Meeting Banner
Abstract #4337

Whole-Tumor Histogram Analysis of Multiparametric MRI for Subtype Classification of Breast Cancer

Tianwen Xie1, Qiufeng Zhao2, Robert Grimm3, Caixia Fu4, Yajia Gu1, and Weijun Peng1

1Department of Radiology, Fudan University Shanghai Cancer Center, Shanghai, China, 2Department of Radiology, Longhua Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, China, 3MR Application Predevelopment, Siemens Healthcare, Erlangen, Germany, 4Application Development, Siemens Shenzhen Magnetic Resonance, Shenzhen, China

Recently, several studies have shown the value of Magnetic Resonance Imaging (MRI) radiomics in non-invasive lesion subtype classification. In this study, we proposed the use of histogram texture features of multiparametric maps to differentiate subtypes of breast cancer. 34 different whole-tumor histogram features were analyzed. Classification was performed between ER-positive and Triple-negative groups resulted in AUROC of 0.94, while classification between ER-positive and HER2-positive groups, and classification between HER2-positive and Triple-negative yielded AUROC of 0.79, and 0.86, respectively.

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

Join Here