Deep Learning Radiomics Nomogram Based on Multiphase Computed Tomography for Predicting Axillary Lymph Node...

Area of Science:

• Radiology
• Oncology
• Artificial Intelligence

Background:

• Accurate prediction of axillary lymph node metastasis (ALNM) is crucial for breast cancer staging and treatment.
• Current methods may have limitations in precisely identifying ALNM.

Purpose of the Study:

• To develop and validate a deep learning radiomics nomogram (DLRN) for predicting ALNM in breast cancer.
• To assess the DLRN's predictive performance and clinical utility.

Main Methods:

• Retrospective analysis of 196 invasive breast cancer patients.
• Extraction of radiomics and deep learning features from CT scans.
• Construction of radiomics and deep learning signatures using support vector machines.

Main Results:

• Radiomics signature, deep learning signature, and clinical n stage were independent predictors of ALNM.
• The DLRN achieved an AUC of 0.893 in the validation set with good calibration.
• Decision curve analysis indicated superior clinical utility of the DLRN.

Conclusions:

• The DLRN demonstrates strong predictive value for ALNM in breast cancer.
• This tool offers valuable insights for tailoring individual patient treatment strategies.