综述
基于纵向时间影像的深度学习在乳腺癌新辅助化疗疗效预测中的研究进展
磁共振成像, 2023,14(3) : 175-178,183. DOI: 10.12015/issn.1674-8034.2023.03.032
摘要

新辅助化疗(neoadjuvant chemotherapy, NACT)已被确立为局部进展期乳腺癌的标准治疗方案。影像学检查在乳腺癌患者治疗过程中具有重要的疗效监测作用。研究已证明基于纵向时间影像动态评估及预测乳腺癌NACT疗效的有效性及必要性,较单时序具有显著优势。本文将分析单一时间影像预测乳腺癌NACT疗效的局限性,并对基于超声和动态对比增强磁共振成像(dynamic contrast enhanced magnetic resonance imaging, DCE-MRI)纵向时间影像预测乳腺癌NACT疗效的研究现状、面临问题及应用前景进行综述。本文认为今后的研究可通过多模态、多序列纵向时间影像来提高预测NACT疗效的准确性,还可以联合多中心、大样本数据进一步提升模型泛化性能。本文旨在为今后纵向时间影像预测乳腺癌NACT疗效提供思路。

引用本文: 黄瑶, 王晓霞, 蒋富杰, 等.  基于纵向时间影像的深度学习在乳腺癌新辅助化疗疗效预测中的研究进展 [J] . 磁共振成像, 2023, 14(3) : 175-178,183. DOI: 10.12015/issn.1674-8034.2023.03.032.
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

乳腺癌是全球最常见的癌症之一,居于女性恶性肿瘤发病率首位,其发病率和死亡率呈逐年上升的趋势[1, 2, 3]。新辅助化疗(neoadjuvant chemotherapy, NACT)具有检测肿瘤对药物的敏感性、缩小术前肿块体积、提高保乳治疗率的优势[4, 5],美国国立综合癌症网络(National Comprehensive Cancer Network, NCCN)乳腺癌临床实践指南和欧洲肿瘤内科学会(European Society for Medical Oncology, ESMO)指南已将新辅助化疗作为局部进展期乳腺癌早期治疗的标准方案[6, 7]。已有研究表明,乳腺癌患者经NACT后若能达到病理完全缓解(pathologic complete response, pCR),意味着可获得更长的总生存期和无病生存期[8, 9]。但因患者个体差异及肿瘤异质性,接受NACT后患者的pCR率为10%~62%,对于无效者反而会带来治疗相关成本及药物毒副作用[10]。因此,在NACT早期阶段进行个体疗效预测并筛选潜在获益患者对于临床诊治具有重要意义。

影像学检查在乳腺癌NACT过程中具有举足轻重的作用,其结果判读将直接指导临床治疗决策[11, 12, 13]。随着成像技术的快速发展以及人工智能方法在医学图像的广泛应用,基于影像图像评估及预测乳腺癌NACT疗效取得了较好效果,尤其是基于纵向时间影像数据[14, 15]。纵向时间影像指的是乳腺癌患者接受NACT前后进行多次影像学检查,以用于动态评估和预测NACT疗效。本文系统地梳理了纵向时间影像预测乳腺癌NACT疗效的研究现状及面临的问题,并对未来应用前景进行了展望。本文旨在为今后纵向时间影像预测乳腺癌NACT疗效的研究提供思路。

1 影像学检查是评估乳腺癌NACT疗效的重要手段

手术病理检测为NACT后是否达到pCR的金标准,但其具有滞后性等缺点,无法在NACT早期指导临床治疗方案决策[16]。影像学检查贯穿乳腺癌患者诊疗全过程,已成为评估NACT疗效的重要方法[17]。目前常用的影像学检查包括:乳腺钼靶、超声及MRI等[18, 19, 20]。临床主要采用实体瘤疗效评价标准(response evaluation criteria in solid tumors, RECIST)1.1评价体系,将疗效分为完全缓解、部分缓解、疾病进展以及疾病稳定四类[21, 22]。然而,凭借肉眼捕捉的宏观形态学改变来评价疗效具有许多局限性,难以满足临床精准预测疗效的需求[23]

