Artificial Intelligence Detects, Classifies, and Describes Lesions in Clinical Breast Ultrasound Images

I presented this work in a poster presentation during the 45th San Antonio Breast Cancer Symposium. My labmate Dustin Valdez also presented some collaborative work on predicting mammographic breast density from breast ultrasound imaging at this conference. In this work we showed preliminary results which show that an adapted Mask RCNN can segment and classify the cancer status and the US BI-RADS mass lexicon characteristics from clinical breast ultrasound imaging.