Related Experiment Video
Updated: May 28, 2025
08:32
Using Computer-based Image Analysis to Improve Quantification of Lung Metastasis in the 4T1 Breast Cancer Model
Published on: October 2, 2020
6.1K
A Radiomic and Clinical Data-Based Risk Model for Malignancy Prediction of Breast BI-RADS 4A Microcalcifications
Nicole Brunetti1, Cristina Campi2, Michele Piana2
1Department of Radiology, IRCCS - Ospedale Policlinico San Martino, Genoa, Italy; Department of Experimental Medicine (DIMES), University of Genova, Genoa, Italy.
Clinical Breast Cancer
|February 12, 2025
Summary
Radiomics combined with clinical data significantly improves the classification of benign and malignant microcalcifications in mammography, potentially reducing unnecessary biopsies for BI-RADS 4A findings.
Area of Science:
- Medical Imaging
- Radiology
- Oncology
Background:
- Mammography is crucial for breast cancer screening but struggles with specificity for microcalcifications.
- Radiomics offers a promising approach to improve lesion risk stratification.
- This study evaluates radiomics for classifying microcalcifications to aid radiological assessment.
Purpose of the Study:
- To assess the reliability of radiomics and clinical data in differentiating benign from malignant microcalcifications.
- To enhance the accuracy of standard radiological assessments.
- To explore the potential reduction in unnecessary biopsies.
Main Methods:
- Retrospective analysis of 167 BI-RADS 4A microcalcification cases (January 2019 - February 2023).
- Extraction of 104 radiomics features from mammographic images.
- Integration of clinical data and Tyrer-Cuzick risk assessment.
- Development of logistic regression models using clinical and radiomics data.
Main Results:
- A refined dataset of 14 radiomics features was utilized.
- The radiomics-only model achieved an AUC of 0.72.
- The combined clinical and radiomics model demonstrated superior performance with an AUC of 0.81.
Conclusions:
- Integrating clinical data with radiomics enhances the classification of BI-RADS 4A microcalcifications.
- This combined approach shows potential for reducing unnecessary biopsies.
- Personalized patient care can be improved through more accurate risk stratification.
