Feedback circuit and noise reduction method for electromagnetic interference signals of non-electromagnetic shielding magnetic resonance imaging equipment

Abstract

The present invention relates to a feedback circuit for electromagnetic interference signals of non-electromagnetic shielding magnetic resonance imaging equipment, belonging to the field of magnetic resonance technology. The circuit comprises: an electromagnetic noise detection sensor, used for real-time acquisition of electromagnetic noise signals on the surface of a human body; an impedance matching circuit, used for matching the impedance of human skin; a preamplifier circuit, used for amplifying filtered electromagnetic noise signal data; a gain adjustment circuit, comprising a gain amplifier circuit and a control circuit, wherein the gain amplifier circuit is used for gain amplification; the control circuit is used for real-time calculation of the gain multiple of the gain amplifier circuit according to the magnitude of the acquired noise voltage, and for controlling the dynamic gain adjustment of the gain amplifier circuit; an inverter, used for converting the amplified signal into a voltage signal of equal magnitude and opposite phase, and outputting the signal to the surface of the human body for noise reduction; and a signal source, used for injecting a signal of specific frequency and amplitude into each branch at an injection point, and performing detection at other detection points.

Description

Technical Field

[0001] The present invention belongs to the field of magnetic resonance technology, and relates to a feedback circuit and a noise reduction method for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device. Background Art

[0002] At present, magnetic resonance imaging systems need to work in a special electromagnetic shielding environment. The existence of electromagnetic shielding facilities blocks the impact of external electromagnetic interference on the system and improves the image signal-to-noise ratio. However, the existence of electromagnetic shielding facilities also greatly increases the requirements of the magnetic resonance imaging system for the use environment and limits the use scenarios of existing magnetic resonance imaging systems. In order to apply magnetic resonance imaging systems to clinical sites, such as bedside monitoring in intensive care units, clinical departments in hospitals with needs, and ambulances before examinations, one of the key issues is to remove the electromagnetic shielding device and realize its use in open scenarios. This requires solving the problem of active noise reduction of electromagnetic noise introduced under open conditions.

[0003] However, the environment in which MRI systems operate presents numerous factors that can affect the stability of the magnetic field in the scanning area, including electromagnetic interference (EMI) generated by 50Hz power frequency currents, large electrical equipment, and electrostatic coupling between the human body and wires. These EMIs cannot be shielded by currently used shielding devices, and their intensity is far weaker than the magnetic field of the MRI magnet itself. Therefore, it is impossible to obtain highly accurate interference data by directly measuring the magnetic field in the scanning area, and it is impossible to eliminate the impact of these interferences on MRI imaging quality. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a feedback circuit and noise reduction method for electromagnetic interference signals of non-electromagnetic shielding magnetic resonance imaging equipment. Based on common-mode voltage detection, the common-mode voltage is extracted, reversely amplified, and then fed back to the human body. The human body common-mode signal is superimposed on this reverse-amplified common-mode signal and is suppressed, thereby offsetting the original noise.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] In one aspect, the present invention provides a feedback circuit for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device, comprising:

[0007] Electromagnetic noise detection sensor, used to collect electromagnetic noise signals on the human body surface in real time;

[0008] An impedance matching circuit, for matching human skin impedance, with an input end electrically connected to the electromagnetic noise detection sensor and an output end electrically connected to the preamplifier circuit;

[0009] A preamplifier circuit amplifies the filtered electromagnetic noise signal data, wherein the input end of the preamplifier circuit is electrically connected to the impedance matching circuit, and the output end of the preamplifier circuit is electrically connected to the gain adjustment circuit;

[0010] The gain adjustment circuit includes a gain amplifier circuit and a control circuit. The gain amplifier circuit is used to perform gain amplification, and its input end is electrically connected to the preamplifier circuit, and its output end is electrically connected to the inverter. The control circuit is used to calculate the gain multiple of the gain amplifier circuit in real time based on the collected noise voltage and control the dynamic gain adjustment of the gain amplifier circuit. The input end is electrically connected to the preamplifier circuit, and the output end is connected to the gain amplifier circuit.

