Meeting Banner
Abstract #2340

Integration of DCE-MRI Texture Features with Clinical Data for Improved Early Prediction of Breast Cancer Therapy Response

Archana Machireddy1, Guillaume Thibault1, Alina Tudorica1, Aneela Afzal1, May Mishal1, Kathleen Kemmer1, Arpana Naik1, Megan Troxell1, Eric Goranson1, Karen Oh1, Nicole Roy1, Neda Jafarian1, Megan Holtorf1, Wei Huang1, and Xubo Song1

1Oregon Health and Science University, Portland, OR, United States

This study investigated the effect of integrating clinical data with DCE-MRI texture features in early prediction of breast cancer therapy response. DCE-MRI data collected from 55 breast cancer patients before and after the first cycle of neoadjuvant chemotherapy were subjected to pharmacokinetic analysis. Texture features were extracted from voxel-based DCE-MRI parametric maps. Predictive performances with imaging features alone and in combination with clinical features were assessed and compared. Addition of clinical features to image texture features increased predictive capability in discriminating pathologic complete response (pCR) vs. non-pCR compared to using imaging features alone.

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

Join Here