An open MRI passive shimming design method

Abstract

The application discloses a design method of an open MRI passive shimming, and first, raw data is subjected to matrix processing through matlab; coordinate values of target points in a target region are calculated through matlab; a magnetic field value generated by a unit thickness magnetic sheet in the target region is calculated through the position of the magnetic sheet, and a sensitivity matrix is calculated; finally, a non-uniformity value is set, and the position and thickness of the magnetic sheet are calculated through an optimization algorithm. The initial solution of the shimming sheet is calculated through the above-mentioned mixed linear programming method, and the genetic algorithm is used for optimization, and the optimal solution is searched in the vicinity of the initial solution, so that the requirement of the MRI on the uniformity of the main magnetic field is met, and the uniformity of the passive shimming algorithm is improved.

Description

TECHNICAL FIELD

[0001] The application relates to a design method of an open MRI passive shimming. BACKGROUND

[0002] A main magnetic field is a basic field of magnetic resonance imaging, and mainly provides a basic field required for imaging. In magnetic resonance imaging (MRI), the higher the main magnetic field uniformity, the better the image quality.

[0003] There are passive shimming and active shimming in a shimming method. Gradient coil shimming belongs to active shimming, and belongs to first-order active shimming. The first-order active shimming cannot meet the uniformity requirement of the main magnetic field of MRI. In order to achieve high uniformity, a high order is required, and the higher the order, the more complex the coil shape is. The passive shimming mainly places permanent magnet pieces with appropriate thickness at appropriate positions in upper and lower pole plates, so that a magnetic field value with extremely high uniformity can be realized.

[0004] In order to solve the problem of the main magnetic field uniformity, the application provides a design method of an open MRI passive shimming, and a two-stage optimization method is provided, that is, an initial solution of the shimming piece is calculated by using a mixed linear programming method, and then the genetic algorithm is optimized to search for an optimal solution near the initial solution, so that the uniformity of the passive shimming algorithm is improved. SUMMARY

[0005] The application aims to provide a design method of an open MRI passive shimming, and a two-stage optimization method is provided, that is, an initial solution of the shimming piece is calculated by using a mixed linear programming method, and then the genetic algorithm is optimized to search for an optimal solution near the initial solution, so that the uniformity of the passive shimming algorithm is improved.

[0006] In order to achieve the above-mentioned purpose, the application provides a design method of an open MRI passive shimming, and the steps are as follows:

[0007] Step 1: according to the text magnetic field data of the target point before shimming, the data is imported into matlab for matrix processing, and the 5th order spherical harmonic coefficient value of the magnetic field data on the spherical surface is calculated.

[0008] Step 2: according to the position and size of the shimming plate, the position matrix of the target point is calculated by matlab.

[0009] Step 3: the magnetic field contribution value of the unit height magnetic piece to the target point is calculated, that is, the sensitivity coefficient matrix is calculated, and the formula is as follows:

[0010] In the formula, Bzi is the target point magnetic field value, Ai and ti are the area and thickness of the magnetic piece respectively, mz is the magnetization intensity of the magnetic piece, n and m are the order and series of Legendre functions, P is the Legendre function, and theta is the theta value of the target point in the target spherical coordinate system. In the target point's spherical coordinate system The value is P(r,θ,). ), The angle values ​​are in the polar coordinate system of the magnetic sheet, and the position of the magnetic sheet is Q(f, α, ...). The distance f from the spherical target point to the magnetic plate on the pole disk is calculated using the following formula:

[0011]

[0012]

[0013] In the formula, z is the perpendicular distance between the magnetic sheet and the center of the target sphere, and r is the distance between the magnetic sheet and the axis of the pole disk.

[0014] Step 4: Construct the sensitivity coefficient matrix Mkj, which has k rows and j columns, representing the magnetic field strength of the k-th shim plate at the j-th measurement point in the target area.

[0015] Step 5: Optimize the core function

[0016]

[0017] In the formula, Bm is the magnetic field value measured before shimming, Bt is the target magnetic field strength, and M is the sensitivity coefficient matrix.

