Meeting Banner
Abstract #1100

On the Feasibility of Noncontrast Valvular Cine MRI with High Spatial Resolution and High Frame Rate Using Deep-learning-powered Acceleration

Peng Lai1, Christopher M Sandino2, Shreyas S Vasanawala3, Anne Menini1, Haonan Wang4, Anja C.S Brau1, and Martin A Janich5
1GE Healthcare, Menlo Park, CA, United States, 2Electrical Engineering, Stanford University, Palo Alto, CA, United States, 3Radiology, Stanford University, Palo Alto, CA, United States, 4GE Healthcare, Waukesha, WI, United States, 5GE Healthcare, Munich, Germany

Valvular imaging is challenging to conventional cine MRI due to its requirement of very high spatial and temporal resolution. This work preliminarily investigated valvular cine MRI with highly accelerated data acquisition powered by deep learning reconstruction. Our results demonstrated the feasibility to resolve valve anatomy and motion with nearly 1mm spatial resolution and 10ms frame rate, while flow-induced dephasing generates shading in blood pool and can complicate valve visualization.

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

Join Here

Abstract Presentation Video