Method and apparatus for dynamic magnetic field control using multiple magnets

• US 7,495,537 B2
• Filed: 08/10/2006
• Issued: 02/24/2009
• Est. Priority Date: 08/10/2005
• Status: Active Grant

• Current Assignee: Stereotaxis, Inc.
• Original Assignee: Stereotaxis, Inc.
• Inventors: Tunay, Ilker
• Primary Examiner(s): Enad; Elvin G
• Assistant Examiner(s): TALPALATSKI, ALEXANDER
• Application Number:
  US11/502,335
• Publication Number:
  US 20070038410A1
• Time in Patent Office: 929 Days
• Field of Search: 335/306
• US Class Current: 335/306
• CPC Class Codes: A61B 5/06 Devices, other than using radiation, for detecting or locating foreign bodies for removing same A61B17/50 ; determining position of probes within or on the body of the patient
• CPC Class Codes: A61B 5/062 using magnetic field

Abstract

A method and apparatus for dynamic magnetic field control using multiple magnets. Control methods and system means are described that allow dynamically changing the magnetic field generated at a point in space by a multiplicity of magnets.

Claims

1. A method of determining the positions and orientations of at least two magnets to apply a magnetic field at an operating point of specified strength and direction, the method comprising:

(1) apportioning the contribution to the desired field of each of the at least two magnets;

(2) individually determining for each of the at least two magnets, a position and orientation to achieve the apportioned contribution using a representation of the magnetic field of each the magnet;

(3) simultaneously determining a position and orientation of each of the at least two magnets two magnets to achieve the desired field, using the individually determined positions and orientations and a representation of the magnetic field of each magnet.

2. The method of claim 1 wherein the representations of the magnetic field used in step (3) are different from the representations of the magnetic field used in step (2).

3. The method of claim 1 wherein step (2) comprises using a nested polynomial representation of the field of each magnet.

4. The method of claim 1 wherein step (3) comprises using a spherical harmonics expansion field representation of the field of each magnet.

5. The method of claim 1 wherein step (3) is performed as a simultaneous optimization of all degrees of freedom.

13. The method of claim 1 wherein the apportioning comprises distributing a target magnetic field at an operating point in space among the at least two magnets subject to design and application related constraints, whereby the sum of the magnetic fields contributed by the at least two magnets at the operating point is equal to the target magnetic field at the operating point.

14. The method of claim 13, further constraining the fields generated by each controlled magnet to be aligned with the target magnetic field and such that the target field magnitude is distributed in equal parts to all controlled magnets.

15. The method of claim 14, where adjustments are made to the field apportionments to account for each magnet lower and upper field magnitudes at the operating point.

6. A method of generating a specific target magnetic field at an operating point in space, the method comprising the steps of:

(a) representing magnet and application related constraints;

(b) defining a series of intermediate field vectors at the operating point;

(c) using an inversion method to find a series of method control parameters for a multiplicity of magnets associated with the series of field vectors of step (b);

(d) finding a time sequence of control parameters for the multiplicity of magnets of step (c) subject to the constraints of step (a) to approximate the series of control parameters of step (c); and

(e) performing the time sequence of step (d),

whereby the target magnetic field at the operating point is achieved at a given level of accuracy and within a specific time interval.

7. The method of claim 6, wherein the series of intermediate field vectors at the operating point lie in the plane formed by the initial and the target magnetic field vectors at the operating point.

8. The method of claim 6, wherein the time sequence is obtained by fitting the series of magnets control parameters of step (c) by representing the distance between the fit functions and the series as one of the Lagrangian terms.

16. The method of claim 6, wherein the inversion method comprises the steps of:

(a) using a magnetic field representation;

(b) representing magnet design and application related constraints;

(c) finding the magnet control parameters using the representation of step (a) and under the constraints of step (b) that will provide a magnetic field at an operating point in space,

whereby a target magnetic field at the operating point in space is obtained at a given level of accuracy.

