This bi-directional, multicentre study aims to assess multiparametric MRI Radiomics-based prediction model for identifying metastasis lymph nodes and prognostic prediction in breast cancer.

Sensitivity for prediction of lymph node metastasis and survival of currently available prognostic scores in limited. This study proposes to establish a deep learning algorithms of multiparametric MRI radiomics and nomogram for identifying lymph node metastasis and prognostic prediction of breast cancer. The study will investigate the relationship between the radiomics and the tumor microenvironment. The study includes the construction of multiparametric MRI radiomics-based prediction model and the validation of the prediction model.

Inclusion Criteria:

• The primary lesion was diagnosed as invasive breast cancer
• Patients can have regional lymph node metastasis,but no distant organ metastasis
• Complete the breast MRI examination before treatment
• Accept breast cancer surgery or lymph node biopsy
• Eastern Cooperative Oncology Group performance status 0-2

Exclusion Criteria:

• Inflammatory breast cancer
• Accompanied with other primary malignant tumors
• Perform surgery,radiotherapy and lymph node biopsy before breast MRI examination
• Patients who have neoadjuvant chemotherapy
• Patients had distant and contralateral axillary lymph node metastasis
• The pathologic diagnosis was extensive ductal carcinoma in situ

Patients who had early stage breast cancer and completed the breast MRI examination before operation,lymph node biopsy,neoadjuvant chemotherapy,and radiotherapy.

Requests for the individual data or study documents will be considered where the proposed use aligns with public good purposes, does not conflict with other requests, and the requestor is willing to sign a data access agreement. Contact is though the corresponding author.