[0011] The inverter, whose input end is electrically connected to the gain adjustment circuit, converts the amplified signal into a voltage signal of equal magnitude and opposite phase, and outputs it to the human body surface for noise reduction;

[0012] The signal source has an input end electrically connected to the gain adjustment circuit and an output end connected to the human body surface, and is used to inject a signal of specific frequency and amplitude into each branch at the injection point and perform detection at other detection points.

[0013] Furthermore, the impedance matching circuit is composed of resistors R4 and R8 and capacitors C2 and C3. One end of the resistors R4 and R8 is used to receive the electromagnetic signal of the sensor 1, and the other end thereof is electrically connected to the gain amplifier in the gain amplifier circuit.

[0014] Furthermore, the preamplifier circuit is composed of resistors R2 and R9 and an operational amplifier U3, the resistor R2 is connected to the positive pole of the operational amplifier, one end of the resistor R9 is connected to the negative pole of the operational amplifier, and the other end is grounded; the signal collected by the electromagnetic noise detection sensor passes through the resistor to obtain a voltage, and the voltage passes through the operational amplifier circuit to amplify the signal.

[0015] Furthermore, the control circuit is composed of a multi-channel analog-to-digital converter ADC, a microprocessor, and a multi-channel digital-to-analog converter DAC, and the gain amplifier circuit is composed of a resistor and a gain amplifier; the voltage detected by the electromagnetic noise detection sensor is amplified by the preamplifier and output to the gain amplifier and the multi-channel analog-to-digital converter. The multi-channel analog-to-digital converter ADC collects the voltage signal and sends it to the microprocessor. The microprocessor calculates the gain amplifier multiple in real time according to the size of the collected noise voltage, and then controls the output DC voltage value of the digital-to-analog converter DAC to realize dynamic adjustment of the gain of the gain amplifier.

[0016] Furthermore, the inverter is composed of resistors R6 and R10 and an operational amplifier U2, one end of the resistor R6 is electrically connected to the gain amplifier, and the other end is connected to the reverse input terminal of the operational amplifier U2; one end of the resistor R10 is connected to the resistor R6, and one end is connected to the output terminal of the operational amplifier U2. The amplified signal is processed to obtain a voltage signal of equal magnitude and opposite phase, which is output to the human body surface through the reverse injection point to suppress or offset the amplitude of the original noise.

[0017] Furthermore, the signal source is composed of a DDS chip, capacitors C1, C4, C5, C6, C7, C8, C9, C10, C11, inductors L2, L3 and resistor R5, wherein the DDS chip provides a signal of specific frequency and amplitude for each branch, capacitors C9 and C10 are connected in parallel to form a power supply filter capacitor, one end of C9 is grounded, and capacitors C1, C4, C6, C7, C8 and inductors L2, L3 form the output filter circuit of the signal source.

[0018] In another aspect, the present invention provides a feedback noise reduction method for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device, comprising the following steps:

[0019] S1: placing electromagnetic noise detection sensors at different locations on the human body surface;

[0020] S2: After the electromagnetic noise detection sensor detects the noise signal, it obtains a voltage signal through impedance matching, amplifies the voltage signal through a preamplifier, and then outputs it to the gain adjustment circuit;

[0021] S3: After receiving the preamplifier signal, the gain adjustment circuit analyzes the voltage signal amplitude, sets the weighted average formula and the objective function, and numbers the n electromagnetic noise detection sensors counterclockwise from 1 to n starting from a certain position according to their location on the human body, and numbering the m injection points counterclockwise from 1 to m. First, the contribution weight and gain of each injection point to the signal strength of each measurement sensor are determined through actual measurement, and the signal is adjusted through the gain amplifier;

[0022] S4: The regulated signal passes through the inverter to obtain voltages of equal magnitude and opposite phase, which are then injected back into the human body through electrodes in contact with the human body.

[0023] S6: Repeat steps S1-S5 until the noise voltage signal collected by the analog-to-digital converter is less than a preset threshold, thus completing the noise interference signal cancellation.

