Share：
Share this content in WeChat
X
[Chinese][PDF]1947
• Clinical Article •
Radiomics based on combined machine learning models for prediction of the response to neoadjuvant chemotherapy in mass enhancement breast cancer using magnetic resonance imaging
YUE Wenyi  ZHANG Hongtao  GAO Shen  ZHOU Juan  CAI Jianming  TIAN Ning  DONG Jinghui  LIU Yuan  BAI Xu  SHENG Fugeng 

Cite this article as: YUE W Y, ZHANG H T, GAO S, et al. Radiomics based on combined machine learning models for prediction of the response to neoadjuvant chemotherapy in mass enhancement breast cancer using magnetic resonance imaging[J]. Chin J Magn Reson Imaging, 2024, 15(3): 93-99, 106. DOI:10.12015/issn.1674-8034.2024.03.016.


[Abstract] Objective To investigate the value of radiomics based on combined machine learning models in predicting the response to neoadjuvant chemotherapy (NAC) in mass enhancement breast cancer.Materials and Methods The clinical and imaging data of ninety-seven patients with mass enhancement breast cancer confirmed by histopathology and underwent NAC from January 2018 to October 2021 in the Fifth Medical Center of Chinese PLA General Hospital were retrospectively analyzed in our study. Based on the results of Response Evaluation Criteria in Solid Tumors (RECIST), the patients were classified into effective group and ineffective group. Based on the radiomics features extracted on the first dynamic contrast-enhanced MRI (DCE-MRI) subtraction images before treatment, a high-pass or low-pass wavelet filter and a Laplace-Gaussian filter with different parameters were also introduced to preprocess the original MR images. For feature screening, feature selection methods based on univariate analysis and multivariate analysis were used. The univariate analysis included F-test, chi-square test and mutual information. The multivariate analysis used the least absolute shrinkage and selection operator (LASSO). Support vector machine (SVM), random forest (RM), and logistic regression (LR) were used for modeling, and finally a total of twelve combinations of feature filters and classifiers were combined by crossover. Ten repetitions of five-fold cross-validation were used for training. Finally, area under the curve (AUC), sensitivity, specificity, accuracy, positive prediction value and negative prediction value were used to evaluate the prediction performance.Results Among all cross-combined schemes, the feature screening method that achieved the best classification performance was the F-test method in univariate analysis, and the best classifier was the SVM. The combination screened a total of 191 imaging features with an overall mean AUC of 0.83 [95% confidence interval (CI): 0.80-0.86] in predicting NAC response, and the accuracy of the model was 77% (95% CI: 74%-80%). Specificity was 81% (95% CI: 78%-84%), sensitivity was 71% (95% CI: 65%-77%), positive predictive value was 67% (95% CI: 62%-72%), and negative predictive value was 85% (95% CI: 83%-87%).Conclusions A combined machine learning model based on F-test and SVM validated good performance of radiomics in predicting the response to NAC for mass enhancement breast cancer patients.
[Keywords] breast cancer；neoadjuvant chemotherapy；radiomics；machine learning；magnetic resonance imaging

YUE Wenyi1, 2   ZHANG Hongtao1   GAO Shen1   ZHOU Juan1   CAI Jianming1   TIAN Ning1   DONG Jinghui1   LIU Yuan1   BAI Xu1   SHENG Fugeng1*  

1 Department of Radiology, the Fifth Medical Center of Chinese PLA General Hospital, Beijing 100071, China

2 Graduate School, Chinese PLA General Medical School, Beijing 100853, China

Corresponding author: SHENG F G, E-mail: fugeng_sheng@163.com

Conflicts of interest   None.

Received  2023-06-13
Accepted  2024-02-26
DOI: 10.12015/issn.1674-8034.2024.03.016
Cite this article as: YUE W Y, ZHANG H T, GAO S, et al. Radiomics based on combined machine learning models for prediction of the response to neoadjuvant chemotherapy in mass enhancement breast cancer using magnetic resonance imaging[J]. Chin J Magn Reson Imaging, 2024, 15(3): 93-99, 106. DOI:10.12015/issn.1674-8034.2024.03.016.

