Meeting Banner
Abstract #0626

DCE time-series characterization with supervised deep learning: Alternative to PK model approaches

Dattesh D Shanbhag1, Vivek Vaidya1, Uday Patil1, Sandeep N Gupta2, and Rakesh Mullick1

1GE Global Research, Bangalore, India, 2GE Global Research, Niskayuna, NY, United States

We demonstrate feasibility of using a supervised deep learning method with DCE time-series data to obtain consistent numerical cutoff for tumor regions. DL based characterization is robust to fluctuations in DCE data due to protocol and patient physiology differences, which typically hinders such a classification with PK maps in clinical practice.

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

Join Here