[0024] Furthermore, the calculation of the noise weight and gain in step S3 specifically includes:

[0025] S31: The electromagnetic noise detection sensors are numbered counterclockwise from 1 to n starting from a certain position according to their positions on the human body, and the m injection points are numbered counterclockwise from 1 to m. First, the contribution weight of each injection point to the signal strength of each measurement sensor is determined through actual measurement. The signal source in the circuit applies a signal of a specific amplitude to injection point No. 1, and detection is performed at detection point No. 1, detection point No. 2, ... detection point No. n respectively. The sensitivity relationship between the injection point No. 1 of the signal source and detection point No. 1, detection point No. 2, ... detection point No. n is obtained, which is defined as the weight and recorded as w 11 、w 12 、w 13 、…w 1m Similarly, the signal source applies a signal of a specific amplitude to the injection point i, and detects it at other detection points. The weight obtained is recorded as w i1 、w i 、w i3 、…w im , and so on to get other weights w n1 、w n2 、w n3 、…w nm , w 11 …w nm Construct a symmetric weight matrix:

[0026]

[0027] S32: Construct the detection signal weighting formula:

[0028] s1=u1+w 11 *x1*v1+w 12 *x2*v2……+w 1m *x m *v m

[0029] s2=u2+w 21 *x1*v1+w 22 *x2*v2……+w 2m *x m *v m

[0030] s3=u3+w 31 *x1*v1+w 32 *x2*v2……+w 3m *x m *v m

[0031] ……(1)

[0032] s n =u n +wi1 *x1*v1+w i2 *x2*u2……+w nm *x m *v m

[0033] Where s represents the actual output voltage signal of the noise sensor at the detection point, s1 represents the actual output voltage signal of the noise sensor at the first detection point, and s n represents the actual output voltage signal of the noise sensor at the nth detection point, n represents the total number of noise sensors, u represents the human body coupled noise voltage signal at the detection point, u1 represents the human body coupled noise voltage signal at the first detection point, w represents the weight, x represents the gain of the gain circuit, v represents the reverse voltage signal output by the reverse circuit in each circuit, m represents the number of reverse injection points, and both n and m are integers not less than 1;

[0034] The matrix expression of formula (1) is:

[0035]

[0036] Rewrite the matrix expression as follows:

[0037] S=A*X+U (3)

[0038] Where A is an m*1 column vector, X is the gain matrix to be calculated, and U is an n*1 column vector, which are expressed as follows:

[0039]

[0040] U=(u1 u2…u n ) T

[0041] S33: Solve the unknown parameters in the gain matrix (x1 x2…x n ), by minimizing the sum of squares of errors, the gap between the actual measured value and the calculated value is used to evaluate the fit, so that the linear fit is achieved to the best possible value, and the function target value is established:

[0042] min||Calculation-Measurement|| 2

[0043] That is: min‖AX+U‖ 2 (4)

[0044] Where X∈R n ,A∈R m×n ,U∈R m ;

[0045] When m = n, the optimal solution of equation (4) is obtained, and the partial derivative of it with respect to parameter X is solved and set to zero, which is represented by the following system of equations:

[0046]

[0047] By combining the system of equations (5), the system of equations has a unique solution X = A -1 U, and x1, x2, x3..x are obtained m , and the gain matrix X is obtained;

[0048] When m < n, the least squares method is used to evaluate the gap between the actual measurement value and the calculated value of equation (1) to achieve the best fit;

[0049] The objective function min||calculation - measurement|| 2 is transformed into a mathematical expression as:

[0050] min‖AX + b‖ 2

[0051] where X is the vector composed of the gain parameters to be solved, (AX + b) is the residual vector. Assuming the calculation function is AX and the measurement sample is b (b = -U), and the number of equations is greater than the number of unknowns, then:

[0052] AX ≈ b

[0053] It is transformed into the least squares expression: min||AX - b|| 2 , and ||AX - b|| 2 is expanded to get:

[0054] ||AX - b|| 2 =(AX - b) T (AX - b) = X T A T AX - 2X T A T b + b T b

[0055] Solve its partial derivative with respect to parameter X and set it to zero:

[0056]

[0057] After simplification, we get: X = (A T A) -1 A T b, that is, X = (A T A) -1 A T (-U), which is expressed in the form of the pseudo-inverse of the matrix: X = A + b;

[0058] If the rectangular matrix A is full rank, in the overdetermined case, the pseudo-inverse of A is expressed as:

[0059] A + =(A T A) -1 A T

[0060] In the underdetermined case, the pseudo-inverse of A is expressed as:

[0061] A + =A T (AA T ) -1

[0062] In summary, the optimal solution of w and x is found, that is, the noise weight and the gain of the gain circuit are found.

[0063] The beneficial effects of the present invention are as follows: the present invention uses an electromagnetic noise detection sensor to collect electromagnetic noise signal data, which is amplified by a preamplifier circuit, inverted and amplified by an inverting circuit, and then fed back to the human body surface to suppress or offset the amplitude of common-mode noise. The entire circuit structure is a feedback network, which adopts closed-loop automatic control, monitors the size of the electromagnetic noise signal in real time, and dynamically adjusts it to minimize the electromagnetic noise finally detected, thereby ensuring imaging quality.

[0064] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0066] Figure 1 This is a structural block diagram of a feedback noise reduction circuit for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance device according to the present invention;

[0067] Figure 2 It is the structure diagram of the signal source circuit;

[0068] Figure 3 This is a block diagram of the human body wearable circuit of this embodiment. DETAILED DESCRIPTION

[0069] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0070] Among them, the accompanying drawings are only for illustrative purposes and represent only schematic diagrams rather than actual pictures, and should not be understood as limiting the present invention. In order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the dimensions of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions may be omitted in the accompanying drawings.

[0071] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "front", "back", etc. indicating directions or positional relationships, they are based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0072] See Figure 1 and Figure 3In order to solve the problem of active noise reduction of electromagnetic noise introduced by magnetic resonance imaging equipment under open conditions without a shielding system, a feedback noise reduction circuit and method for electromagnetic interference signals of magnetic resonance imaging equipment without electromagnetic shielding are designed. The circuit includes: an electromagnetic noise detection sensor, a preamplifier circuit, a gain adjustment circuit, and an inverter. Its working principle is to use the electromagnetic noise detection sensor to collect electromagnetic noise signal data, amplify it through the preamplifier circuit, and then amplify it through the inverting circuit before feeding it back to the human body surface to suppress or offset the amplitude of common-mode noise, thereby further improving the common-mode rejection ratio of magnetic resonance. The human body electromagnetic noise collection sensor is a conductive device that can directly contact the human body surface or contact the human body surface through a conductive agent, including patch type, ball suction type, clamp type or coil type. The number of human body electromagnetic noise collection sensors is one or more, and they are set on the human body surface, such as on the left arm, right arm, left shoulder, right shoulder or left chest, right chest and other symmetrical positions of the human body. A certain place on the human body surface can also be selected as needed. Each electromagnetic noise detection sensor is equipped with corresponding amplification, collection, and reverse branches, including an adaptive impedance matching circuit, a preamplifier circuit, a reverse circuit, and a signal source.

[0073] like Figure 1 The figure shows the circuit design diagram of the present invention, in which the impedance matching circuit is composed of resistors R4 and R8, and capacitors C2 and C3. One end of resistor R4 is used to receive the electromagnetic signal of sensor 1, and the other end thereof is connected to R8 in the preamplifier circuit. One end of resistor R8 is used to receive the electromagnetic signal of sensor 2, and the other end thereof is connected to R9 in the preamplifier circuit.

[0074] The preamplifier circuit consists of resistors R1 and R7 and an operational amplifier U2. Resistor R7 is connected to the positive pole of the operational amplifier, and resistor R7 is connected to the negative pole of the operational amplifier. The collected signal passes through the resistor to obtain a voltage, and the voltage passes through the operational amplifier circuit to amplify and collect the signal.