[1]
DEMICHELE A, YEE D, ESSERMAN L. Mechanisms of resistance to neoadjuvant chemotherapy in breast cancer[J]. N Engl J Med, 2017, 377(23): 2287-2289. DOI: 10.1056/NEJMcibr1711545.
[2]
HAYASHI M, YAMAMOTO Y, IWASE H. Clinical imaging for the prediction of neoadjuvant chemotherapy response in breast cancer[J/OL]. Chin Clin Oncol, 2020, 9(3): 31 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/32594748/. DOI: 10.21037/cco-20-15.
[3]
HOLMES D, COLFRY A, CZERNIECKI B, et al. Performance and practice guideline for the use of neoadjuvant systemic therapy in the management of breast cancer[J]. Ann Surg Oncol, 2015, 22(10): 3184-3190. DOI: 10.1245/s10434-015-4753-3.
[4]
CAIN E H, SAHA A, HAROWICZ M R, et al. Multivariate machine learning models for prediction of pathologic response to neoadjuvant therapy in breast cancer using MRI features: a study using an independent validation set[J]. Breast Cancer Res Treat, 2019, 173(2): 455-463. DOI: 10.1007/s10549-018-4990-9.
[5]
LIU W F, CHEN W, ZHANG X X, et al. Higher efficacy and reduced adverse reactions in neoadjuvant chemotherapy for breast cancer by using pegylated liposomal doxorubicin compared with pirarubicin[J/OL]. Sci Rep, 2021, 11(1): 199 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/33420241/. DOI: 10.1038/s41598-020-80415-w.
[6]
BIAN T T, WU Z J, LIN Q, et al. Radiomic signatures derived from multiparametric MRI for the pretreatment prediction of response to neoadjuvant chemotherapy in breast cancer[J/OL]. Br J Radiol, 2020, 93(1115): 20200287 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/32822542/. DOI: 10.1259/bjr.20200287.
[7]
BITENCOURT A, DAIMIEL NARANJO I, GULLO R L, et al. AI-enhanced breast imaging: where are we and where are we heading?[J/OL]. Eur J Radiol, 2021, 142: 109882 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/34392105/. DOI: 10.1016/j.ejrad.2021.109882.
[8]
PARK J, CHAE E Y, CHA J H, et al. Comparison of mammography, digital breast tomosynthesis, automated breast ultrasound, magnetic resonance imaging in evaluation of residual tumor after neoadjuvant chemotherapy[J/OL]. Eur J Radiol, 2018, 108: 261-268 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/30396666/. DOI: 10.1016/j.ejrad.2018.09.032.
[9]
EOM H J, CHA J H, CHOI W J, et al. Predictive clinicopathologic and dynamic contrast-enhanced MRI findings for tumor response to neoadjuvant chemotherapy in triple-negative breast cancer[J/OL]. AJR Am J Roentgenol, 2017, 208(6): W225-W230 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/28350486/. DOI: 10.2214/AJR.16.17125.
[10]
LIU J, SUN D, CHEN L L, et al. Radiomics analysis of dynamic contrast-enhanced magnetic resonance imaging for the prediction of sentinel lymph node metastasis in breast cancer[J/OL]. Front Oncol, 2019, 9: 980 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/31632912/. DOI: 10.3389/fonc.2019.00980.
[11]
LIU H, ZHANG F X, ZHANG F. Research status of DCE-MRI semi-quantitative and quantitative analysis in distinguishing benign and malignant cervical lymph nodes[J]. Chin J Magn Reson Imag, 2021, 12(1): 103-105. DOI: 10.12015/issn.1674-8034.2021.01.024.
[12]
KUHL C K. RECIST needs revision: a wake-up call for radiologists[J]. Radiology, 2019, 292(1): 110-111. DOI: 10.1148/radiol.2019190785.
[13]
