Related Experiment Video
Updated: Jan 13, 2026
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
Preoperative Prediction of Axillary Lymph Node Metastasis in Breast Cancer Using Radiomics Features of Voxel-Wise
Ya Ren1, Kexin Chen2,3, Meng Wang1
1Department of Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen 518116, China.
This study demonstrates that a radiomics model using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) time-intensity-curve (TIC) profiles can effectively predict axillary lymph node (ALN) metastasis in breast cancer patients. The type-19-combined model showed the highest accuracy, offering a noninvasive approach for treatment guidance.
Area of Science:
- Oncology
- Radiology
- Medical Imaging
Background:
- Axillary lymph node (ALN) status is critical for breast cancer treatment and prognosis.
- Predicting ALN metastasis noninvasively is essential for personalized treatment strategies.
Purpose of the Study:
- To evaluate the efficacy of a radiomics model based on voxel-wise dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) time-intensity-curve (TIC) profiles for predicting ALN metastasis in breast cancer.
- To compare the performance of different radiomics feature sets derived from DCE-MRI.
Main Methods:
- Retrospective analysis of 615 breast cancer patients' preoperative DCE-MRI data.
- Segmentation of 3D lesions and categorization of voxels into 19 TIC subtypes.
- Derivation of three feature sets: type-19, type-19-radiomics, and phase-3-radiomics.
- Construction and evaluation of four predictive models (type-19, type-19-radiomics, type-19-combined, phase-3-radiomics) using support vector machines (SVM).
Main Results:
- The type-19-combined model achieved the highest area under the curve (AUC) of 0.779 in cross-validation and 0.674 in the testing set.
- The type-19-combined model significantly outperformed the phase-3-radiomics and type-19 models.
- The type-19-radiomics model also demonstrated superior performance compared to the phase-3-radiomics and type-19 models.
Conclusions:
- Radiomics analysis of voxel-wise DCE-MRI TIC profiles offers an effective, noninvasive method for predicting ALN metastasis in breast cancer.
- This approach quantifies temporal and spatial hemodynamic heterogeneity, aiding in treatment decisions.