17. The inversion method of claim 16, using a nested polynomial magnetic field representation and further comprising the steps of:

(a) pivoting the magnet by an angle Φ

such that the operating point lies on the Z^m magnet axis and with a plane of constant Y^m;

(b) calculating the rotation angle θ

such that the apportioned target magnetic field lies in a plane of constant Y^m in the magnet under consideration;

(c) calculating the target magnetic field angle β

with respect to the Magnet—

X^m axis;

(d) calculating the cylindrical coordinate angle α

of a point within the magnet such that the associated magnetic field generated by the magnet lies at an angle α

+β

with respect to magnet axis—

X^m;

(e) calculating the cylindrical coordinate ρ

of a point within the magnet at an angle α

+β

such that the magnet field magnitude at that point equals that of the apportioned field;

(f) calculating from planar geometry the magnet translation Z and pivoting φ

that will bring the target point to the desired position within the magnet for generation of the apportioned field at the operating point; and

(g) carrying out the rotation of angle θ

, pivoting of angle φ

, and translation along Z^m to bring the target point to the desired point within the magnet.

18. The inversion method of claim 16, using a spherical harmonics expansion representation for the magnetic field of each magnet and further comprising the step of formulating the problem as a non-linear constrained optimization method.

19. The inverse method of claim 18, further comprising the steps of:

(a) at each algorithm iteration, locally modeling non-linear cost functions via their linear approximations and including the square of the linear terms in a sub-problem optimization cost function;

(b) adding a quadratic term to the cost function that varies according to the square of the distance between the current iteration state vector x and the initial optimization state vector x_0;

(c) solving at each iteration step the local quadratic sub-problem using known techniques of constrained optimization; and

(d) iterating on steps (a) through (c) above till a solution to the initial optimization problem has been reached.

20. The method of claim 6, wherein the inversion method for the multiplicity of magnets comprises the steps of:

(a) using a multiplicity of magnetic field representations for each the magnets;

(b) representing magnet design and application related constraints;

(c) using a first inversion method in a first stage to find the magnets control parameters using a first level of magnetic field representation of step (a) and under the constraints of step (b) that will provide a magnetic field at an operating point in space at a first level of accuracy;

(d) using a second inversion method in a second stage, the magnet control parameters of step (c) as a starting point, and a second level of magnetic field representation of step (a), finding under the constraints of step (b) the magnet control parameters that will provide a magnetic field at the operating point in space at a second, higher, level accuracy,

whereby the target magnetic field at the operating point in space is achieved at a high level of accuracy and within a specified time interval by the multiplicity of magnets under control.

9. A method of representing the magnetic field generated by a magnet, the method comprising the steps of:

(a) providing a least a first level of magnetic field representation such that the magnet control parameters that will provide a specific magnetic field generated by the magnet at an operating point in space within a first level of accuracy can be found efficiently; and

(b) providing at least a second level of magnetic field representation such that the magnet control parameters that will provide a specific magnetic field generated by a multiplicity of magnets at an operating point in space at a second, higher, level of accuracy can be found efficiently; and

(c) designing the magnetic field representations of steps (a) and (b) such that the control parameters found by using the representations provided in step (a) for the multiplicity of magnets under consideration in step (b) will provide an initial point in the neighborhood of the final solution for the procedure of finding the control parameters using the representation provided in step (b);

whereby the control parameters for a multiplicity of magnets that will generate a specific magnetic field at an operating point in space within a second, higher level of accuracy subject to design and application related constraints can be found efficiently.

10. The method of claim 9, wherein the step (a) of providing at least a first level of magnetic field representation comprises the use of nested polynomials.

11. The method of claim 10, wherein the nested polynomials provide a one-to-one correspondence between a couple consisting of a magnetic field magnitude and magnetic field orientation within a magnet plane of constant coordinate Y^m and a triplet of magnet articulations consisting of a rotation angle θ

, a pivoting angle Φ

, and a translation distance Z.

12. The method of claim 9, wherein the step (b) of providing a second level of magnetic field representation comprises using a spherical harmonics function expansion, and calculating the expansion coefficients for optimal accuracy given a number of expansion terms by minimizing the representation error in a local volume;

said local volume always containing the operating point.