随着人工智能在医学图像中的发展,将深度学习(deep learning, DL)与影像图像相结合的处理方式备受青睐。DL可将原始影像数据直接输入模型,减少人工操作,降低主观因素对数据的影响[24],其主要通过卷积神经网络(convolutional neural network, CNN)自动提取影像潜在细微特征以完成分类和预测任务。越来越多的研究表明,DL在乳腺癌NACT疗效的预测中具有极大的发展潜力[25, 26]

2 单一时间影像预测乳腺癌NACT疗效具有局限性

目前,大多数研究是基于单一时间影像来预测NACT疗效[27, 28]。如AGHAEI等[29]提出了一种计算机辅助检测方法,从68例乳腺癌患者接受NACT前的MRI影像中提取动力学特征,通过建立的人工神经网络实现了对NACT疗效的预测。然而,已有研究发现,肿瘤内部药物代谢的动力学特征、纹理特征等均会在治疗过程产生变化[30, 31, 32, 33]。因此,单个时间点的影像学检查不足以表征肿瘤在治疗过程中产生变化的特有属性,难以准确评估乳腺癌NACT疗效,而NACT前后的肿瘤异质性变化是预测疗效的重要信息。基于此,深入开展纵向时间影像特征的动态变化研究,以便早期精准预测NACT反应势在必行。

3 纵向时间影像预测乳腺癌NACT疗效研究进展
3.1 影像学检查在乳腺癌NACT疗效预测中的应用

获取纵向时间影像需要乳腺癌患者在接受NACT前后多次进行影像学检查。虽然有学者指出,乳腺钼靶可用于监测NACT前后乳腺组织成分的定量变化[34],但目前缺少相关研究来进一步验证乳腺钼靶的纵向时间影像在预测乳腺癌NACT疗效中的有效性。需要注意的是,使用具有辐射的乳腺钼靶并不适合进行多次检查。相比之下,超声设备普及广泛、检查费用较低,是临床中最便捷的NACT疗效评估方式。另外,MRI被认为是监测乳腺癌NACT疗效最准确的医学影像学检查方式。但现有基于纵向时间MRI预测NACT疗效的研究较少,且主要是基于动态对比增强磁共振成像(dynamic contrast enhanced MRI, DCE-MRI)进行的[35]。研究已证明,与乳腺钼靶和超声相比,DCE-MRI在评估残余肿瘤大小体积和预测NACT疗效等方面的性能更佳[36, 37]

3.2 基于超声的乳腺癌NACT疗效预测

BYRA等[38]构建了2种CNN模型(先验模型和孪生网络)用于预测乳腺癌患者早期NACT反应,先验模型的输入仅为治疗前的超声图像,而孪生网络的输入为治疗前和NACT后的超声图像,两种模型的AUC分别为0.797和0.847。由此可见,纵向时间影像的使用有助于提升NACT疗效预测准确性。LIU等[39]继而在传统孪生网络的基础上,构建了多任务孪生网络(siamese multi-task network, SMTN)用于乳腺癌pCR预测,SMTN从治疗前和NACT后一个或两个周期的纵向超声影像中提取动态变化信息并结合患者病理信息,该混合模型预测pCR的AUC高达0.986。该研究结果表明SMTN预测性能较好,可为临床医生早期调整非pCR患者的治疗方案提供帮助。近年来,将DL与传统影像组学结合用于癌症疗效预测等也取得了较好成效。JIANG等[40]建立了一套基于DL的影像组学(DL radiomics, DLR)的pCR预测模型。先利用深度CNN(DenseNet-201)和开源工具包(Pyradiomics)分别对乳腺癌NACT前后的超声数据提取DL特征和影像组学特征,继而进行特征筛选并结合临床病理预后因素,通过多因素逻辑回归构建模型,在外部验证集中AUC和模型的分类准确度分别为0.94和0.84,预测效果优于临床模型和两名经验丰富的影像科医生。另外,在MANI等[41]和HUSSAIN等[42]的研究中也都证实了NACT前后的影像组学特征在NACT疗效预测中具有积极作用。

与其他采用单个DLR模型研究不同的是,GU等[43]使用了两个DLR模型(DLR-2和DLR-4)来识别患者对NACT的反应。DLR-2中输入患者治疗前和治疗后2个周期后的超声影像,DLR-4中输入患者治疗前和治疗4个周期后的超声影像,用于识别患者对NACT的反应,两个DLR模型的AUC分别为0.979和0.981。该研究还使用梯度加权类激活映射对DLR模型进行可视化解释,结果显示,DLR-2更加关注超声影像中的肿瘤内部特征,DLR-4则侧重于肿瘤周围组织特征。这项研究表明使用DLR模型能够有效地识别NACT过程中患者的反应,为临床治疗提供决策支持。