[0075] The gain adjustment circuit includes a gain amplifier circuit and a control circuit. The control circuit is composed of a multi-channel analog-to-digital converter (ADC), a microprocessor, and a multi-channel digital-to-analog converter (DAC). The gain amplifier circuit consists of a resistor and a gain amplifier. The voltage detected by the electromagnetic noise detection sensor is amplified by a preamplifier and output to the gain amplifier and the multi-channel analog-to-digital converter. The multi-channel analog-to-digital converter (ADC) collects the voltage signal and sends it to the microprocessor. The microprocessor calculates the gain amplifier's multiple in real time based on the collected noise voltage, and then controls the output DC voltage value of the digital-to-analog converter (DAC), achieving dynamic adjustment of the gain amplifier's gain.

[0076] The inverter consists of resistors R4 and R8 and operational amplifier U1. The input end of resistor R4 is connected to the preamplifier. Resistor R8 and operational amplifier U1 pass the amplified signal through the inverter to obtain a voltage signal of equal magnitude and opposite phase, which is output to the right leg to suppress or offset the amplitude of the original noise.

[0077] like Figure 2 As shown, the signal source consists of a DDS chip, capacitors C1, C4, C5, C6, C7, C8, C9, C10, and C11, inductors L2 and L3, and resistor R5. The DDS chip provides a signal of specific frequency and amplitude for each branch. Capacitors C9 and C10 are connected in parallel to form a power supply filter capacitor, and one end of C9 is grounded. Capacitors C1, C4, C6, C7, and C8 and inductors L2 and L3 form the output filter circuit of the signal source. The injection electrode and the injection point electrode of the signal source are shared, and both are controlled by a microprocessor.

[0078] The feedback noise reduction method for electromagnetic interference signals comprises the following steps:

[0079] S1: Setting electromagnetic noise detection sensors at different locations on the human body surface;

[0080] S2: After the electromagnetic noise detection sensor detects the noise signal, it obtains a voltage signal through impedance matching. The voltage signal is amplified by the preamplifier and then output to the gain adjustment circuit;

[0081] S3: After receiving the preamplifier signal, the gain adjustment circuit analyzes and calculates the voltage signal amplitude, frequency, and phase, sets the weighted average formula and objective function, and dynamically adjusts the gain;

[0082] S4: The regulated signal passes through the inverter to obtain voltages of equal magnitude and opposite phase, which are then injected back into the human body through electrodes in contact with the human body.

[0083] S6: Repeat steps S1-S5 until the noise voltage signal collected by the analog-to-digital converter is less than a preset threshold, thus completing the noise interference signal cancellation.

[0084] The specific method of dynamically adjusting the gain in S3 is:

[0085] S31: The electromagnetic noise detection sensors are numbered counterclockwise from 1 to n starting from a certain position according to their positions on the human body, and the m injection points are numbered counterclockwise from 1 to m. First, the contribution weight of each injection point to the signal strength of each measurement sensor is determined through actual measurement. The signal source in the circuit applies a signal of a specific amplitude to injection point No. 1, and detection is performed at detection point No. 1, detection point No. 2, ... detection point No. n respectively. The sensitivity relationship between the injection point No. 1 of the signal source and detection point No. 1, detection point No. 2, ... detection point No. n is obtained (this sensitivity relationship is named "weight" here, the same below), which is recorded as w 11 、w 12 、w 13 、…w 1m Similarly, the signal source applies a signal of a specific amplitude to the injection point i, and detects it at other detection points respectively. The obtained weight is recorded as w i1 、w i 、w i3 、…w im , and so on to get other weights w n1 、w n2 、w n3 、…w nm , w 11 …w nm Construct a symmetric weight matrix:

[0086]

[0087] S32: Construct a detection signal weighting formula:

[0088] s1=u1+w 11 *x1*v1+w 12 *x2*v2……+w 1m *x m *v m

[0089] s2=u2+w 21 *x1*v1+w 22 *x2*v2……+w 2m *x m *v m

[0090] s3=u3+w 31 *x1*v1+w 32 *x2*v2……+w 3m *x m *v m

[0091] ……(1)

