Browse Prior Art Database

AN ALGORITHM FOR FASTER VOLUMETRIC ACQUISITION IN MAGNETIC RESONANCE IMAGING

IP.com Disclosure Number: IPCOM000129733D
Publication Date: 2005-Oct-07
Document File: 7 page(s) / 1M

Publishing Venue

The IP.com Prior Art Database

Abstract

In an embodiment an algorithm for Faster Volumetric Acquisition in Magnetic Resonance Imaging is disclosed, which achieves saving in acquisition time by only acquiring data in an elliptical shape within rectangular space. It is established that corner of k-space might not need to be sampled fully to achieve the desired resolution. This invention enables the acquisition of Elliptical shape within k-space.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 33% of the total text.

AN ALGORITHM FOR FASTER VOLUMETRIC ACQUISITION IN MAGNETIC RESONANCE IMAGING

FIELD OF THE INVENTION

The Invention is generally in the field of Magnetic Resonance Imaging and specifically in the field of Volumetric Acquisition in MR.

BACKGROUND OF THE INVENTION

Bernstein et.al.[1] has demonstrated previously that corner of k-space might not need to be sampled fully to achieve the desired resolution.  In conventional 3D Cartesian sampling, phase encoding is performed on a rectangular grid along two directions ky and kz as depicted in figure 1.  There is significant time saving if the corner ky-kz-space is not fully sampled.  As shown in [1], sampling data within an ellipse enclosed inside the rectangle with k - k dimension can result in 1-p/4 (21.46%) saving in acquisition time.  However, there is currently no known general algorithm to achieve true elliptical sampling with various type of known phase encoding order such as linear (sequential), centric, elliptical centric, radial fan beam etc. 

DETAILED DESCRIPTION OF THE INVENTION

 This invention describes a simple yet robust algorithm that can be deployed to achieve true elliptical 3D Cartesian sampling.  Using k-space coordinate convention as shown in figure 1 below

Figure 1.  Eccentric polar angle q (counter clockwise angle from Kz axis) and maximum radius (from the origin) using conventional k-space coordinate r(q) = where a and b are the semi-major and semi-minor axis as described above.

 

  1. Compute the semi-major axis a and semi-minor axis b based on the desired overall sampling fraction. 

a. For sampling fraction (sf) of p/4 (78.53%), the semi-major axis a and semi-minor axis b can be chosen as k and k, respectively.

b.For sampling fraction (sf) differs from p/4, there are different possibilities for semi-major/minor axis setting

§          Semi-major axis a = k * and semi-minor axis b = k * i.e. both axis are scaled equally to achieve desired sampling fraction sf.

§          Semi-major axis a = k* (sf *4/ p)  and semi-minor axis b =  k i.e. scaled semi-major axis only.

§          Semi-major axis a = k  and semi-minor axis b =  k* (sf *4/ p)  i.e. scaled semi-minor axis only.

  1. Calculate the eccentric polar angle q for all encoding steps (kz ,ky) where q = tan-1 (ky/kz) .
  2. Calculate the Euclidean distance from center of k-space as for all encoding step (kz ,ky) = .
  3.  For each encoding step (kz, ky, q), compare the Euclidean distance to the maximum radius                 r(q) =   where a and b are the semi-major and semi-minor axis as described above.
  4. If the Euclidean distance is larger than that of maximum radius r(q) (i.e. point lies outside the ellipse with the desired dimension),  the phase encoding step will be flagged to be skipped over at run-time.
  5. For Elliptical Centric (EC) phase encoding order family  (EC, delay/recessed EC, reverse EC), the Euclidean distance of the skipped encoding step is temporarily changed to

a. +µ or very large value (>> ) for Elliptical Centric or de...