然而,超声在预测乳腺癌的NACT疗效方面具有固有的一些局限性:(1)超声切线位图像受操作医生主观影响大[44];(2)超声信噪比相对低,图像是2D而不是3D;(3)超声对治疗后的肿块边界显示不清楚[45]。综上所述,超声的纵向时间影像对NACT前后肿瘤及其周围组织动态变化反映不够灵敏从而降低了NACT疗效预测的准确性。

3.3 基于DCE-MRI的乳腺癌NACT疗效预测

麻省理工的研究团队首次通过使用乳腺癌患者NACT前后的DCE-MRI中肿瘤最大径及其上下的切片作为CNN的输入来预测治疗反应,结果显示乳腺癌分子亚型与DL模型相结合可提高AUC[46]。DRUKKER等[47]考虑到NACT过程中可能会抑制血管生成,导致乳腺癌的血流动力学特征发生变化,限制肿瘤供血并阻碍其增殖。该研究通过长短期记忆网络验证了乳腺癌血流动力学特征可用于NACT疗效预测。此外,还有研究比较了单一时间影像与纵向时间影像在预测NACT疗效的性能,QU等[48]选取NACT前后的DCE-MRI六个增强序列输入神经网络并与单一使用治疗前图像进行对比,结果也表明,使用纵向DCE-MRI图像的预测效果更好,使用NACT前、后以及组合图像的AUC分别为0.553、0.968和0.970。

上述研究旨在通过DL揭示纵向时间影像在预测NACT疗效中的重要性和有效性,未深究肿瘤异质性与NACT疗效之间的关系。FAN等[49]选取DCE-MRI中增强峰值图像,通过纹理特征评估了肿瘤异质性在治疗前后的变化,发现乳腺癌患者接受NACT后,肿瘤异质性的降低与对NACT反应良好相关。该研究进一步验证了GU等[43]基于超声提出的肿瘤异质性在治疗中的变化与NACT疗效有关这一观点。

DCE-MRI不仅可以反映病变的形态特征,确定残余肿瘤的大小和边界,还可以评估组织功能、微环境特征和代谢的变化[50, 51]。基于DCE-MRI预测乳腺癌NACT疗效的研究更加多元化,不仅可以将治疗前后的DCE-MRI图像直接作为模型输入,还能将治疗前后的血流动力学特征等作为预测因子。然而,由于DCE-MRI检查成本较高且部分患者具有MRI检查禁忌证,造成DCE-MRI在乳腺癌NACT过程中难以采集完整的纵向数据;此外,不同医院及不同型号MRI设备,造成扫描图像参数难以标准化,因此在开展基于DCE-MRI的纵向时间影像预测乳腺癌NACT疗效研究时,大样本、多中心的标准化数据库不足成为一大难题。

4 基于纵向时间影像预测乳腺癌NACT疗效问题与挑战

以上研究表明,肿瘤对NACT有反应时,肿瘤及其微环境将发生变化,这些变化通常不容易被影像医生肉眼观察到,而使用DL模型结合纵向时间影像将有助于挖掘肿瘤内部微观特征变化,从而更全面地监测肿瘤动态变化,为乳腺癌患者是否继续接受NACT提供临床决策参考。目前,该研究的发展亦面临一些问题与挑战:(1)现有研究绝大多数为回顾性研究,缺少前瞻性研究验证纵向时间影像在预测乳腺癌NACT疗效中的作用;纵向时间影像数据样本量少,国内目前暂无公开的乳腺癌纵向时间影像数据集,而国外公开的数据集病例较少且大多需要重新勾画感兴趣区(region of interest, ROI)。(2)纵向时间影像数据采集较为困难,乳腺癌患者接受NACT过程中的纵向时间影像学检查的最佳时间节点缺乏统一的规范和标准。乳腺影像学方法多种多样,MRI常规扫描也包含多个序列,目前大部分研究主要聚焦于单一模态或单一序列,忽视了多模态影像及多序列MRI的互补作用[52, 53]。因此,多模态、多序列纵向研究的组合方案可能是未来的研究方向之一。(3)使用DL结合纵向时间影像需要海量样本,而临床数据通常是稀有的、不平衡的,难以满足DL模型的训练需求。现有的研究模型泛化性能普遍较差且DL的“黑匣子”问题依然存在,无法为医生提供可视化过程,造成模型的可解释性差,因此导致其在临床的推广应用较为困难[54, 55]