[0018] Step Six: Calculate the preliminary optimal solution using Mixed Integer Linear Programming (MILP). Given that x is the thickness of the magnetic sheet, an integer between 0mm and 4mm, and f is a column vector with all elements equal to 1, satisfying the following formula:

[0019]

[0020] In the formula, Bmax and Bmin are the maximum and minimum values ​​of the measured magnetic field, respectively, and ε is the allowable error value;

[0021] M is the sensitivity matrix, Meq is the coefficient of the fifth-order spherical harmonic matrix, and Beq is the coefficient of the fifth-order spherical harmonic matrix of the target magnetic field; the value range of x is [0,1,2,3,4].

[0022] Step 7: Perform secondary optimization using a genetic algorithm. In the vicinity of the MILP optimization solution [xi-t, xi+t], find the optimal integer solution and compare it with the target value to determine whether the design requirements are met. If the design requirements are met, stop the iteration; otherwise, continue to modify the relevant parameters, including coil thickness, coil width, and coil spacing.

[0023] The beneficial effects of this invention are as follows:

[0024] A two-stage optimization method is proposed: first, a hybrid linear programming method is used to calculate the initial solution of the shimming patch, and then a genetic algorithm is used to optimize the solution by searching for the optimal solution near the initial solution, which improves the uniformity of the passive shimming algorithm.

[0025] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0026] Figure 1 This is a design flowchart of the present invention.

[0027] Figure 2 This is a schematic diagram of the placement of the field uniform plate of the present invention.

[0028] Figure 3 This is a schematic diagram of the magnetic field measurement point of the present invention. Detailed Implementation

[0029] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0030] Figure 1 This is a design flowchart of the present invention; Figure 2 This is a schematic diagram of the electrode head uniformizing plate placement of the present invention, including upper and lower electrode plates and upper and lower uniformizing plates; the uniformizing plates are fixed to the electrode head, and the uniformizing plates have circular holes of different diameters distributed on them. The uniformizing plate is 5mm thick; the uniformizing plate has a small hole in the middle with a diameter of 5mm, mainly used for uniformizing the 1-layer area. The hole diameter gradually increases from the center to the edge of the uniformizing plate, mainly used for uniformizing layers 2-7. The uniformizing plate for the 7-layer layer is the largest, with a diameter of 20mm. The uniformizing plate for the 6-layer layer has a diameter of 15mm; the uniformizing plates for layers 2-5 have a diameter of 10mm. Figure 3 This is a schematic diagram of the magnetic field measurement points of the present invention. The diameter of the uniform field sphere is Ø400mm, with a total of 31 measurement points in the circumferential direction and 16 measurement points in the height direction, for a total of 496 measurement points.

[0031] The specific design steps of this embodiment are as follows.

[0032] Step 1: Import the text magnetic field data of the target point before shimming into MATLAB for matrix processing, and calculate the 5th order spherical harmonic coefficients of the magnetic field data on the sphere.

[0033] Step 2: Calculate the position matrix of the target point using MATLAB based on the position and dimensions of the shimming plate.

[0034] Step 3: Calculate the magnetic field contribution of the magnetic sheet per unit height to the target point, i.e., calculate the sensitivity coefficient matrix, as shown in the following formula:

[0035] In the formula, Bzi is the magnetic field value at the target point, Ai and ti are the area and thickness of the magnetic sheet, respectively, and mz is the magnetization intensity of the magnetic sheet; n and m are the order and series of the Legendre function. P is the Legendre function, and θ is the value of θ in the target spherical coordinate system. In the target point's spherical coordinate system The value is P(r,θ,). ), The angle values ​​are in the polar coordinate system of the magnetic sheet, and the position of the magnetic sheet is Q(f, α, ...). The distance f from the spherical target point to the magnetic plate on the pole disk is calculated using the following formula:

[0036]

[0037]

[0038] In the formula, z is the perpendicular distance between the magnetic sheet and the center of the target sphere, and r is the distance between the magnetic sheet and the axis of the pole disk.

[0039] Step 4: Construct the sensitivity coefficient matrix Mkj, which has k rows and j columns, representing the magnetic field strength of the k-th shim plate at the j-th measurement point in the target area.

[0040] Step 5: Optimize the core function

[0041]

[0042] In the formula, Bm is the magnetic field value measured before shimming, Bt is the target magnetic field strength, and M is the sensitivity coefficient matrix.

