METHOD OF REMOVING DYNAMIC NONLINEAR PHASE ERRORS FROM MRI DATA

Patent №

US 6,853,191

Granted

2005-02-08

Filed 2003

Owner

BOARD OF TRUSTEES OF THE LELAND STANFORD JUNIOR UNIVERSITY THE

AI components

2

vision · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

10733011

Disclosed is a generalized reconstruction method that corrects for non-linear phase errors based on least-squares estimation. An approximation of the least squares estimate utilizes refocusing reconstruction in which high-resolution data is multiplied by the phase conjugate of a navigator in image-space. The multiplication rephases the unaliased signal in the high-resolution data. The high-resolution data can then be added together coherently. The multiplication can be effected in k-space as a convolution using a gridding reconstruction of the high-resolution data using the low resolution navigator.

VisionAI hardwareG01R 33/56509G01R 33/5676

AI classification

Vision0.91
AI hardware0.81
Natural language0.00
Planning0.00
Machine learning0.00
Knowledge representation0.00
Speech0.00
Evolutionary computation0.00

Ownership

BOARD OF TRUSTEES OF THE LELAND STANFORD JUNIOR UNIVERSITY THE

assignment · 147920400

Assignors

MILLER, KARLA L., PAULY, JOHN M.

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC