Share：
Share this content in WeChat
X
[Chinese] [PDF] 65 2
• Clinical Article •
A study of an MRI-based radiomics and deep learning model for predicting axillary lymph node metastasis in breast cancer with 1 to 2 positive sentinel lymph nodes
CAO Ziwei  XU Mengting  LIAO Jun  LI Shunian  WANG Danyang  LI Yongli  TAN Hongna 

DOI:10.12015/issn.1674-8034.2026.07.009.


[Abstract] Objective To evaluate the clinical value of magnetic resonance imaging (MRI)-based radiomics and deep learning (DL) features in predicting axillary lymph node (ALN) status in breast cancer patients with 1 to 2 positive sentinel lymph nodes (SLNs).Materials and Methods We retrospectively analyzed clinicopathological and MRI data from breast cancer patients with pathologically confirmed 1 to 2 positive SLNs at our institution between January 2017 and December 2024. The tumor regions of interest (ROIs) were manually delineated on diffusion-weighted imaging (DWI) and third-phase dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) images using ITK-SNAP 3.8.0 software. Radiomics features were extracted using PyRadiomics, and DL features were extracted using a pretrained Inception-v3 model. Features strongly associated with ALN metastasis were selected using statistical tests, minimum redundancy maximum relevance (mRMR), Spearman correlation analysis, and the least absolute shrinkage and selection operator (LASSO) algorithm. Clinicopathological risk factors associated with ALN metastasis were identified using multivariate logistic regression analysis. Radiomics model, DL model, deep learning radiomics (DLR) model, clinical model and combined model incorporating clinical features were constructed to predict ALN status. The area under the curve (AUC), sensitivity, specificity, and accuracy were calculated to evaluate model performance. Decision curve analysis was performed to assess clinical utility.Results A total of 256 patients were enrolled, with 105 patients in the ALN-positive group and 151 in the ALN-negative group. Logistic regression analysis demonstrated that maximum tumor diameter, lymphovascular invasion (LVI), and the number of positive SLNs were independent risk factors for ALN metastasis (all P < 0.05). In the validation cohort, the DLR model achieved an AUC of 0.800 (95% CI: 0.661 to 0.909), outperforming the radiomics model (AUC = 0.734, 95% CI: 0.580 to 0.861) and the DL model (AUC = 0.774, 95% CI: 0.632 to 0.903). The combined model incorporating clinical features showed optimal performance, achieving an AUC, sensitivity, specificity, and accuracy of 0.851 (95% CI: 0.722 to 0.953), 76.2%, 90.3%, and 84.6%, respectively. It also achieved a higher clinical net benefit within a reasonable threshold probability range.Conclusions Multiparametric MRI-based radiomics, DL, and DLR models can assist in assessing the risk of ALN metastasis in breast cancer patients with 1 to 2 positive SLNs preoperatively. The combined model incorporating clinical features demonstrated favorable clinical utility for comprehensive risk stratification of ALN metastasis after sentinel lymph node biopsy (SLNB). Following further multicenter prospective validation, it may provide auxiliary support for individualized clinical decision-making regarding axillary management.
[Keywords] breast cancer；sentinel lymph node；axillary lymph node；radiomics；deep learning；magnetic resonance imaging

CAO Ziwei1   XU Mengting2   LIAO Jun1   LI Shunian1   WANG Danyang1   LI Yongli3   TAN Hongna3*  

1 Department of Medical Imaging, People's Hospital of Zhengzhou University, Zhengzhou 450003, China

2 Department of Medical Imaging, People's Hospital of Henan University, Zhengzhou 450003, China

3 Department of Medical Imaging, Henan Provincial People's Hospital, Zhengzhou 450003, China

Corresponding author: TAN H N, E-mail: natan2000@126.com

Conflicts of interest   None.

Received  2026-03-06
Accepted  2026-06-22
DOI: 10.12015/issn.1674-8034.2026.07.009
DOI:10.12015/issn.1674-8034.2026.07.009.

[1]
BRAY F, LAVERSANNE M, SUNG H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2024, 74(3): 229-263. DOI: 10.3322/caac.21834.
[2]
LI C Z, ZHANG P, LV J, et al. Axillary management in patients with clinical node-negative early breast cancer and positive sentinel lymph node: a systematic review and meta-analysis[J/OL]. Front Oncol, 2024, 13: 1320867 [2025-10-27]. https://www.frontiersin.org/articles/10.3389/fonc.2023.1320867/full. DOI: 10.3389/fonc.2023.1320867.
[3]
YU Y F, TAN Y J, XIE C M, 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/OL]. JAMA Netw Open, 2020, 3(12): e2028086[2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/33289845/. DOI: 10.1001/jamanetworkopen.2020.28086.
[4]
SUN B Q, SHAO G D, SHI M M, et al. Preoperative comprehensive risk estimation for axillary lymph node metastasis in breast cancer: development and verification of a network-based prediction model[J/OL]. Sci Rep, 2025, 15: 1524 [2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/39789023/. DOI: 10.1038/s41598-024-84904-0.
[5]
SAMIEI S, GRANZIER R W Y, IBRAHIM A, et al. Dedicated axillary MRI-based radiomics analysis for the prediction of axillary lymph node metastasis in breast cancer[J/OL]. Cancers, 2021, 13(4): 757 [2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/33673071/. DOI: 10.3390/cancers13040757.
[6]
LANGER I, GULLER U, BERCLAZ G, et al. Morbidity of sentinel lymph node biopsy (SLN) alone versus SLN and completion axillary lymph node dissection after breast cancer surgery: a prospective Swiss multicenter study on 659 patients[J]. Ann Surg, 2007, 245(3): 452-461. DOI: 10.1097/01.sla.0000245472.47748.ec.
[7]
NOGUCHI M, INOKUCHI M, YOKOI-NOGUCHI M, et al. Conservative axillary surgery may prevent arm lymphedema without increasing axillary recurrence in the surgical management of breast cancer[J/OL]. Cancers, 2023, 15(22): 5353 [2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/38001613/. DOI: 10.3390/cancers15225353.
[8]
ZAHOOR S, HAJI A, BATTOO A, et al. Sentinel lymph node biopsy in breast cancer: a clinical review and update[J/OL]. J Breast Cancer, 2017, 20(3): 217 [2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/28970846/. DOI: 10.4048/jbc.2017.20.3.217.
[9]
LYMAN G H, SOMERFIELD M R, BOSSERMAN L D, et al. Sentinel lymph node biopsy for patients with early-stage breast cancer: American society of clinical oncology clinical practice guideline update[J]. J Clin Oncol, 2017, 35(5): 561-564. DOI: 10.1200/jco.2016.71.0947.
[10]
VERONESI U, PAGANELLI G, VIALE G, et al. Sentinel-lymph-node biopsy as a staging procedure in breast cancer: update of a randomised controlled study[J]. Lancet Oncol, 2006, 7(12): 983-990. DOI: 10.1016/S1470-2045(06)70947-0.
[11]
KRAG D N, ANDERSON S J, JULIAN T B, et al. Sentinel-lymph-node resection compared with conventional axillary-lymph-node dissection in clinically node-negative patients with breast cancer: overall survival findings from the NSABP B-32 randomised phase 3 trial[J]. Lancet Oncol, 2010, 11(10): 927-933. DOI: 10.1016/S1470-2045(10)70207-2.
[12]
LIANG Y, CHEN X S, TONG Y W, et al. Higher axillary lymph node metastasis burden in breast cancer patients with positive preoperative node biopsy: may not be appropriate to receive sentinel lymph node biopsy in the post-ACOSOG Z0011 trial era[J/OL]. World J Surg Oncol, 2019, 17(1): 37 [2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/30786903/. DOI: 10.1186/s12957-019-1582-z.
[13]
GIULIANO A E, BALLMAN K V, MCCALL L, et al. Effect of axillary dissection vs No axillary dissection on 10-year overall survival among women with invasive breast cancer and sentinel node metastasis: the ACOSOG Z0011 (alliance) randomized clinical trial[J]. JAMA, 2017, 318(10): 918-926. DOI: 10.1001/jama.2017.11470.
[14]
