A Mammography-Based Radiomic Nomogram for Predicting Malignancy in Breast Suspicious Microcalcifications

Area of Science:

• Radiology
• Medical Imaging
• Oncology

Background:

• Accurate preoperative identification of breast lesions is crucial for individualized treatment.
• Distinguishing benign from malignant breast microcalcifications (MCs) remains a clinical challenge.

Purpose of the Study:

• To develop and validate a mammography-based radiomic nomogram for predicting the malignant risk of breast suspicious MCs.
• To improve preoperative diagnostic accuracy for breast lesions.

Main Methods:

• Radiomics features were extracted from mammography images of 496 patients.
• A radiomics score (Rad-score) was calculated using the least absolute shrinkage and selection operator algorithm.
• A clinical-radiomics model was built incorporating the Rad-score and clinical factors, visualized as a nomogram.

Main Results:

• A nomogram was developed using 29 radiomics features, menopause status, MCs morphology, and distribution.
• The combined model achieved an area under the curve of 0.926 in the validation set.
• The model demonstrated favorable calibration and clinical utility.

Conclusions:

• The developed clinical-radiomics nomogram is a potential imaging marker for predicting malignant risk in breast suspicious MCs.
• This tool can aid in personalized treatment strategies for breast cancer patients.