—强度曲线(time-intensity curve,TIC)的半定量参数,如:达峰时间(time-to-peak enhancement,Tpeak)、最大强化斜率(maximum rise slope,Slopemax)等和基于药代动力学计算模型的定量参数,如:转运常数(transfer constant,Ktrans)、回流常数(rate constant,Kep)和细胞外血管外体积分数(extracellular extravascular volume fraction,Ve)等。

       基于TIC计算得到的半定量参数中,不同研究的结果各异。相对预后良好的患者,肿瘤内较快的初始增强[24]、更高的峰值增强[25],TIC的Slopemax[26]、更短的Tpeak[27]以及较多的流出成分[28]是较差生存结局的重要预测参数。Niukkanen等[29]利用3D分割技术分析得到瘤内、瘤周的信号增强比(signal enhancement ratio,SER)与肿瘤大小、组织学分级、Ki-67表达相关,进一步证实了瘤周区域TIC相关的半定量参数与乳腺癌不良预后的关联性。

       研究表明,Ktrans、Kep与雌激素受体(estrogen receptor,ER)、孕激素受体(progesterone receptor,PR)表达呈负相关;Ktrans与组织学分级、人表皮生长因子受体2 (the human epidermal growth factor receptor 2,HER-2)表达、淋巴结转移及Ki-67表达呈正相关;Kep与HER-2、Ki-67表达呈正相关[30, 31],这可能与新生毛细血管增加导致对比剂流出及回流的速度加快有关。基于DCE-MRI定量参数的直方图分析结果显示,高侵袭性乳腺癌(三阴性乳腺癌)、高Ki-67表达(>20%)和高核级癌症(2级或3级)表现出更高的Ve值变异系数和偏度,且具有显著差异性[32],而这些癌症被认为具有高侵袭及高增殖性,往往预后不良。

       虽然TIC形态与复杂的定量灌注参数显著相关[33],但定量参数显然具有更好的客观性和可比性。基于药代动力学模型定量参数的精准计算需要依赖高时间分辨率的DCE-MRI序列扫描,并受不同设备、参数设置和所选计算模型的影响。但随着乳腺癌定量研究的不断深入,基于DCE-MRI影像定量参数的乳腺癌侵袭性或预后仍然是未来研究的重要趋势。随着磁共振波谱成像(magnetic resonance spectroscopic imaging,MRSI)、化学交换饱和转移成像技术(chemical exchange saturation transfer,CEST)、血氧水平依赖(blood oxygenation level-dependent,BOLD)、弛豫时间等更多定量扫描序列在乳腺领域的逐步应用,功能学定量参数不断丰富,我们对多参数定量模型的开发充满期待。

2.2 弥散加权成像及衍生技术

       弥散加权成像(diffusion-weighted imaging,DWI)已在临床实践中常规使用,先进的DWI技术如体素内不相干运动(intravoxel incoherent motion,IVIM)、弥散张量成像(diffusion tensor imaging,DTI)、弥散峰度成像(diffusion kurtosis imaging,DKI)等在乳腺癌预后评估中也表现出潜在价值[34, 35, 36]