5 小结及展望

综上所述,基于纵向时间影像预测乳腺癌NACT疗效在临床治疗决策中具有较大的发展潜力。虽然该研究的发展面临一些问题,但随着人工智能的发展和医学水平的不断进步,基于纵向时间影像预测乳腺癌NACT疗效的方法可能有助于提高临床决策的准确性,具有广阔的发展前景。

未来可以开展研究评估不同时间节点对于预测NACT疗效的准确性,或通过临床实践经验来确定纵向时间影像学检查的最佳时间节点。还可以通过多模态、多序列的纵向时间影像来提高预测NACT疗效的准确性。此外,由于目前大多研究样本量较少且缺乏对外部数据集的验证。因此,今后的研究还可以联合多中心、大样本数据进一步提升模型的预测性能和泛化性能。

本文引用格式:

黄瑶, 王晓霞, 蒋富杰, 等. 基于纵向时间影像的深度学习在乳腺癌新辅助化疗疗效预测中的研究进展[J]. 磁共振成像, 2023, 14(3): 175-178, 183.

Cite this article as:

HUANG Y, WANG X X, JIANG F J, et al. The research progress in predicting the efficacy of neoadjuvant chemotherapy for breast cancer according to longitudinal images-based deep learning[J]. Chin J Magn Reson Imaging, 2023, 14(3): 175-178, 183.

ACKNOWLEDGMENTS

2021 SKY Imaging Research Fund of the Chinese International Medical Foundation (No. Z-2014-07-2101); Chongqing Natural Science Foundation (No. cstc2021jcyj-msxmX0387, CSTB2022NSCQ-MSX1158); Medical Research Project of Chongqing Municipal Health Commission (No. 2022WSJK027).

利益冲突
利益冲突:

全体作者均声明无利益冲突。

参考文献References
[1]
SUNGH, FERLAYJ, SIEGELR 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]
WANGX X, TANY C, LIUD H, et al. Chemotherapy-associated steatohepatitis was concomitant with epicardial adipose tissue volume increasing in breast cancer patients who received neoadjuvant chemotherapy[J]. Eur Radiol, 2022, 32(7): 4898-4908. DOI: 10.1007/s00330-022-08581-1.
[3]
ARNOLDM, MORGANE, RUMGAYH, et al. Current and future burden of breast cancer: global statistics for 2020 and 2040[J]. Breast, 2022, 66: 15-23. DOI: 10.1016/j.breast.2022.08.010.
[4]
LIUZ Y, LIZ L, QUJ R, et al. Radiomics of multiparametric MRI for pretreatment prediction of pathologic complete response to neoadjuvant chemotherapy in breast cancer: a multicenter study[J]. Clin Cancer Res, 2019, 25(12): 3538-3547. DOI: 10.1158/1078-0432.CCR-18-3190.
[5]
DERKSM G M, VAN DE VELDEC J H. Neoadjuvant chemotherapy in breast cancer: more than just downsizing[J]. Lancet Oncol, 2018, 19(1): 2-3. DOI: 10.1016/S1470-2045(17)30914-2.
[6]
GRADISHARW J, ANDERSONB O, ABRAHAMJ, et al. Breast cancer, version 3.2020, NCCN clinical practice guidelines in oncology[J]. J Natl Compr Canc Netw, 2020, 18(4): 452-478. DOI: 10.6004/jnccn.2020.0016.
[7]
DENDULURIN, SOMERFIELDM R, CHAVEZ-MACGREGORM, et al. Selection of optimal adjuvant chemotherapy and targeted therapy for early breast cancer: ASCO guideline update[J]. J Clin Oncol, 2021, 39(6): 685-693. DOI: 10.1200/JCO.20.02510.
[8]
VON MINCKWITZG, BLOHMERJ U, COSTAS D, et al. Response-guided neoadjuvant chemotherapy for breast cancer[J]. J Clin Oncol, 2013, 31(29): 3623-3630. DOI: 10.1200/JCO.2012.45.0940.
[9]
YANGY, CHENX, FANZ, et al. Pathological response-guided postoperative treatment strategy for positive HR/negative HER2 and positive lymphnodes breast cancer patients with non-responsive pathological results to neoadjuvant chemotherapy[J/OL]. Cancer Res, 2022, 82(4) [2022-11-13]. https://doi.org/10.1158/1538-7445.SABCS21-P2-12-06. DOI: 10.1158/1538-7445.SABCS21-P2-12-06.
[10]
GOUTSOULIAKK, VEERARAGHAVANJ, SETHUNATHV, et al. Towards personalized treatment for early stage HER2-positive breast cancer[J]. Nat Rev Clin Oncol, 2020, 17(4): 233-250. DOI: 10.1038/s41571-019-0299-9.
[11]
LIUY G, WUM X, TANW Y, et al. Efficacy evaluation of neoadjuvant chemotherapy in breast cancer by MRI[J/OL]. Contrast Media Mol Imaging, 2022, 2022: 4542288 [2022-11-13]. https://doi.org/10.1155/2022/4542288. DOI: 10.1155/2022/4542288.
[12]
YANGM, LIUH, DAIQ L, et al. Treatment response prediction using ultrasound-based pre-, post-early, and delta radiomics in neoadjuvant chemotherapy in breast cancer[J/OL]. Front Oncol, 2022, 12: 748008 [2022-11-13]. https://doi.org/10.3389/fonc.2022.748008. DOI: 10.3389/fonc.2022.748008.
[13]
GAOY M, REIGB, HEACOCKL, et al. Magnetic resonance imaging in screening of breast cancer[J]. Radiol Clin North Am, 2021, 59(1): 85-98. DOI: 10.1016/j.rcl.2020.09.004.
[14]
单慧明, 周靖宇, 谢婷婷, . MRI影像学特征预测乳腺癌新辅助化疗疗效的可行性[J]. 中国医学影像学杂志, 2019, 27(12): 905-909, 913. DOI: 10.3969/j.issn.1005-5185.2019.12.006.
SHANH M, ZHOUJ Y, XIET T, et al. Feasibility of MRI features in predicting efficacy of neoadjuvant chemotherapy for breast cancer[J]. Chin J Med Imaging, 2019, 27(12): 905-909, 913. DOI: 10.3969/j.issn.1005-5185.2019.12.006.
[15]
YOSHIDAK, KAWASHIMAH, KANNONT, et al. Prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer using radiomics of pretreatment dynamic contrast-enhanced MRI[J/OL]. Magn Reson Imaging, 2022, 92: 19-25 [2022-11-13]. https://doi.org/10.1016/j.mri.2022.05.018. DOI: 10.1016/j.mri.2022.05.018.
[16]
