Abstract #4877

Characterization of Sub-centimeter Enhancing Breast Masses on MRI with Radiomics and Machine Learning in BRCA Mutation Carriers

Roberto Lo Gullo¹, Isaac Daimiel ¹, Carolina Saccarelli¹, Almir Bitencourt¹, Peter Gibbs², michael James Fox², Sunitha B. Thakur^2,3, Danny F. Martinez¹, Elisabeth A. Mossir¹, and Katja Pinker¹

¹Radiology, Breast imaging, Memorial Sloan Kettering Cancer Center, New York, NY, United States, ²Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, United States, ³medical physics, Memorial Sloan Kettering Cancer Center, New York, NY, United States

The purpose of our study was to investigate whether radiomics features extracted from MRI of BRCA-positive patients with breast masses smaller than 1 cm coupled with machine learning can differentiate benign from malignant lesions using model-free parameter maps. We included 96 patients with 116 lesions assessed by two readers according to the BI-RADS lexicon. Radiomics features were calculated and included in a machine learning model, along with clinical factors, to discriminate between malignant and benign lesions. The machine learning model, combining two clinical and three radiomics features, achieved higher diagnostic accuracy (81.5%) compared to morphological assessment alone (53.4%)

This abstract and the presentation materials are available to members only; a login is required.

Join Here