WEIS J A, MIGA M I, YANKEELOV T E. Three-dimensional image-based mechanical modeling for predicting the response of breast cancer to neoadjuvant therapy[J/OL]. Comput Methods Appl Mech Eng, 2017, 314: 494-512 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/28042181/. DOI: 10.1016/j.cma.2016.08.024.
[14]
HA R, CHIN C, KARCICH J, et al. Prior to initiation of chemotherapy, can we predict breast tumor response? deep learning convolutional neural networks approach using a breast MRI tumor dataset[J]. J Digit Imaging, 2019, 32(5): 693-701. DOI: 10.1007/s10278-018-0144-1.
[15]
CHAMMING'S F, UENO Y, FERRÉ R, et al. Features from computerized texture analysis of breast cancers at pretreatment MR imaging are associated with response to neoadjuvant chemotherapy[J]. Radiology, 2018, 286(2): 412-420. DOI: 10.1148/radiol.2017170143.
[16]
HORVAT N, VEERARAGHAVAN H, KHAN M, et al. MR imaging of rectal cancer: radiomics analysis to assess treatment response after neoadjuvant therapy[J]. Radiology, 2018, 287(3): 833-843. DOI: 10.1148/radiol.2018172300.
[17]
MOKRANE F Z, LU L, VAVASSEUR A, et al. Radiomics machine-learning signature for diagnosis of hepatocellular carcinoma in cirrhotic patients with indeterminate liver nodules[J]. Eur Radiol, 2020, 30(1): 558-570. DOI: 10.1007/s00330-019-06347-w.
[18]
CABALLO M, SANDERINK W B G, HAN L Y, et al. Four-dimensional machine learning radiomics for the pretreatment assessment of breast cancer pathologic complete response to neoadjuvant chemotherapy in dynamic contrast-enhanced MRI[J]. J Magn Reson Imaging, 2023, 57(1): 97-110. DOI: 10.1002/jmri.28273.
[19]
BITENCOURT A G V, GIBBS P, ROSSI SACCARELLI C, et al. MRI-based machine learning radiomics can predict HER2 expression level and pathologic response after neoadjuvant therapy in HER2 overexpressing breast cancer[J/OL]. EBioMedicine, 2020, 61: 103042 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/33039708/. DOI: 10.1016/j.ebiom.2020.103042.
[20]
LIU Z Y, LI Z L, QU J 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.
[21]
SUN C J, SHI L H, GU Y S, et al. Clinical effects of neoadjuvant chemotherapy in treating breast cancer[J]. Cancer Biother Radiopharm, 2021, 36(2): 174-179. DOI: 10.1089/cbr.2019.3545.
[22]
BRAMAN N M, ETESAMI M, PRASANNA P, et al. Erratum to: 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): 80 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/28693537/. DOI: 10.1186/s13058-017-0862-1.
[23]
CHOUDHERY S, GOMEZ-CARDONA D, FAVAZZA C 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(Suppl 1): S145-S154 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/33160859/. DOI: 10.1016/j.acra.2020.10.020.
[24]
Breast Cancer Committee of Chinese anti-Cancer Association. Guidelines and specifications for diagnosis and treatment of breast cancer of China Anti-Cancer Association (2017 edition)[J]. China Oncol, 2017, 27(9): 695-759. DOI: 10.19401/j.cnki.1007-3639.2017.09.004.
[25]
COATES A S, WINER E P, GOLDHIRSCH A, et al. Tailoring therapies: improving the management of early breast cancer: St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2015[J]. Ann Oncol, 2015, 26(8): 1533-1546. DOI: 10.1093/annonc/mdv221.
[26]
PARK H, SHOLL L M, HATABU H, et al. Imaging of precision therapy for lung cancer: current state of the art[J]. Radiology, 2019, 293(1): 15-29. DOI: 10.1148/radiol.2019190173.