邵志敏, 吴炅, 江泽飞, . 中国乳腺癌新辅助治疗专家共识(2022年版)[J]. 中国癌症杂志, 2022, 32(1): 80-89. DOI: 10.19401/j.cnki.1007-3639.2022.01.011.
SHAOZ M, WUJ, JIANGZ f, et al. Expert consensus on neoadjuvant treatment of breast cancer in China(2021 edition)[J]. China Oncol, 2022, 32(1): 80-89. DOI: 10.19401/j.cnki.1007-3639.2022.01.011.
[17]
赵智锦, 常才, 李佳伟. 影像学技术对三阴性乳腺癌临床诊断及预后分析价值的研究进展[J]. 中华超声影像学杂志, 2022, 31(2): 181-185. DOI: 10.3760/cma.j.cn131148-20210630-00451.
ZHAOZ J, CHANGC, LIJ W. Value of imaging technology in clinical diagnosis and prognosis prediction for triple negative breast cancer: a state-of-art review[J]. Chin J Ultrason, 2022, 31(2): 181-185. DOI: 10.3760/cma.j.cn131148-20210630-00451.
[18]
STEINHOF-RADWAŃSKAK, GRAŻYŃSKAA, LOREKA, et al. Contrast-enhanced spectral mammography assessment of patients treated with neoadjuvant chemotherapy for breast cancer[J]. Curr Oncol, 2021, 28(5): 3448-3462. DOI: 10.3390/curroncol28050298.
[19]
HANX, JINS Y, YANGH J, et al. Application of conventional ultrasonography combined with contrast-enhanced ultrasonography in the axillary lymph nodes and evaluation of the efficacy of neoadjuvant chemotherapy in breast cancer patients[J/OL]. Br J Radiol, 2021, 94(1125): 20210520 [2022-11-13]. https://www.birpublications.org/doi/10.1259/bjr.20210520. DOI: 10.1259/bjr.20210520.
[20]
YINX X, JINY B, GAOM Y, et al. Artificial intelligence in breast MRI radiogenomics: towards accurate prediction of neoadjuvant chemotherapy responses[J]. Curr Med Imaging, 2021, 17(4): 452-458. DOI: 10.2174/1573405616666200825161921.
[21]
陈志庚, 李响, 沙琳. 机器学习基于MRI预测乳腺癌对新辅助化疗反应的研究进展[J]. 磁共振成像, 2021, 12(12): 102-104. DOI: 10.12015/issn.1674-8034.2021.12.024.
CHENZ G, LIX, SHAL. Research progress of machine learning for predicting breast cancer response to neoadjuvant chemotherapy based on MRI[J]. Chin J Magn Reson Imaging, 2021, 12(12): 102-104. DOI: 10.12015/issn.1674-8034.2021.12.024.
[22]
WANGH, MAOX Y. Evaluation of the efficacy of neoadjuvant chemotherapy for breast cancer[J]. Drug Des Devel Ther, 2020, 14: 2423-2433. DOI: 10.2147/DDDT.S253961.
[23]
LEES H, JANGM J, KIMS M, et al. Factors affecting breast cancer detectability on digital breast tomosynthesis and two-dimensional digital mammography in patients with dense breasts[J]. Korean J Radiol, 2019, 20(1): 58-68. DOI: 10.3348/kjr.2018.0012.
[24]
ZHANGX Y, LIH, WANGC Y, et al. Evaluating the accuracy of breast cancer and molecular subtype diagnosis by ultrasound image deep learning model[J/OL]. Front Oncol, 2021, 11: 623506 [2022-11-13]. https://doi.org/10.3389/fonc.2021.623506. DOI: 10.3389/fonc.2021.623506.
[25]
REZAEIJOS M, GHORVEIM, MOFIDB. Predicting breast cancer response to neoadjuvant chemotherapy using ensemble deep transfer learning based on CT images[J]. J Xray Sci Technol, 2021, 29(5): 835-850. DOI: 10.3233/XST-210910.
[26]
CHOIJ H, KIMH A, KIMW, et al. Early prediction of neoadjuvant chemotherapy response for advanced breast cancer using PET/MRI image deep learning[J/OL]. Sci Rep, 2020, 10(1): 21149 [2022-11-13]. https://www.nature.com/articles/s41598-020-77875-5. DOI: 10.1038/s41598-020-77875-5.
[27]
CHOUDHERYS, GOMEZ-CARDONAD, FAVAZZAC P, et al. MRI radiomics for assessment of molecular subtype, pathological complete response, and residual cancer burden in breast cancer patients treated with neoadjuvant chemotherapy[J/OL]. Acad Radiol, 2022, 29: S145-S154 [2022-11-13]. https://doi.org/10.1016/j.acra.2020.10.020. DOI: 10.1016/j.acra.2020.10.020.
[28]
BRAMANN M, ETESAMIM, PRASANNAP, et al. Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI[J/OL]. Breast Cancer Res, 2017, 19(1): 57 [2022-11-13]. https://breast-cancer-research.biomedcentral.com/articles/10.1186/s13058-017-0846-1. DOI: 10.1186/s13058-017-0862-1.