       传统的DWI因出现时间早、扫描序列和后处理相对简便易行,已被广泛用于乳腺癌预后的研究中。肿瘤内表观弥散系数(apparent diffusion coefficient,ADC)差异值越大(>0.698×10-3 mm2/s),发生远处转移的风险越高(HR=4.5,P<0.001)[37]。经过7.2年的平均随访时间,瘤周与瘤内ADC比率较高的患者相对比率较低的患者具有更差的总生存率[38, 39],这可能与肿瘤内异质性及肿瘤侵袭性诱导的瘤周液体渗出相关[40]。IVIM实现体素内水分子弥散和灌注成分的分离,反映组织细胞的数量和血管分布[41]。但它与乳腺癌预后之间的关系存在争议[34,42, 43]。有研究[42, 43]表明,其参数伪扩散系数与激素受体、Ki-67表达、组织学分级之间具有相关性,Song等[34]却认为IVIM参数与预后因素之间并无显著关联,原因可能是定量参数的计算及拟合b值的选取存在差异。DTI获得体素内的各向异性程度及弥散情况,以表征乳腺癌微观结构的差异。先前的研究认为较大的肿瘤(>2 cm)、高组织学分级(3级)、腋窝淋巴结转移与较低的平均弥散率(mean diffusivity,MD)显著相关,而淋巴结状态与各向异性分数(fractional anisotropy,FA)差异并无统计学意义[35,44];ER阴性和Ki-67高表达与FA值呈显著负相关[45],这与肿瘤细胞密度增加、微观结构破坏,使得水分子弥散的幅度和方向降低相关[45]。DKI用于研究b值超过1000 s/mm2时水分子的非高斯弥散。组织学分级和Ki-67表达被认为与平均峰度(mean kurtosis,MK)呈正相关,与MD呈负相关,但与肿瘤大小无明显相关性[46, 47]。高侵袭性肿瘤的MK值升高,反映了组织微观结构较复杂,偏离高斯分布较大,它增加了单体素中的细胞密度,进而影响水分子弥散[48]。但也有关于组织学分级的阴性研究[36],其结果的差异可能是由b值选取、扫描参数不同等多种因素造成。

       DWI及其衍生技术在乳腺癌预后方面显示出巨大的预测潜力,但DWI序列在乳腺扫描中也存在一些不足:信噪比和空间分辨率较低、空间失真较大等,会导致图像损失一些重要信息。目前DWI序列仍作为DCE-MRI的补充序列。DTI和IVIM技术更详细地表征了组织微观结构,似乎更具优越性,但是其扫描和计算模型的复杂性更容易产生不一致的结果[34, 35]。近年来,弥散相关序列扫描技术不断进步,有望弥补现有序列的缺陷,并将DWI及衍生技术与人工智能相结合,以挖掘更多有价值的预后信息。

3 基于MRI的影像组学和人工智能

       影像组学和人工智能是近几年来在各个领域科学研究的重点和热点,同样也应用在乳腺癌预后预测的相关研究中,并已经取得了很多乐观、积极的结果。

       影像组学特征已被证明是预测乳腺癌预后的独立生物标志物[49]。一项基于肿瘤异质性纹理分析的研究发现,T2WI的高熵值(≥6.013,HR=9.84)和动态增强T1加权减影图像的低熵值(<5.057,HR=4.55)与较差的无复发生存率显著相关[50]。虽然纹理分析可以在一定程度上测量整个肿瘤内的异质性,但这种测量依赖于肿瘤内异质性的良好混合,忽略了肿瘤内的区域表型变异[51]。而亚区分割技术更加关注肿瘤内的灌注异质性,相对于临床、影像组学等其他4种模型,利用组学特征构建联合多亚区的复发风险评估模型表现出最好的预测性能(C-Index=0.760)[52],因此,灌注异质性的量化是预测乳腺癌预后的一种潜在方法。然而,Yoon等[53]从ADC图中提取的纹理参数均未显示出对疾病无进展生存期的显著预测价值。造成结果不同的原因,可能是不同的研究在参数选择、量化标准、亚型组成及计算方法等方面存在差异。

       目前,针对乳腺癌预后进行预测的机器学习(machine learning,ML)模型主要使用MRI组学特征融合临床特征进行训练,并能达到满意的预测效能[3]。当使用包含多参数MRI特征的8种ML算法(如:随机森林、决策树、K-最邻近等)对乳腺癌的重要预后因素Ki-67和组织学分级进行评估时,6个模型显示出对Ki-67的等效性能,受试者工作特征(receiver operating characteristic,ROC)曲线下面积(area under the curve,AUC)为0.70;对组织学分级的预测,贝叶斯算法表现最佳(AUC=0.79)[54]。ML允许提取内容丰富的成像信息,且能够量化人眼无法察觉的组织之间的差异,因此基于MRI影像组学的ML在乳腺癌预后方面有望取得更多进展。

