Meeting Banner
Abstract #3923

Early Prediction of Soft Tissue Sarcoma Response to Preoperative Therapy Using DCE-MRI Texture Features

Tristan Xiao1, Archana Machireddy2, Xubo Song2, Alina Tudorica2, Aneela Afzal2, May Mishal2, Brooke Beckett2, Megan Holtorf2, Torrie Aston2, Christopher Ryan2, Wei Huang2, and Guillaume Thibault2

1Saratoga High School, Saratoga, CA, United States, 2OHSU, Portland, OR, United States

23 patients with soft tissue sarcoma (STS) (25 tumors) underwent DCE-MRI before and after one cycle of preoperative chemoradiotherapy. Extended Tofts model (ETM) and Shutter-Speed model (SSM) were used for pharmacokinetic (PK) analysis of DCE-MRI data and generating voxel based PK parametric maps, from which texture features were extracted using different statistical matrix methods. Changes in SZM and RLM features consistently provided good early prediction of therapy response, while more features from the SSM PK maps were good predictors of response than the ETM maps.

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

Join Here