[29]
AGHAEIF, TANM, HOLLINGSWORTHA B, et al. Computer-aided breast MR image feature analysis for prediction of tumor response to chemotherapy[J]. Med Phys, 2015, 42(11): 6520-6528. DOI: 10.1118/1.4933198.
[30]
DOGANB E, YUANQ, BASSETTR, et al. Comparing the performances of magnetic resonance imaging size vs pharmacokinetic parameters to predict response to neoadjuvant chemotherapy and survival in patients with breast cancer[J]. Curr Probl Diagn Radiol, 2019, 48(3): 235-240. DOI: 10.1067/j.cpradiol.2018.03.003.
[31]
TUDORICAA, OHK Y, CHUIS Y, et al. Early prediction and evaluation of breast cancer response to neoadjuvant chemotherapy using quantitative DCE-MRI[J]. Transl Oncol, 2016, 9(1): 8-17. DOI: 10.1016/j.tranon.2015.11.016.
[32]
HENDERSONS, PURDIEC, MICHIEC, et al. Interim heterogeneity changes measured using entropy texture features on T2-weighted MRI at 3.0 T are associated with pathological response to neoadjuvant chemotherapy in primary breast cancer[J]. Eur Radiol, 2017, 27(11): 4602-4611. DOI: 10.1007/s00330-017-4850-8.
[33]
EUNN L, KANGD, SONE 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.
[34]
JANGH J, KIMH J, CHAEY S, et al. Effect of neoadjuvant chemotherapy on breast tissue composition: a longitudinal mammographic study with automated volumetric measurement[J]. Eur Radiol, 2020, 30(9): 4785-4794. DOI: 10.1007/s00330-020-06830-9.
[35]
DIALANIV, CHADASHVILIT, SLANETZP J. Role of imaging in neoadjuvant therapy for breast cancer[J]. Ann Surg Oncol, 2015, 22(5): 1416-1424. DOI: 10.1245/s10434-015-4403-9.
[36]
MARINOVICHM L, MACASKILLP, IRWIGL, et al. Agreement between MRI and pathologic breast tumor size after neoadjuvant chemotherapy, and comparison with alternative tests: individual patient data meta-analysis[J]. BMC Cancer, 2015, 15: 662. DOI: 10.1186/s12885-015-1664-4.
[37]
FOWLERA M, MANKOFFD A, JOEB N. Imaging neoadjuvant therapy response in breast cancer[J]. Radiology, 2017, 285(2): 358-375. DOI: 10.1148/radiol.2017170180.
[38]
BYRAM, DOBRUCH-SOBCZAKK, KLIMONDAZ, et al. Early prediction of response to neoadjuvant chemotherapy in breast cancer sonography using Siamese convolutional neural networks[J]. IEEE J Biomed Health Inform, 2021, 25(3): 797-805. DOI: 10.1109/JBHI.2020.3008040.
[39]
LIUY, WANGY, WANGY X,. Early prediction of treatment response to neoadjuvant chemotherapy based on longitudinal ultrasound images of HER2-positive breast cancer patients by Siamese multi-task network: a multicentre, retrospective cohort study[J/OL]. eClinicalMedicine, 2022, 52: 101562 [2022-11-13]. https://doi.org/10.1016/j.eclinm.2022.101562. DOI: 10.1016/j.eclinm.2022.101562.
[40]
JIANGM, LIC L, LUOX M, et al. Ultrasound-based deep learning radiomics in the assessment of pathological complete response to neoadjuvant chemotherapy in locally advanced breast cancer[J]. Eur J Cancer, 2021, 147: 95-105. DOI: 10.1016/j.ejca.2021.01.028.
[41]
MANIS, CHENY K, LIX, et al. Machine learning for predicting the response of breast cancer to neoadjuvant chemotherapy[J]. J Am Med Inform Assoc, 2013, 20(4): 688-695. DOI: 10.1136/amiajnl-2012-001332.
[42]
HUSSAINL, HUANGP, NGUYENT, et al. Machine learning classification of texture features of MRI breast tumor and peri-tumor of combined pre- and early treatment predicts pathologic complete response[J/OL]. BioMed Eng OnLine, 2021, 20(1): 63 [2022-11-13]. https://biomedical-engineering-online.biomedcentral.com/articles/10.1186/s12938-021-00899-z. DOI: 10.1186/s12938-021-00899-z.
