Unipolar fast spin echo for permanent magnet MRI

• US 10,261,146 B2
• Filed: 04/30/2015
• Issued: 04/16/2019
• Est. Priority Date: 05/21/2014
• Status: Active Grant

• Current Assignee: Aspect Imaging Ltd.
• Original Assignee: Aspect Imaging Ltd.
• Inventors: Shapira, Boaz, Rapoport, Uri
• Primary Examiner(s): Phan, Huy Q
• Assistant Examiner(s): Clarke, Adam
• Application Number:
  US14/700,207
• Publication Number:
  US 20150338486A1
• Time in Patent Office: 1,447 Days
• CPC Class Codes: G01R 33/383 using permanent magnets
• CPC Class Codes: G01R 33/5617 using RF refocusing, e.g. RARE
• CPC Class Codes: G01R 33/56563 caused by a distortion of the main magnetic field B0, e.g. temporal variation of the magnitude or spatial inhomogeneity of B0 G01R33/56509, G01R33/56518, G01R33/56536 take precedence

Abstract

A method of reducing artifacts produced during Fast Spin Echo measurements made using permanent magnet NMR instruments. The method includes applying encoding gradients that do not switch signs throughout the experiment. Prior to the 90° RF pulse, a strong RM gradient pulse is given to produce a dominant and constant residual magnetization. The encoding is done through the combination of encoding gradients with the aid of the 180° RF pulses of the echo train. A first constant encoding gradient is given before the first 180 pulse. Then two variable encoding gradients are provided after each 180 pulse; one applied prior to and one applied subsequent to each acquisition in the echo train.

Claims

1. A method for reducing artifacts produced during a Fast Spin Echo (FSE) sequence in a nuclear magnetic resonance instrument comprising a gradient power amplifier and a permanent magnet comprising a plurality of pole pieces, said permanent magnet characterized by a gradient-dependent residual magnetization B_g, said FSE pulse sequence comprising, in order:

a 90°

radio frequency (RF) pulse characterized by a duration of the 90°

RF pulse (T₉₀);

an echo train comprising a first 180°

RF pulse, a first echo, X−

1 additional 180°

RF pulses and X−

1 additional echoes, where X≥

1; and

,

a navigator pulse;

wherein said method comprises;

applying a residual magnetization (RM) gradient pulse prior to said 90°

RF pulse, said RM gradient pulse having an amplitude G_RM^pe along a phase axis and a positive or a negative sign relative to an origin along said phase axis;

applying a shim gradient pulse of amplitude at least sufficient to cancel any net effect on said gradient-dependent residual magnetization B_g produced by said RM gradient pulse;

applying, prior to said first 180°

RF pulse, a single preparation gradient pulse having an amplitude G_shift^pe that is less than or equal to the RM gradient pulse amplitude G_RM^pe, whereby for a gradient pulse duration T, a k-value of the single preparation gradient pulse (k_shift^pe) is equal to the single preparation gradient pulse amplitude G_shift^pe times the gradient pulse duration T along said phase axis;

applying, for each X−

1 additional 180°

RF pulses i (1≤

i≤

X) in said echo train;

a first encoding gradient pulse applied subsequent to said 180°

pulse i and prior to echo i, said first encoding gradient pulse having an amplitude G_ai^pe such that a k-value for the first encoding gradient pulse k_ai^pe=G_ai^pe times T along said phase axis; and

,

a second encoding gradient pulse applied subsequent to said echo i and prior to said 180°

pulse i+1, said second encoding gradient pulse G_bi^pe having an amplitude G_bi^pe such that a k-value for the second encoding gradient pulse k_bi^pe=G_bi^pe times T along said phase axis, said second encoding gradient G_bi^pe having an identical sign to the sign of said RM gradient pulse; and

,

applying, subsequent to said navigator pulse, a final gradient pulse, said final gradient pulse having a sign identical to the sign of said RM gradient pulse;

obtaining a magnetic resonance images based on the FSE sequence.

2. The method according to claim 1, wherein said preparation gradient pulse is characterized by k^pe_shift≥

k^pe_max/2, where k^pe is being incremented between [−

k^pe_max/2, k^pe_max/2] k^pe_max is defined by the required spatial resolution, dr^pe, along the phase axis, such as;

k^pe_max=1/dr^pe.

3. The method according to claim 1, wherein for 1≤

i≤

X,k^pe_ai=G^pe_aiT=k^pe_shift+k^pe_i, where k^pe_ai, is the imposed k value after the gradient, G^pe_ai.

4. The method according to claim 1, wherein for 1≤

i≤

X,k^pe_bi=G^pe_biT=k^pe_shift−

k^pe_i where k^pe_bi, is the imposed k value after the gradient, G^pe_bi.

5. The method according to claim 1, wherein said final gradient pulse is characterized by k=k_shift^pe.

6. The method according to claim 1, wherein said FSE sequence is a 3D FSE sequence comprising:

an embedded 2D FSE sequence;

an encoding axis k^pe2 along said slice axis;

N_pe2 increments along said slice axis;

j increments of k^pe2, 1≤

j≤

N_pe2, the j^th increment of k^pe2 (k_j^pe2) being restricted to values in the range of

7. The method according to claim 6, wherein

8. The method according to claim 6, wherein k_aj^pe2=G_aj^pe2=k_shift^pe2+k_j^pe2.

9. The method according to claim 6, wherein k_bj^pe2=k_shift^pe2−

k_j^pe2.

10. The method according to claim 6, wherein said final slice gradient pulse is characterized by k=k_shift^pe2.