|Table of Contents|

Prediction of axillary lymph node metastatic state in breast cancer with mass like enhancement by MRI radiomics-based model

Journal Of Modern Oncology[ISSN:1672-4992/CN:61-1415/R]

Issue:
2023 03
Page:
506-512
Research Field:
Publishing date:

Info

Title:
Prediction of axillary lymph node metastatic state in breast cancer with mass like enhancement by MRI radiomics-based model
Author(s):
WEN Jie1; WANG Meng1; LIU Zhou1; REN Ya1; YANG Qian1; XIANG Lu1; GENG Yayuan2; LUO Dehong1; 3
1.National Cancer Center/National Clinical Research Center for Cancer/Department of Radiology,Cancer Hospital & Shenzhen Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College,Guangdong Shenzhen 518116,China;2.Huiying Medical Technology Co.,Ltd.,Beijing 100089,China;3.National Cancer Center/National Clinical Research Center for Cancer/Department of Radiology,Cancer Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College,Beijing 100021,China.
Keywords:
breast cancer; axillary lymph node metastasis; multi-parameter magnetic resonance imaging; radiology; machine learning model
PACS:
R737.9
DOI:
10.3969/j.issn.1672-4992.2023.03.023
Abstract:
Objective:To investigate the value of a machine learning model based on radiomics features derived from primary breast cancer lesions with mass like enhancement on multi-parametric MR images in predicting the metastatic status of axillary lymph nodes.Methods:We enrolled 98 breast cancer patients with 114 breast lesions.107 radiomics features(histogram-based,shape-based and texture-based features) were extracted from multi-parametric images(T2,DWI,ADC map and DCE) on the Radcloud platform(Huiying Medical Technology Co.,Ltd.).To reduce redundant features,variance threshold(variance threshold=0.8) and least absolute shrinkage and selection operator(lasso) were sequentially used.Based on selected optimal features,3 classifiers,including k-Nearest Neighbor(KNN),support vector machine(SVM) and logistic regression(LR) were independently performed to establish the radiomics-based prediction model and receiver operating characteristic(ROC) analysis was used to evaluate the prediction performance with mean area under the curve(AUC) calculated in test set.Results:In this study,there are 46 cases(56 breast cancer lesions) with lymph node metastasis，52 cases(58 breast cancer lesions) without lymph node metastasis.After feature selection,the optimal radiomic features(12,10,29,10,16 features for each of the five folds with low consistency) were used to build the prediction model.Among the three radiomics-based models,SVM model had the best performance with mean AUC of 0.805 in the validation set,higher than the mean AUC of KNN and LR(0.783,0.802).Conclusion:MRI texture analysis of breast cancer can be used as a non-invasive index to help predict lymph node metastasis status in advance,which warranted further study.

References:

[1]TA D,KHAN M,ISHAQUE A,et al.Reliability of 3D texture analysis:A multicenter MRI study of the brain［J］.Journal of Magnetic Resonance Imaging,2020,51(4):1200-1209.
[2] NAGANAWA S,KIM J,YIP S,et al.Texture analysis of T2-weighted MRI predicts SDH mutation in paraganglioma［J］.Neuroradiology,2021,63(4):547-554.
[3] 许东,张文军,张学喜.MRI纹理参数对乳腺癌分子分型鉴别的初步探讨［J］.中国中西医结合影像学杂志,2019,17(2):132-136. XU D,ZHANG WJ,ZHANG XX.Preliminary study on the classification of molecular subtypes in breast cancer by the texture parameters［J］.Chinese Imaging Journal of Integrated Traditional and Western Medicine,2019,17(2):132-136.
[4] 陈文静,牟玮,张文馨,等.MR动态增强图像纹理分析判断乳腺结节良恶性的价值［J］.中国医学影像技术,2017,33(5):647-651. CHEN WJ,MOU W,ZHANG WX,et al.Value of texture feature analysis of MRI dynamic contrast enhancement in diagnosis of benign and malignant breast nodules［J］.Chinese Journal of Medical Imaging Technology,2017,33(5):647-651.
[5]WANG Y,LIAO X,XIAO F,et al.Magnetic resonance imaging texture analysis in differentiating benign and malignant breast lesions of breast imaging reporting and data system 4:A preliminary study［J］.Journal of Computer Assisted Tomography,2020,44(1):83-89.
[6] HENDERSON S,PUEDIE C,MICHIE C,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］.European Radiology,2017,27(11):4602-4611.
[7]LAE EN,DAESUNG K,JU SE,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.
[8]刘晓东,王新宇,宁刚.MRI影像组学术前预测乳腺浸润性导管癌Ki-67表达［J］.中国医学影像技术,2022,38(2):210-214. LIU XD,WANG XY,NING G.MRI radiomics for peroperative predicting Ki-67 expression of breast invasive ductal carcinoma［J］.Chinese Journal of Medical Imaging Technology,2022,38(2):210-214.
[9] YANG JB,WANG T,YANG LF,et al.Preoperative prediction of axillary lymph node metastasis in breast cancer using mammography-based radiomics method［J］.Scientific Reports,2019,9(1):4429-4440.
[10]MARINO MA,AVENDAO D,ZAPATA P,et al.Lymph node imaging in patients with primary breast cancer:Concurrent diagnostic tools［J］.The Oncologist,2020,2(25):231-242.
[11]CUI X,WANG N,ZHAO Y,et al.Preoperative prediction of axillary lymph node metastasis in breast cancer using radiomics features of DCE-MRI［J］.Scientific Reports,2019,9(1):2240-2248.
[12]张东蕾,钱银锋,李伟,等.影响乳腺癌腋窝淋巴结转移的因素分析及不同诊断方式的对比研究［J］.中国临床医学影像杂志,2021,32(2):94-99. ZHANG DL,QIAN YF,LI W,et al.Analysis of factors affecting axillary lymph node metastasis of breast cancer and comparative study of different diagnositic methods［J］.Journal of China Clinic Medical Imaging,2021,32(2):94-99.
[13]薛梅,李静,车树楠,等.乳腺癌多模态磁共振影像特征与腋窝淋巴结转移的相关性研究 ［J］.磁共振成像,2020,11(07):540-545. XUE M,LI J,CHE SN,et al.The correlation between multiparametric MR imaging characteristics of breast cancer and axillary lymph node metastasis［J］.Chin J Magn Reson Imaging,2020,11(07):540-545.
[14]LIN YP,YIN WJ,YAN TT,et al.Site-specific relapse pattern of the triple negative tumors in Chinese breast cancer patients［J］.BMC Cancer,2009,9(24):342-349.
[15]TAGLIAFICO AS,PIANA M,SCHENONE D,et al.Overview of radiomics in breast cancer diagnosis and prognostication［J］.Breast,2020,2(49):74-80.
[16] LIU CL,DING J,SPUHLER K,et al.Preoperative prediction of sentinel lymph node metastasis in breast cancer by radiomic signatures from dynamic contrast-enhanced MRI［J］.Journal of Magnetic Resonance Imaging,2019,49(1):131-140.
[17] LIU Z,FENG B,LI C,et al.Preoperative prediction of lymphovascular invasion in invasive breast cancer with dynamic contrast-enhanced-MRI-based radiomics［J］.Journal of Magnetic Resonance Imaging,2019,50(3):847-857.
[18] CHAI R,MA H,XU M,et al.Differentiating axillary lymph node metastasis in invasive breast cancer patients:A comparison of radiomic signatures from multiparametric breast MR sequences［J］.Journal of Magnetic Resonance Imaging,2019,50(4):1125-1132.
[19]邢滔,陈基明,颜秀芳,等.MRI纹理分析预测乳腺癌腋窝淋巴结转移的价值［J］.临床放射学杂志,2019,38(12):2290-2294. XING T,CHEN JM,YAN XF,et al.The value of MRI texture analysis to predict axillary lymph node metastasis in breast cancer patients［J］.J Clin Radiol,2019,38(12):2290-2294.
[20]单嫣娜,龚向阳,丁忠祥,等.动态增强MRI影像组学特征预测乳腺癌腋窝淋巴结转移的价值 ［J］.中华放射学杂志,2019,53(9):742-747. SHAN YN,GONG XY,DING ZX,et al.Dynamic contrast-enhanced MRI radiomic features predict axillary lymph node metastasis of breast cancer［J］.Chin J Radiol,2019,53(9):742-747.
[21]黄栎有,高先聪,尤传文.基于乳腺X线图像纹理特征建立机器学习模型在鉴别良恶性乳腺肿块中的价值［J］.放射学实践,2021,36(4):480-483. HUANG LY,GAO XC,YOU CW.Application of machine learning model based on the texture features on mammgraphy in the differeniation diagnosis of benign and malignant breast masses［J］.Radiol Practice,2021,36(4):480-483.
[22] CHEN C,QIN Y,CHEN H,et al.A meta-analysis of the diagnostic performance of machine learning-based MRI in the prediction of axillary lymph node metastasis in breast cancer patients［J］.Insights Imaging,2021,12(1):156-168.

Memo

Memo:
中国医学科学院肿瘤医院深圳医院院内科研课题资助（编号：E010321005）
Last Update: 2022-12-30