       卷积神经网络(convolutional neural network,CNN)是目前最流行的用于图像分析的DL结构[55]。近年来,乳腺癌预后预测相关的DL方法不断深入:Liu等[56]在DCE-MRI对比前后图像生成的肿瘤掩模基础上改进了3D-CNN框架,认为由肿瘤掩模引导的3D-CNN预测模型比掩模引导的完整图像或仅掩模体素的图像具有更高的5年无复发生存预测精度。相比传统的3D-CNN方法[57],由掩模引导的预测模型能将注意力集中在与掩模不同的乳腺肿瘤内未知区域,更强调了与肿瘤预后相关的图像特征[56]。4D-CNN (随时间变化的3D-CNN)模型增加了时间维度,补充了2D和3D所缺乏的时空背景,在融合临床信息后,对腋窝淋巴结转移的预测效能及稳定性均有较大提升,也进一步证实了肿瘤和瘤周信息对预测乳腺癌转移的重要价值[58]。DL已经在乳腺癌预后预测中显示出重要价值,其无创、可预知的特点有望将患者利益实现最大化。

       由于缺乏明确的生物学原理来解释组学特征和预后之间的联系,影像组学的临床应用价值受到限制。同时,由于图像采集方法的不一致和计算方法选择的差异,影像组学研究结果的难以重复性也是其在各领域普遍处于瓶颈期的原因之一。尽管ML和DL对乳腺肿瘤的分割、诊断及预后等方面产生了深远影响,但其预测结果的低可解释性同样阻碍了他们的临床应用[59, 60],当前影像组学研究和人工智能相关研究正在致力于准确性、科学性和泛化能力,探索模型对预测结果的解释能力。另外在给予人工智能更大自主权的同时,对不能提供理由的黑箱决策尝试进行监督,尊重公众对人工智能决策的看法[60]

4 结论和展望

       综上所述,随着MRI技术的不断开发与应用,术前对乳腺癌患者的预后进行无创预测的设想成为可能,并及时为预后不佳的患者提供恰当的辅助治疗,对提高患者生存率具有重要意义。除了已建立的乳腺癌预后相关的MRI形态学、功能学特征外,基于MRI的影像组学进一步揭示了更多与预后相关的高维度参数,计算机引导的人工智能正在兴起。然而,由于乳腺癌的异质性和影像表现的复杂性,未来在乳腺癌筛查、诊断、治疗和预后方面的研究还有很长的路要走,大队列、前瞻性、多中心研究值得进一步探索。