[43]
GUJ H, TONGT, HEC, et al. Deep learning radiomics of ultrasonography can predict response to neoadjuvant chemotherapy in breast cancer at an early stage of treatment: a prospective study[J]. Eur Radiol, 2022, 32(3): 2099-2109. DOI: 10.1007/s00330-021-08293-y.
[44]
车树楠, 薛梅, 李静, . 基于MRI多期增强影像组学联合临床影像特征模型术前预测乳腺癌Ki-67表达状态[J]. 中华放射学杂志, 2022, 56(9): 967-975. DOI: 10.3760/cma.j.cn112149-20211224-01137.
CHES N, XUEM, LIJ, et al. Preoperative prediction of Ki-67 expression status in breast cancer based on dynamic contrast enhanced MRI radiomics combined with clinical imaging features model[J]. Chin J Radiol, 2022, 56(9): 967-975. DOI: 10.3760/cma.j.cn112149-20211224-01137.
[45]
GRANZIERR Y, VANNIJNATTEN T J A, WOODRUFFH C, et al. Exploring breast cancer response prediction to neoadjuvant systemic therapy using MRI-based radiomics: a systematic review[J/OL]. Eur J Radiol, 2019, 121: 108736 [2022-11-13]. https://doi.org/10.1016/j.ejrad.2019.108736. DOI: 10.1016/j.ejrad.2019.108736.
[46]
RAVICHANDRANK, BRAMANN, JANOWCZYKA, et al. A deep learning classifier for prediction of pathological complete response to neoadjuvant chemotherapy from baseline breast DCE-MRI[C/OL]//SPIE Medical Imaging 2018: Computer-Aided Diagnosis, Houston, 2018 [2022-11-13]. https://doi.org/10.1117/12.2294056. DOI: 10.1117/12.2294056.
[47]
DRUKKERK, EDWARDSA, PAPAIOANNOUJ, et al. Deep learning predicts breast cancer recurrence in analysis of consecutive MRIs acquired during the course of neoadjuvant chemotherapy[C/OL]//SPIE Medical Imaging 2020: Computer-Aided Diagnosis, Houston, 2020 [2022-11-13]. https://doi.org/10.1117/12.2549044. DOI: 10.1117/12.2549044.
[48]
QUY H, ZHUH T, CAOK, et al. Prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer using a deep learning (DL) method[J]. Thorac Cancer, 2020, 11(3): 651-658. DOI: 10.1111/1759-7714.13309.
[49]
FANM, CHENH, YOUC, et al. Radiomics of tumor heterogeneity in longitudinal dynamic contrast-enhanced magnetic resonance imaging for predicting response to neoadjuvant chemotherapy in breast cancer[J/OL]. Front Mol Biosci, 2021, 8: 622219 [2022-11-13]. https://doi.org/10.3389/fmolb.2021.622219. DOI: 10.3389/fmolb.2021.622219.
[50]
CHITALIAR D, KONTOSD. Role of texture analysis in breast MRI as a cancer biomarker: a review[J]. J Magn Reson Imaging, 2019, 49(4): 927-938. DOI: 10.1002/jmri.26556.
[51]
LAFORGIA D, VESTITOA, LASCIARREAM, et al. Response predictivity to neoadjuvant therapies in breast cancer: a qualitative analysis of background parenchymal enhancement in DCE-MRI[J/OL]. J Pers Med, 2021, 11(4): 256 [2022-11-13]. https://doi.org/10.3390/jpm11040256. DOI: 10.3390/jpm11040256.
[52]
KEUNEJ D, JEFFED B, SCHOOTMANM, et al. Accuracy of ultrasonography and mammography in predicting pathologic response after neoadjuvant chemotherapy for breast cancer[J]. Am J Surg, 2010, 199(4): 477-484. DOI: 10.1016/j.amjsurg.2009.03.012.
[53]
CHENP X, WANGC, LUR L, et al. Multivariable models based on baseline imaging features and clinicopathological characteristics to predict breast pathologic response after neoadjuvant chemotherapy in patients with breast cancer[J]. Breast Care, 2022, 17(3): 306-315. DOI: 10.1159/000521638.
[54]
LEEJ G, JUNS, CHOY W, et al. Deep learning in medical imaging: general overview[J]. Korean J Radiol, 2017, 18(4): 570-584. DOI: 10.3348/kjr.2017.18.4.570.
[55]
ESTEVAA, KUPRELB, NOVOAR A, et al. Dermatologist-level classification of skin cancer with deep neural networks[J]. Nature, 2017, 542(7639): 115-118. DOI: 10.1038/nature21056.
 
 
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