Abstract #2321

Comparison of Breast Cancer Diagnostic Performance Using Radiomics Models Built Based on Features Extracted from DCE-MRI and Mammography

Jiejie Zhou¹, Yang Zhang², Kyoung Eun Lee³, Jeon-Hor Chen², Xiaxia He¹, Nina Xu¹, Shuxin Ye¹, Ouchen Wang¹, Jiance Li¹, Yezhi Lin⁴, Meihao Wang¹, and Min-Ying Su²

¹First Affiliate Hospital of Wenzhou Medical University, Wenzhou, China, ²University of California, Irvine, CA, United States, ³Inje University Seoul Paik Hospital, Seoul, Korea, Republic of, ⁴Wenzhou Medical University, Wenzhou, China

A total of 89 patients receiving both DCE-MRI and mammography were analyzed, including 56 malignant and 33 benign lesions. The 3D tumor mask on MRI was generated using computer algorithms. A total of 99 texture and histogram features were extracted from three DCE parameters maps. The suspicious area on mammography was outlined using MRI findings as guidance, and a similar radiomics method was applied to extract features from the mass and the margin. Random forest was applied to select features for building diagnostic models. The overall accuracy was 0.80 for MRI, 0.75 for mammography, and improved to 0.85 when combined.

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

Join Here