[1]
Siegel RL, Miller KD, Fuchs HE, et al. Cancer statistics, 2022[J]. CA Cancer J Clin, 2022, 72(1): 7-33. DOI: 10.3322/caac.21708.
[2]
Colleoni M, Sun ZX, Price KN, et al. Annual hazard rates of recurrence for breast cancer during 24 years of follow-up: results from the international breast cancer study group trials I to V[J]. J Clin Oncol, 2016, 34(9): 927-935. DOI: 10.1200/JCO.2015.62.3504.
[3]
Yu YF, Tan YJ, Xie CM, et al. Development and validation of a preoperative magnetic resonance imaging radiomics-based signature to predict axillary lymph node metastasis and disease-free survival in patients with early-stage breast cancer[J]. JAMA Netw Open, 2020, 3(12): e2028086. DOI: 10.1001/jamanetworkopen.2020.28086.
[4]
Choi WJ, Cha JH, Kim HH, et al. Long-term survival outcomes of primary breast cancer in women with or without preoperative magnetic resonance imaging: a matched cohort study[J]. Clin Oncol (R Coll Radiol), 2017, 29(10): 653-661. DOI: 10.1016/j.clon.2017.06.015.
[5]
Hayashi Y, Satake H, Ishigaki S, et al. Kinetic volume analysis on dynamic contrast-enhanced MRI of triple-negative breast cancer: associations with survival outcomes[J]. Br J Radiol, 2020, 93(1106): 20190712. DOI: 10.1259/bjr.20190712.
[6]
Pickles MD, Lowry M, Gibbs P. Pretreatment prognostic value of dynamic contrast-enhanced magnetic resonance imaging vascular, texture, shape, and size parameters compared with traditional survival indicators obtained from locally advanced breast cancer patients[J]. Invest Radiol, 2016, 51(3): 177-185. DOI: 10.1097/RLI.0000000000000222.
[7]
Drukker K, Li H, Antropova N, et al. Most-enhancing tumor volume by MRI radiomics predicts recurrence-free survival "early on" in neoadjuvant treatment of breast cancer[J]. Cancer Imaging, 2018, 18(1): 12. DOI: 10.1186/s40644-018-0145-9.
[8]
Nam SY, Ko ES, Lim Y, et al. Preoperative dynamic breast magnetic resonance imaging kinetic features using computer-aided diagnosis: association with survival outcome and tumor aggressiveness in patients with invasive breast cancer[J]. PLoS One, 2018, 13(4): e0195756. DOI: 10.1371/journal.pone.0195756.
[9]
Lee J, Kim SH, Kang BJ, et al. Imaging characteristics of young age breast cancer (YABC) focusing on pathologic correlation and disease recurrence[J]. Sci Rep, 2021, 11(1): 20205. DOI: 10.1038/s41598-021-99600-6.
[10]
Ryu MJ, Kim YS, Lee SE. Association between imaging features using the BI-RADS and tumor subtype in patients with invasive breast cancer[J]. Curr Med Imaging, 2022, 18(6): 648-657. DOI: 10.2174/1573405617666210520155157.
[11]
Choi BB. Dynamic contrast enhanced-MRI and diffusion-weighted image as predictors of lymphovascular invasion in node-negative invasive breast cancer[J]. World J Surg Oncol, 2021, 19(1): 76. DOI: 10.1186/s12957-021-02189-3.
[12]
Kim HJ, Choi WJ, Kim HH, et al. Association between Oncotype DX recurrence score and dynamic contrast-enhanced MRI features in patients with estrogen receptor-positive HER2-negative invasive breast cancer[J]. Clin Imaging, 2021, 75: 131-137. DOI: 10.1016/j.clinimag.2021.01.021.
[13]
Schmitz AM, Loo CE, Wesseling J, et al. Association between rim enhancement of breast cancer on dynamic contrast-enhanced MRI and patient outcome: impact of subtype[J]. Breast Cancer Res Treat, 2014, 148(3): 541-551. DOI: 10.1007/s10549-014-3170-9.
[14]
Bitencourt AGV, Eugênio DSG, Souza JA, et al. Prognostic significance of preoperative MRI findings in young patients with breast cancer[J]. Sci Rep, 2019, 9(1): 3106. DOI: 10.1038/s41598-019-39629-w.
[15]