[0043] Step Six: Calculate the preliminary optimal solution using Mixed Integer Linear Programming (MILP). Given that x is the thickness of the magnetic sheet, an integer between 0mm and 4mm, and f is a column vector with all elements equal to 1, satisfying the following formula:

[0044]

[0045] In the formula, Bmax and Bmin are the maximum and minimum values ​​of the measured magnetic field, respectively, and ε is the allowable error value;

[0046] M is the sensitivity matrix, Meq is the coefficient of the fifth-order spherical harmonic matrix, and Beq is the coefficient of the fifth-order spherical harmonic matrix of the target magnetic field; the value range of x is [0,1,2,3,4].

[0047] Step 7: Perform secondary optimization using a genetic algorithm. In the vicinity of the MILP optimization solution [xi-t, xi+t], find the optimal integer solution and compare it with the target value to determine whether the design requirements are met. If the design requirements are met, stop the iteration; otherwise, continue to modify the relevant parameters, including coil thickness, coil width, and coil spacing.

[0048] In this embodiment, the open MRI utilizes the passive shimming method provided by the present invention, which significantly improves the uniformity of the main magnetic field. Specific experimental data are shown in the table below.

[0049]

[0050] Therefore, this invention provides a design method for passive shimming in open MRI, proposing a two-stage optimization method: first, a hybrid linear programming method is used to calculate the initial solution of the shimming patch, and then a genetic algorithm is used to optimize the search for the optimal solution near the initial solution, thereby improving the uniformity of the passive shimming algorithm.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A design method for passive shimming in open MRI, characterized in that: Based on the magnetic field data before shimming, the shimming positions of the upper and lower pole disks, and the allowable thickness range, the thickness and position of the magnetic sheet were calculated using a hybrid integer programming and genetic algorithm optimization method. The specific steps are as follows: Step 1: Based on the text magnetic field data of the target point before shimming, import it into MATLAB for matrix processing and calculate the 5th order spherical harmonic coefficients of the magnetic field data on the sphere. Step 2: Calculate the position matrix of the target point using MATLAB based on the position and dimensions of the shimming plate; Step 3: Calculate the magnetic field contribution of the magnetic sheet per unit height to the target point, i.e., calculate the sensitivity coefficient matrix, as shown in the following formula: In the formula, Bzi is the magnetic field value at the target point, Ai and ti are the area and thickness of the magnetic sheet, respectively, mz is the magnetization of the magnetic sheet; n and m are the order and series of the Legendre function, P is the Legendre function, and θ is the value of θ in the target point's spherical coordinate system. In the target point's spherical coordinate system The value is P(r,θ,). ), The angle values ​​are in the polar coordinate system of the magnetic sheet, and the position of the magnetic sheet is Q(f, α, ...). The distance f from the spherical target point to the magnetic plate on the pole disk is calculated using the following formula: In the formula, z is the perpendicular distance between the magnetic sheet and the center of the target sphere, and r is the distance between the magnetic sheet and the axis of the pole disk; Step 4: Construct the sensitivity coefficient matrix Mkj, which has k rows and j columns, representing the magnetic field strength of the k-th shim plate at the j-th measurement point in the target area; Step 5: Optimize the core function In the formula, Bm is the magnetic field value measured before shimming, Bt is the target magnetic field strength, and M is the sensitivity coefficient matrix. Step 6: Calculate the preliminary optimal solution using Mixed Integer Linear Programming (MILP). Given that x is the thickness of the magnetic sheet, an integer between 0mm and 4mm; and f is a column vector with all elements equal to 1. Satisfy the following formula: In the formula, Bmax and Bmin are the maximum and minimum values ​​of the measured magnetic field, respectively, and ε is the allowable error value; M is the sensitivity matrix, Meq is the coefficient of the fifth-order spherical harmonic matrix, and Beq is the coefficient of the fifth-order spherical harmonic matrix of the target magnetic field; the value range of x is [0,1,2,3,4]. Step 7: Perform secondary optimization using a genetic algorithm. In the vicinity of the MILP optimization solution [xi-t, xi+t], find the optimal integer solution and compare it with the target value to determine whether the design requirements are met. If the design requirements are met, stop the iteration; otherwise, continue to modify the relevant parameters, including coil thickness, coil width, and coil spacing.

Images