Guillaume Thibault1, Alina Tudorica2, Aneela Afzal2, Stephen Chui2, Arpana Naik2, Megan Troxell3, Kathleen Kemmer2, Karen Oh2, Nicole Roy2, Megan Holtorf2, Wei Huang2, and Xubo Song2
1BME, OHSU, Portland, OR, United States, 2OHSU, Portland, OR, United States, 3OHSU, portland, OR, United States
36 breast cancer patients underwent research
DCE-MRI before and after one cycle of neoadjuvant chemotherapy. 3D tumor
imaging texture features were extracted from parametric maps of quantitative
pharmacokinetic (PK) and semi-quantitative DCE-MRI parameters, and correlated
with pathologically measured post-therapy residual cancer burden (RCB).
Texture features from quantitative PK parameters were found to be more useful
than those from semi-quantitative metrics for early prediction of therapy
response, while the features from the SSM PK parameters were superior to the SM
counterparts for prediction of response.