Koh J, Park AY, Ko KH, et al. Can enhancement types on preoperative MRI reflect prognostic factors and surgical outcomes in invasive breast cancer?[J]. Eur Radiol, 2019, 29(12): 7000-7008. DOI: 10.1007/s00330-019-06236-2.
[16]
Teifke A, Behr O, Schmidt M, et al. Dynamic MR imaging of breast lesions: correlation with microvessel distribution pattern and histologic characteristics of prognosis[J]. Radiology, 2006, 239(2): 351-360. DOI: 10.1148/radiol.2392050205.
[17]
Song SE, Shin SU, Moon HG, et al. MR imaging features associated with distant metastasis-free survival of patients with invasive breast cancer: a case-control study[J]. Breast Cancer Res Treat, 2017, 162(3): 559-569. DOI: 10.1007/s10549-017-4143-6.
[18]
Cheon H, Kim HJ, Kim TH, 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.
[19]
Santucci D, Faiella E, Cordelli E, et al. The impact of tumor edema on T2-weighted 3T-MRI invasive breast cancer histological characterization: a pilot radiomics study[J]. Cancers, 2021, 13(18): 4635. DOI: 10.3390/cancers13184635.
[20]
Costantini M, Belli P, Distefano D, et al. Magnetic resonance imaging features in triple-negative breast cancer: comparison with luminal and HER2-overexpressing tumors[J]. Clin Breast Cancer, 2012, 12(5): 331-339. DOI: 10.1016/j.clbc.2012.07.002.
[21]
Jones EF, Sinha SP, Newitt DC, et al. MRI enhancement in stromal tissue surrounding breast tumors: association with recurrence free survival following neoadjuvant chemotherapy[J]. PLoS One, 2013, 8(5): e61969. DOI: 10.1371/journal.pone.0061969.
[22]
Luo J, Johnston BS, Kitsch AE, et al. Ductal carcinoma in situ: quantitative preoperative breast MR imaging features associated with recurrence after treatment[J]. Radiology, 2017, 285(3): 788-797. DOI: 10.1148/radiol.2017170587.
[23]
Lim Y, Ko ES, Han BK, 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.
[24]
Dietzel M, Zoubi R, Vag T, et al. Association between survival in patients with primary invasive breast cancer and computer aided MRI[J]. J Magn Reson Imaging, 2013, 37(1): 146-155. DOI: 10.1002/jmri.23812.
[25]
Kim JY, Kim JJ, Hwangbo L, et al. Kinetic heterogeneity of breast cancer determined using computer-aided diagnosis of preoperative MRI scans: relationship to distant metastasis-free survival[J]. Radiology, 2020, 295(3): 517-526. DOI: 10.1148/radiol.2020192039.
[26]
Tuncbilek N, Tokatli F, Altaner S, et al. Prognostic value DCE-MRI parameters in predicting factor disease free survival and overall survival for breast cancer patients[J]. Eur J Radiol, 2012, 81(5): 863-867. DOI: 10.1016/j.ejrad.2011.02.021.
[27]
van der Velden BHM, Elias SG, Bismeijer T, et al. Complementary value of contralateral parenchymal enhancement on DCE-MRI to prognostic models and molecular assays in high-risk ER+/HER2- breast cancer[J]. Clin Cancer Res, 2017, 23(21): 6505-6515. DOI: 10.1158/1078-0432.CCR-17-0176.
[28]
Yamaguchi A, Honda M, Ishiguro H, et al. Kinetic information from dynamic contrast-enhanced MRI enables prediction of residual cancer burden and prognosis in triple-negative breast cancer: a retrospective study[J]. Sci Rep, 2021, 11(1): 10112. DOI: 10.1038/s41598-021-89380-4.
[29]
Niukkanen A, Okuma H, Sudah M, et al. Quantitative three-dimensional assessment of the pharmacokinetic parameters of intra- and peri-tumoural tissues on breast dynamic contrast-enhanced magnetic resonance imaging[J]. J Digit Imaging, 2021, 34(5): 1110-1119. DOI: 10.1007/s10278-021-00509-3.
[30]
Liu L, Mei N, Yin B, et al. Correlation of DCE-MRI perfusion parameters and molecular biology of breast infiltrating ductal carcinoma[J]. Front Oncol, 2021, 11: 561735. DOI: 10.3389/fonc.2021.561735.
