Abstract #0573

Deep Generative Adversarial Neural Networks for Compressed Sensing (GANCS) Automates MRI

Morteza Mardani¹, Enhao Gong², Joseph Cheng³, Shreyas Vasanawala⁴, Greg Zaharchuk⁴, Lei Xing^1,5, and John Pauly⁶

¹Electrical Engineering, Stanford University, Stanford, CA, United States, ²Electrical Engineering, Stanford University, stanford, CA, United States, ³Stanford University, stanford, CA, United States, ⁴Radiology, Stanford University, Stanford, CA, United States, ⁵Radiation Oncology, Stanford University, Stanford, CA, United States, ⁶Stanford University, Stanford, CA, United States

MRI suffers from aliasing artifacts when undersampled for real-time imaging. Conventional compressed sensing (CS) is not however cognizant of image diagnostic quality, and substantially trade-off accuracy for speed in real-time imaging. To cope with these challenges we put forth a novel CS framework that permeates benefits from generative adversarial networks (GAN) to modeling a manifold of MR images from historical patients. Evaluations on a large abdominal MRI dataset of pediatric patients by expert radiologists corroborate that GANCS retrieves improved images with finer details relative to CS-MRI and deep learning schemes with pixel-wise costs, at 100 times faster speed than CS-MRI.

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

Join Here