[27]
RUDIE J D, RAUSCHECKER A M, BRYAN R N, et al. Emerging applications of artificial intelligence in neuro-oncology[J]. Radiology, 2019, 290(3): 607-618. DOI: 10.1148/radiol.2018181928.
[28]
XIONG Q, ZHOU X, LIU Z, et al. Multiparametric MRI-based radiomics analysis for prediction of breast cancers insensitive to neoadjuvant chemotherapy[J/OL]. Breast, 2019, 44: S68 [2023-06-12]. https://www.thebreastonline.com/article/S0960-9776(19)30256-5/abstract DOI: 10.1016/s0960-9776(19)30256-5.
[29]
YAO C, YANG Z Q, YANG J D, et al. A nomogram based on DCE-MRI Radscore and hormone receptor status for predicting drug insensitive to neoadjuvant chemotherapy in breast cancer patients[J]. Int J Med Radiol, 2022, 45(2): 130-134. DOI: 10.19300/j.2022.L19006.
[30]
WANG L L, HOU J, JIN K. First-order texture features of DCE-MRI in predicting the molecular subtypes of breast cancer[J]. Mod Med Imageology, 2022, 31(4): 595-599. DOI: 10.3969/j.issn.1006-7035.2022.04.001.
[31]
WANG Y, ZHANG L. Run-length matrix analysis of ADC images of solid occupying liver[J]. J Imag Res Med Appl, 2021, 5(9): 219-220, 223. DOI: 10.3969/j.issn.2096-3807.2021.09.105.
[32]
MAO N, WANG Q L, LIU M J, et al. Computerized image analysis to differentiate benign and malignant breast tumors on magnetic resonance diffusion weighted image: a preliminary study[J]. J Comput Assist Tomogr, 2019, 43(1): 93-97. DOI: 10.1097/RCT.0000000000000793.
[33]
LEE Y, RAGGUETT R M, MANSUR R B, et al. Corrigendum to "Applications of machine learning algorithms to predict therapeutic outcomes in depression: A meta-analysis and systematic review." J Affect Disord. 241 (2018) 519-532 [J]. Journal of affective disorders, 2020, 274: 1211-5. DOI: 10.1016/j.jad.2020.02.037LEE Y N, RAGGUETT R M, MANSUR R B, et al. Applications of machine learning algorithms to predict therapeutic outcomes in depression: a meta-analysis and systematic review[J/OL]. J Affect Disord, 2018, 241: 519-532 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/30153635/. DOI: 10.1016/j.jad.2018.08.073.
[34]
ERICKSON B J, KORFIATIS P, AKKUS Z, et al. Machine learning for medical imaging[J]. Radiographics, 2017, 37(2): 505-515. DOI: 10.1148/rg.2017160130.
[35]
ZHANG B, TIAN J, DONG D, et al. Radiomics features of multiparametric MRI as novel prognostic factors in advanced nasopharyngeal carcinoma[J]. Clin Cancer Res, 2017, 23(15): 4259-4269. DOI: 10.1158/1078-0432.CCR-16-2910.
[36]
XIA B Q, LI C P, QIAN C X, et al. Radiomics based on machine learning in predicting the long-term prognosis for triple-negative breast cancer after neoadjuvant chemotherapy[J]. Chin J Radiol, 2021, 55(10): 1059-1064. DOI: 10.3760/cma.j.cn112149-20201215-01308.
[37]
SUN J Y, ZHONG G Q, HUANG K Z, et al. Banzhaf random forests: cooperative game theory based random forests with consistency[J/OL]. Neural Netw, 2018, 106: 20-29 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/30007122/. DOI: 10.1016/j.neunet.2018.06.006.
[38]
HUANG S J, CAI N G, PACHECO P P, et al. Applications of support vector machine (SVM) learning in cancer genomics[J]. Cancer Genomics Proteomics, 2018, 15(1): 41-51. DOI: 10.21873/cgp.20063.
[39]