[31]
Liu F, Wang M, Li HG. Role of perfusion parameters on DCE-MRI and ADC values on DWMRI for invasive ductal carcinoma at 3.0 Tesla[J]. World J Surg Oncol, 2018, 16(1): 239. DOI: 10.1186/s12957-018-1538-8.
[32]
Nagasaka K, Satake H, Ishigaki S, et al. Histogram analysis of quantitative pharmacokinetic parameters on DCE-MRI: correlations with prognostic factors and molecular subtypes in breast cancer[J]. Breast Cancer, 2019, 26(1): 113-124. DOI: 10.1007/s12282-018-0899-8.
[33]
Yi B, Kang DK, Yoon D, et al. Is there any correlation between model-based perfusion parameters and model-free parameters of time-signal intensity curve on dynamic contrast enhanced MRI in breast cancer patients?[J]. Eur Radiol, 2014, 24(5): 1089-1096. DOI: 10.1007/s00330-014-3100-6.
[34]
Song SE, Cho KR, Seo BK, et al. Intravoxel incoherent motion diffusion-weighted MRI of invasive breast cancer: correlation with prognostic factors and kinetic features acquired with computer-aided diagnosis[J]. J Magn Reson Imaging, 2019, 49(1): 118-130. DOI: 10.1002/jmri.26221.
[35]
Kim JY, Kim JJ, Kim S, et al. Diffusion tensor magnetic resonance imaging of breast cancer: associations between diffusion metrics and histological prognostic factors[J]. Eur Radiol, 2018, 28(8): 3185-3193. DOI: 10.1007/s00330-018-5429-8.
[36]
Yang ZL, Li Y, Zhan CA, et al. Evaluation of suspicious breast lesions with diffusion kurtosis MR imaging and connection with prognostic factors[J]. Eur J Radiol, 2021, 145: 110014. DOI: 10.1016/j.ejrad.2021.110014.
[37]
Kim JY, Kim JJ, Hwangbo L, et al. Diffusion-weighted imaging of invasive breast cancer: relationship to distant metastasis-free survival[J]. Radiology, 2019, 291(2): 300-307. DOI: 10.1148/radiol.2019181706.
[38]
Okuma H, Sudah M, Kettunen T, et al. Peritumor to tumor apparent diffusion coefficient ratio is associated with biologically more aggressive breast cancer features and correlates with the prognostication tools[J]. PLoS One, 2020, 15(6): e0235278. DOI: 10.1371/journal.pone.0235278.
[39]
Kettunen T, Okuma H, Auvinen P, et al. Peritumoral ADC values in breast cancer: region of interest selection, associations with hyaluronan intensity, and prognostic significance[J]. Eur Radiol, 2020, 30(1): 38-46. DOI: 10.1007/s00330-019-06361-y.
[40]
Baltzer PAT, Yang F, Dietzel M, et al. Sensitivity and specificity of unilateral edema on T2w-TSE sequences in MR-Mammography considering 974 histologically verified lesions[J]. Breast J, 2010, 16(3): 233-239. DOI: 10.1111/j.1524-4741.2010.00915.x.
[41]
Iima M, Reynaud O, Tsurugizawa T, et al. Characterization of glioma microcirculation and tissue features using intravoxel incoherent motion magnetic resonance imaging in a rat brain model[J]. Invest Radiol, 2014, 49(7): 485-490. DOI: 10.1097/RLI.0000000000000040.
[42]
Lee YJ, Kim SH, Kang BJ, et al. Intravoxel incoherent motion (IVIM)-derived parameters in diffusion-weighted MRI: associations with prognostic factors in invasive ductal carcinoma[J]. J Magn Reson Imaging, 2017, 45(5): 1394-1406. DOI: 10.1002/jmri.25514.
[43]
Kim Y, Ko K, Kim D, et al. Intravoxel incoherent motion diffusion-weighted MR imaging of breast cancer: association with histopathological features and subtypes[J]. Br J Radiol, 2016, 89(1063): 20160140. DOI: 10.1259/bjr.20160140.
[44]
Ozal ST, Inci E, Gemici AA, et al. Can 3.0 Tesla diffusion tensor Imaging parameters be prognostic indicators in breast cancer?[J]. Clin Imaging, 2018, 51: 240-247. DOI: 10.1016/j.clinimag.2018.03.022.
[45]
