Shimming method for NMR magnet having large magnetic field inhomogeneities

• US 5,539,316 A
• Filed: 08/25/1995
• Issued: 07/23/1996
• Est. Priority Date: 08/25/1995
• Status: Expired due to Term

• Current Assignee: Bruker Instruments Incoporated (Bruker Corporation)
• Original Assignee: Bruker Instruments Incoporated (Bruker Corporation)
• Inventors: Sukumar, Subramaniam
• Primary Examiner(s): TOKAR, MICHAEL J
• Application Number:
  US08/519,656
• Time in Patent Office: 333 Days
• Field of Search: 324/300, 324/307, 324/309, 324/310, 324/311, 324/312, 324/313, 324/314, 324/318, 324/319, 324/320, 324/322, 128/653.2, 335/301, 335/216
• US Class Current: 324/320
• CPC Class Codes: G01R 33/3875 using correction coil assemblies, e.g. active shimming

Abstract

A method of shimming an NMR magnet uses a plurality of 1D projections through a sample volume to determine the inhomogeneities of the field of the NMR magnet. The frequency distributions obtained are assembled from the phase signals of the various projections. In order to avoid frequency errors due to phase wrapping, the phase of each signal is monitored over time for discontinuities indicative of aliasing. For each phase wrap, the signal is "unwrapped" by adjusting the value of the phase signal by 2π. The same method is used to establish a shim field map for each of the shim coils being used. With one shim coil at a time being driven with a predetermined current, the detection method is repeated to acquire a shim-base frequency map for each shim coil. The base frequency map is then subtracted from the shim-base frequency maps to obtain shim field maps. The proper shim currents are then acquired through matrix operations on the shim field maps and the base frequency map.

Claims

1. A method of determining the inhomogeneities in the magnetic field of a magnet which generates a magnetic field in a region containing a sample material, the method comprising:

A) generating an RF magnetic pulse which is incident on the sample material;

B) generating a phase encoded gradient pulse in a gradient direction which is incident on the sample;

C) detecting a first set of free induction decay signals s(t_r)₁ corresponding to a plurality of spatial points in the sample at a first time T₁ ;

D) detecting a second set of free induction decay signals s(t_r)₂ corresponding to said plurality of spatial points in the sample at a second time T₂ ;

E) transforming s(T₁)₁ and s(T₁)₂ from the time domain to image signal sets S(r)₁ and S(r)₂ in the frequency domain;

F) converting S(r)₁ and S(r)₂ to polar coordinate signals P_w (r)₁ and P_w (r)₂, respectively;

G) for each of P_w (r)₁ and P_w (r)₂, detecting a change in phase at each of a plurality of spatial points relative to a phase of a next previously detected point and, when said change in phase at a particular point exceeds a predetermined threshold indicative of a phase wrap, unwrapping the phase by adjusting the phase at said particular point by 2π

, the phase unwrapping for all the spatial points resulting in phase unwrapped signals P(r)₁ and P(r)₂ ;

H) subtracting P(r)₁ from P(r)₂ to obtain a phase difference signal P(r)_D ; and

I) acquiring a base frequency distribution, F_B, in the gradient direction by calculating F_B =(P(r)_D /(t₂ -t₁)) for each of said spatial points.

2. A method according to claim 1 further comprising performing each of steps A) through I) for a plurality of gradient directions.

3. A method according to claim 1 wherein the method is also a method of shimming a magnet by providing electrical currents through a plurality of magnetic shim coils, the electrical currents being selected such that the magnetic fields generated by the shim coils compensate for the determined inhomogeneities in the magnetic field of the magnet.

4. A method according to claim 3 further comprising performing each of steps A) through I) for a plurality of gradient directions.

5. A method according to claim 1 wherein step H) is performed before step G), such that the phase difference signal P_D (r) is obtained prior to unwrapping the phase.

6. A method according to claim 1 wherein the method is also a method of acquiring a shim frequency distribution in the sample as a result of a magnetic field generated by a shim coil, the method further comprising the steps of:

J) passing a first current through a shim coil;

K) repeating steps A) through I) to acquire a shim-base frequency distribution, F_SB, the shim-base frequency distribution being the base frequency distribution F_B acquired while the current is passed through the shim coil; and

L) determining the shim frequency distribution by subtracting the base frequency distribution, F_S, from the shim-base frequency distribution, F_SB.

7. A method according to claim 6 further comprising repeating steps J) through L) for each of a plurality of shim coils.

8. A method according to claim 7 wherein each execution of steps A) through I) comprises executing steps A) through I) for a plurality of gradient directions so as to acquire a plurality of projections, each projection being a frequency distribution acquired in one of said gradient directions, the method also being a method for shimming the magnetic field of the magnet and further comprising the steps of:

M) calculating the appropriate shim currents for compensating for inhomogeneities of the magnetic field of the magnet by calculating values for a vector v which minimizes the term A.v-b)², where A is a matrix of dimension c×

m, containing c of said shim frequency distributions each having m data points, and b is a vector containing m data points from the projections of the base frequency distribution F_B, said data points corresponding spatially to the datapoints of the shim maps; and

N) applying the shim currents to the appropriate shim coils.

9. A method of determining the frequency signal for a particular spatial point in the sample volume of an NMR apparatus, the method comprising:

measuring the time-domain free induction decay signal of the spatial point at a plurality times during a decay period;

determining the phase of the free induction decay signal of said spatial point as a function of time; and

taking the derivative of said phase as a function of time, and averaging said derivative over a predetermined time period.

10. A method according to claim 9 further comprising detecting anomalies in the phase signal, and discarding values of said derivative which correspond in time to said anomalies.

11. A method according to claim 9 further comprising detecting anomalies in said derivative, and discarding values of said derivative from said averaging which correspond in time to said anomalies.