MARINO M A, AVENDANO D, ZAPATA P, et al. Lymph node imaging in patients with primary breast cancer: concurrent diagnostic tools[J/OL]. Oncologist, 2020, 25(2): e231-e242 [2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/32043792/. DOI: 10.1634/theoncologist.2019-0427.
[15]
PAN G H, PAN Z J, CHEN W B, et al. Integration of MRI radiomics and clinical data for preoperative prediction of vascular invasion in breast cancer: a deep learning approach[J/OL]. Magn Reson Imaging, 2025, 118: 110339 [2025-10-27]. https://www.sciencedirect.com/science/article/pii/S0730725X25000219via%3Dihub. DOI: 10.1016/j.mri.2025.110339.
[16]
CONTI A, DUGGENTO A, INDOVINA I, et al. Radiomics in breast cancer classification and prediction[J/OL]. Semin Cancer Biol, 2021, 72: 238-250 [2025-10-27]. https://pubmed.ncbi.nlm.nih.gov/32371013/. DOI: 10.1016/j.semcancer.2020.04.002.
[17]
MA W H, YANG L, ZHANG Y, et al. Heterogeneity assessment of breast cancer tumor microenvironment: multiparametric quantitative analysis with DCE-MRI and discovery of radiomics biomarkers[J/OL]. Breast Cancer Targets Ther, 2025, 17: 573-581 [2026-03-05]. https://pubmed.ncbi.nlm.nih.gov/40657027/. DOI: 10.2147/BCTT.S530834.
[18]
GILLIES R J, KINAHAN P E, HRICAK H. Radiomics: images are more than pictures, they are data[J]. Radiology, 2016, 278(2): 563-577. DOI: 10.1148/radiol.2015151169.
[19]
LIN G H, CHEN W Y, FAN Y Y, et al. Machine learning radiomics-based prediction of non-sentinel lymph node metastasis in Chinese breast cancer patients with 1-2 positive sentinel lymph nodes: a multicenter study[J]. Acad Radiol, 2024, 31(8): 3081-3095. DOI: 10.1016/j.acra.2024.02.010.
[20]
JIANG W, WANG X P, GUO Y, et al. Preoperative multiparametric magnetic resonance imaging and contrast-enhanced mammography radiomics based nonsentinel lymph node metastasis prediction in breast cancer[J]. J Intell Med, 2025, 2(2): 103-114. DOI: 10.1002/jim4.70007.
[21]
CHEN Y H, WANG L J, DONG X, et al. Deep learning radiomics of preoperative breast MRI for prediction of axillary lymph node metastasis in breast cancer[J]. J Digit Imaging, 2023, 36(4): 1323-1331. DOI: 10.1007/s10278-023-00818-9.
[22]
LI Z Y, GAO J, ZHOU H, et al. Multiregional dynamic contrast-enhanced MRI-based integrated system for predicting pathological complete response of axillary lymph node to neoadjuvant chemotherapy in breast cancer: multicentre study[J/OL]. eBioMedicine, 2024, 107: 105311 [2026-03-05]. https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(24)00347-5/fulltext. DOI: 10.1016/j.ebiom.2024.105311.
[23]
ZHANG B Y, YU Y M, MAO Y, et al. Development of MRI-based deep learning signature for prediction of axillary response after NAC in breast cancer[J]. Acad Radiol, 2024, 31(3): 800-811. DOI: 10.1016/j.acra.2023.10.004.
[24]
WANG M Y, ZHANG X L, YANG X Y, et al. Multiphase dynamic contrast-enhanced magnetic resonance imaging radiomics nomogram for predicting axillary lymph node metastasis in breast cancer[J]. Gland Surg, 2025, 14(10): 1975-1989. DOI: 10.21037/gs-2025-317.
[25]
LI X G, DONG Q J, CONG C, et al. MRI-based radiomics nomogram for predicting CD8-positive tumor-infiltrating lymphocytes levels in HER2-positive breast cancer[J/OL]. Front Oncol, 2025, 15: 1612631 [2026-03-05]. https://pubmed.ncbi.nlm.nih.gov/41142610/. DOI: 10.3389/fonc.2025.1612631.
[26]
HAMMOND M E H, HAYES D F, WOLFF A C, et al. American society of clinical oncology/college of American pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer[J]. J Oncol Pract, 2010, 6(4): 195-197. DOI: 10.1200/jop.777003.
[27]
Guideline for HER2 detection in breast cancer, the 2019 version Writing Group. Guideline for HER2 detection in breast cancer, the 2019 version[J/OL]. Chinese Journal of Pathology, 2019, 48(3): 169-175 [2026-03-04]. https://rs.yiigle.com/cmaid/1117746. DOI: 10.3760/cma.j.issn.0529-5807.2019.03.001.
[28]
Chinese Anti-Cancer Association Breast Cancer Professional Committee. Guidelines and norms for diagnosis and treatment of breast cancer of China Anti-Cancer Association (2021 edition)[J]. 2021, 31(10): 954-1040. DOI: 10.19401/j.cnki.1007-3639.2021.10.013.
[29]
ZHANG D Y, SHEN M Y, ZHANG L, et al. Establishment of an interpretable MRI radiomics-based machine learning model capable of predicting axillary lymph node metastasis in invasive breast cancer[J/OL]. Sci Rep, 2025, 15(1): 26030 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/40676103/. DOI: 10.1038/s41598-025-10818-0.
[30]
RYU Y J, KANG S J, CHO J S, et al. Lymphovascular invasion can be better than pathologic complete response to predict prognosis in breast cancer treated with neoadjuvant chemotherapy[J/OL]. Medicine, 2018, 97(30): e11647[2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/30045313/. DOI: 10.1097/MD.0000000000011647.
[31]
QIU Y, ZHANG X, WU Z Y, et al. MRI-based radiomics nomogram: prediction of axillary non-sentinel lymph node metastasis in patients with sentinel lymph node-positive breast cancer[J/OL]. Front Oncol, 2022, 12: 811347 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/35296027/. DOI: 10.3389/fonc.2022.811347.
[32]
WU X Q, GUO Y, SA Y, et al. Contrast-enhanced spectral mammography-based prediction of non-sentinel lymph node metastasis and axillary tumor burden in patients with breast cancer[J/OL]. Front Oncol, 2022, 12: 823897 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/35615151/. DOI: 10.3389/fonc.2022.823897.
[33]
HUANG Z, WU Z, ZOU Q Q, et al. Risk factors of non-sentinel lymph node metastasis in breast cancer with 1-2 sentinel lymph node macrometastases underwent total mastectomy: a case-control study[J/OL]. World J Surg Oncol, 2023, 21(1): 125 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/37024930/. DOI: 10.1186/s12957-023-02888-z.
[34]
MCBEE M P, AWAN O A, COLUCCI A T, et al. Deep learning in radiology[J]. Acad Radiol, 2018, 25(11): 1472-1480. DOI: 10.1016/j.acra.2018.02.018.
[35]
ZHANG X D, LIU M H, REN W Q, et al. Predicting of axillary lymph node metastasis in invasive breast cancer using multiparametric MRI dataset based on CNN model[J/OL]. Front Oncol, 2022, 12: 1069733 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/36561533/. DOI: 10.3389/fonc.2022.1069733.
[36]
LIANG T, HU B, DU H W, et al. Predictive value of T2-weighted magnetic resonance imaging for the prognosis of patients with mass-type breast cancer with peritumoral edema[J/OL]. Oncol Lett, 2020, 20(6): 314 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/33133250/. DOI: 10.3892/ol.2020.12177.
[37]
LI X, YANG L F, JIAO X. Comparison of traditional radiomics, deep learning radiomics and fusion methods for axillary lymph node metastasis prediction in breast cancer[J]. Acad Radiol, 2023, 30(7): 1281-1287. DOI: 10.1016/j.acra.2022.10.015.
[38]
LIU W C, CHEN W B, XIA J, et al. Lymph node metastasis prediction and biological pathway associations underlying DCE-MRI deep learning radiomics in invasive breast cancer[J/OL]. BMC Med Imaging, 2024, 24(1): 91 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/38627678/. DOI: 10.1186/s12880-024-01255-y.
[39]
LIU H, ZOU L W, XU N, et al. Deep learning radiomics based prediction of axillary lymph node metastasis in breast cancer[J/OL]. npj Breast Cancer, 2024, 10: 22 [2026-03-04]. https://pubmed.ncbi.nlm.nih.gov/38472210/. DOI: 10.1038/s41523-024-00628-4.

PREV Cardiac magnetic resonance tissue characteristics and their association with systemic inflammation in acute Takotsubo cardiomyopathy: A longitudinal quantitative study
NEXT Analysis of the fat fraction histogram of the brown adipose depot in the supraclavicular region of patients with colorectal cancer based on magnetic resonance mDIXON technology
  


LINK


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