Onaygil C, Kaya, Ugurlu MU, et al. Diagnostic performance of diffusion tensor imaging parameters in breast cancer and correlation with the prognostic factors[J]. J Magn Reson Imaging, 2017, 45(3): 660-672. DOI: 10.1002/jmri.25481.
[46]
Huang Y, Lin Y, Hu W, et al. Diffusion kurtosis at 3.0T as an in vivo imaging marker for breast cancer characterization: correlation with prognostic factors[J]. J Magn Reson Imaging, 2019, 49(3): 845-856. DOI: 10.1002/jmri.26249.
[47]
Sun K, Chen XS, Chai WM, et al. Breast cancer: diffusion kurtosis MR imaging-diagnostic accuracy and correlation with clinical-pathologic factors[J]. Radiology, 2015, 277(1): 46-55. DOI: 10.1148/radiol.15141625.
[48]
Suo ST, Cheng F, Cao MQ, et al. Multiparametric diffusion-weighted imaging in breast lesions: association with pathologic diagnosis and prognostic factors[J]. J Magn Reson Imaging, 2017, 46(3): 740-750. DOI: 10.1002/jmri.25612.
[49]
Chitalia RD, Rowland J, McDonald ES, 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.
[50]
Kim JH, Ko ES, Lim Y, et al. Breast cancer heterogeneity: MR imaging texture analysis and survival outcomes[J]. Radiology, 2017, 282(3): 665-675. DOI: 10.1148/radiol.2016160261.
[51]
Gatenby RA, Grove O, Gillies RJ. Quantitative imaging in cancer evolution and ecology[J]. Radiology, 2013, 269(1): 8-15. DOI: 10.1148/radiol.13122697.
[52]
Cho HH, Kim H, Nam SY, et al. Measurement of perfusion heterogeneity within tumor habitats on magnetic resonance imaging and its association with prognosis in breast cancer patients[J]. Cancers (Basel), 2022, 14(8): 1858. DOI: 10.3390/cancers14081858.
[53]
Yoon HJ, Kim Y, Chung J, et al. Predicting neo-adjuvant chemotherapy response and progression-free survival of locally advanced breast cancer using textural features of intratumoral heterogeneity on F-18 FDG PET/CT and diffusion-weighted MR imaging[J]. Breast J, 2019, 25(3): 373-380. DOI: 10.1111/tbj.13032.
[54]
Song SE, Cho KR, Cho Y, et al. Machine learning with multiparametric breast MRI for prediction of Ki-67 and histologic grade in early-stage luminal breast cancer[J]. Eur Radiol, 2022, 32(2): 853-863. DOI: 10.1007/s00330-021-08127-x.
[55]
Rawat W, Wang ZH. Deep convolutional neural networks for image classification: a comprehensive review[J]. Neural Comput, 2017, 29(9): 2352-2449. DOI: 10.1162/NECO_a_00990.
[56]
Liu GB, Mitra D, Jones EF, et al. Mask-guided convolutional neural network for breast tumor prognostic outcome prediction on 3D DCE-MR images[J]. J Digit Imaging, 2021, 34(3): 630-636. DOI: 10.1007/s10278-021-00449-y.
[57]
Li J, Fan M, Zhang J, et al. Discriminating between benign and malignant breast tumors using 3D convolutional neural network in dynamic contrast enhanced-MR images[C]//Medical Imaging 2017: Imaging Informatics for Healthcare, Research, and Applications. Orlando, Florida, USA: SPIE, 2017, 10138: 44-51. DOI: 10.1117/12.2254716.
[58]
Nguyen S, Polat D, Karbasi P, et al. Preoperative prediction of lymph node metastasis from clinical DCE MRI of the primary breast tumor using a 4D CNN[J]. Med Image Comput Comput Assist Interv, 2020, 12262: 326-334. DOI: 10.1007/978-3-030-59713-9_32.
[59]
Banegas-Luna AJ, Peña-García J, Iftene A, et al. Towards the interpretability of machine learning predictions for medical applications targeting personalised therapies: a cancer case survey[J]. Int J Mol Sci, 2021, 22(9): 4394. DOI: 10.3390/ijms22094394.
[60]
Ou WC, Polat D, Dogan BE. Deep learning in breast radiology: current progress and future directions[J]. Eur Radiol, 2021, 31(7): 4872-4885. DOI: 10.1007/s00330-020-07640-9.

上一篇 创伤性脑损伤后血管源性脑水肿的MRI研究进展
下一篇 肝胆特异性对比剂MRI在肝脏弥漫性疾病的应用进展
  
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