A Transparent Machine Learning Approach Using Ultrasound Radiomics for Preoperative Distinction Between Granulomatous Lobular Mastitis and Breast Cancer

1

A machine learning model was developed to differentiate granulomatous lobular mastitis from breast cancer using ultrasound radiomics.
2

The study analyzed 237 patients, extracting 1,161 radiomic features from ultrasound images to construct predictive models.
3

The combined model integrating clinical factors and radiomics achieved an AUC of 0.935 in training and 0.833 in validation cohorts.
4

Decision curve analysis indicated the combined model's favorable clinical applicability for preoperative differentiation.
5

The methodology enhances diagnostic precision and supports clinical decision-making in distinguishing between GLM and breast cancer.

Frontiers In Oncology

by Jinhong Zhou, Zhongcun Lai, Cishun Yu, Dan Liu, Yuguo Wei, Xiaowei Han, Guozheng Zhang
April 24, 2026

1