STOLTZFUS J C. Logistic regression: a brief primer[J]. Acad Emerg Med, 2011, 18(10): 1099-1104. DOI: 10.1111/j.1553-2712.2011.01185.x.
[40]
BLAGUS R, LUSA L. SMOTE for high-dimensional class-imbalanced data[J/OL]. BMC Bioinformatics, 2013, 14: 106 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/23522326/. DOI: 10.1186/1471-2105-14-106.
[41]
ROMEO V, ACCARDO G, PERILLO T, et al. Assessment and prediction of response to neoadjuvant chemotherapy in breast cancer: a comparison of imaging modalities and future perspectives[J/OL]. Cancers, 2021, 13(14): 3521 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/34298733/. DOI: 10.3390/cancers13143521.
[42]
RAUCH G M, ADRADA B E, KUERER H M, et al. Multimodality imaging for evaluating response to neoadjuvant chemotherapy in breast cancer[J]. AJR Am J Roentgenol, 2017, 208(2): 290-299. DOI: 10.2214/AJR.16.17223.
[43]
GIANNI L, PIENKOWSKI T, IM Y H, et al. 5-year analysis of neoadjuvant pertuzumab and trastuzumab in patients with locally advanced, inflammatory, or early-stage HER2-positive breast cancer (NeoSphere): a multicentre, open-label, phase 2 randomised trial[J]. Lancet Oncol, 2016, 17(6): 791-800. DOI: 10.1016/S1470-2045(16)00163-7.
[44]
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.
[45]
PENG Y S, CHENG Z L, GONG C, et al. Pretreatment DCE-MRI-based deep learning outperforms radiomics analysis in predicting pathologic complete response to neoadjuvant chemotherapy in breast cancer[J/OL]. Front Oncol, 2022, 12: 846775 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/35359387/. DOI: 10.3389/fonc.2022.846775.
[46]
TAHMASSEBI A, WENGERT G J, HELBICH T H, et al. Impact of machine learning with multiparametric magnetic resonance imaging of the breast for early prediction of response to neoadjuvant chemotherapy and survival outcomes in breast cancer patients[J]. Invest Radiol, 2019, 54(2): 110-117. DOI: 10.1097/RLI.0000000000000518.
[47]
FAN M, WU G L, CHENG H, et al. Radiomic analysis of DCE-MRI for prediction of response to neoadjuvant chemotherapy in breast cancer patients[J/OL]. Eur J Radiol, 2017, 94: 140-147[2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/28712700/. DOI: 10.1016/j.ejrad.2017.06.019.
[48]
LI H J, CAO K, ZHENG H, et al. Whole-lesion histogram analysis of apparent diffusion coefficient in prediction of response to neoadj uvant chemotherapy in patients with cervical carcinoma[J]. J Pract Radiol, 2022, 38(1): 85-89. DOI: 10.3969/j.issn.1002-1671.2022.01.021.
[49]
TAMAYO CARABAÑO D, ÁLVAREZ PÉREZ R, DE BONILLA DAMIÁ Á, et al. Sentinel lymph node biopsy in N+ breast cancer with conversion into N0 after neoadjuvant chemotherapy[J]. Rev Esp Med Nucl Imagen Mol, 2019, 38(3): 140-146. DOI: 10.1016/j.remn.2018.11.006.
[50]
KIM J Y, PARK H S, KIM S, et al. Prognostic nomogram for prediction of axillary pathologic complete response after neoadjuvant chemotherapy in cytologically proven node-positive breast cancer[J/OL]. Medicine, 2015, 94(43): e1720 [2023-06-12]. https://pubmed.ncbi.nlm.nih.gov/26512562/. DOI: 10.1097/MD.0000000000001720.

PREV A MRI-Radiomics-based model predicts EGFR mutation status in brain metastases in lung cancer patients
NEXT Value of MRI radiomics in predicting molecular subtypes of invasive breast carcinoma of no special type
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn