Meeting Banner
Abstract #0128

Improving K-T Auto-Calibrating Parallel Imaging for 3D Cardiac Cine MRI Using Prior-Reconstruction Static Tissue Estimation and Elimination

Peng Lai1, Shreyas S. Vasanawala2, Atsushi Nozaki3, Maggie Fung4, Anja C.S Brau5

1MR Applications & Workflow , GE Healthcare, Menlo Park, CA, United States; 2Radiology, Stanford University, Stanford, CA, United States; 3MR Applications & Workflow, GE Healthcare, Asahigaoka, Hino, Japan; 4MR Applications & Workflow, GE Healthcare, Jersey City, NJ, United States; 5MR Applications & Workflow, GE Healthcare, Garching, Munchen, Germany

High acceleration needed for 3D cine MRI potentially results in residual artifacts. High density coils and k-t acceleration methods demand long reconstruction time. In this work, we developed a method that automatically estimates and eliminates static tissue signals from k-t accelerated 3D cine datasets. This method enables k-t reconstruction on dynamic tissue signals with reduced aliasing and furthermore enables selective reconstruction on locations and coil channels around the heart only. Based on our evaluations, the proposed method can improve 3D cardiac cine MRI in both image quality and computation efficiency.

Abstract PDF Presentation Video

Keywords

accelerated acceleration acquisition adaptive adding address aliasing aorta applications around arterial artifacts auto back blurring calibrating cardiac central challenging channel channels chest cine city clinical coil coils collected combined combining computation condition containing content contribute converted correlation curvature cycle datasets decrease degrades density detection diagnosis differentiate dynamic effective effectively elements eliminated elimination entire equivalent especially estimation evaluated exploiting extended faster filter final flicking furthermore generated generates grayed healthy heart highly hybrid implementation improve improves increasing index induces intensive japan limited location locations losing loss materials mostly motion near next noise original parallel peripheral perspectives pixel prior processing produces proportion proposed quality reconstructed reconstruction recover reduce reduces reduction reformatted removes representative requires residual resolution resolved respectively sacrificing scheme scouting selected selection selective selectively sensitivity sequentially significantly slice slices space spatial square static step subsequent subsequently substantially subtracted synthesis thereby thick thin though thread threshold tissue tissues transition validated variation visually volume whole