[0092] s n =un +w i1 *x1*v1+w i2 *x2*u2……+w nm *x m *v m

[0093] Where s represents the actual output voltage signal of the noise sensor at the detection point, s1 represents the actual output voltage signal of the noise sensor at the first detection point, and s n represents the actual output voltage signal of the noise sensor at the nth detection point, n represents the total number of noise sensors, u represents the human body coupled noise voltage signal at the detection point, u1 represents the human body coupled noise voltage signal at the first detection point, w represents the weight, x represents the gain of the gain circuit, v represents the reverse voltage signal output by the reverse circuit in each circuit, m represents the number of reverse injection points, and both n and m are integers not less than 1;

[0094] The matrix expression of formula (1) is:

[0095]

[0096] S33: Rewrite the above formula into the following formula:

[0097] S=A*X+U (3)

[0098] Where A is an m*1 column vector, X is the gain matrix to be calculated, and U is an n*1 column vector, which are expressed as follows:

[0099]

[0100] U=(u1 u2…u n ) T

[0101] S34: Solve the unknown parameters in the gain matrix (x1 x2…x n ), we can evaluate the goodness of fit by minimizing the sum of squares of errors and using the gap between actual measured values ​​and calculated values ​​to achieve the best linear fit and establish the function target value:

[0102] min||Calculation-Measurement|| 2

[0103] That is: min‖AX+U‖ 2 (4)

[0104] Where X∈R n ,A∈R m×n ,U∈R m ;

[0105] When m = n, the optimal solution (i.e., the minimum value) of Equation (4) is obtained by solving. Solve its partial derivative with respect to the parameter X and set it equal to zero, which is expressed by the following system of equations:

[0106]

[0107] By solving the above system of equations (5) simultaneously, the system of equations has a unique solution X = A -1 U, then x1, x2, x3..x can be obtained m , and thus the gain matrix X is obtained;

[0108] When m < n, in Equation (1), the number of equations is greater than the number of unknowns, and this equation is called an overdetermined equation. Conversely, when m > n, that is, the number of equations is less than the number of unknowns, this equation is called an underdetermined equation. In both cases, the least squares method can be used to evaluate the gap between the actual measured values and the calculated values to achieve the best fit;

[0109] The objective function min||calculated - measured|| 2 , is transformed into a mathematical expression as:

[0110] min‖AX + b‖ 2

[0111] where X is the vector composed of the gain parameters to be determined, and (AX + b) is the residual vector. Now assume that the calculation function is AX and the measurement sample is b (b = -U), and the number of equations is greater than the number of unknowns, then there is:

[0112] AX ≈ b

[0113] It is transformed into a least squares expression as: min||AX - b|| 2 , for ||AX - b|| 2 Expanding it gives:

[0114] ||AX - b|| 2 =(AX - b) T (AX - b)= X T A T AX - ²X T A T b + b T b

[0115] To obtain the optimal solution (i.e., the minimum value) of the equation, solve its partial derivative with respect to the parameter X and set it equal to zero:

[0116]

[0117] After simplification, we get: X = (A T A) -1 A T b, that is, X = (AT A) -1 A T (-U)

[0118] The result can also be expressed as the pseudo-inverse form of the matrix (pseudo-inverse is a generalized form of a quasi-matrix. A singular matrix or a non-square matrix does not have an inverse matrix, but a pseudo-inverse matrix can be calculated): X = A + b,

[0119] If the rectangular matrix A is full rank, in the overdetermined case, the pseudo-inverse of A can be expressed as:

[0120] A + =(A T A) -1 A T

[0121] In the underdetermined case, the pseudo-inverse of A can be expressed as:

[0122] A + =A T (AA T ) -1

[0123] In summary, find the optimal solution of w and x, that is, find the noise weight and the gain of the gain circuit;

[0124] The entire circuit structure of the feedback noise reduction circuit for electromagnetic interference signals is a feedback network, which adopts closed-loop automatic control, monitors the size of electromagnetic noise signals in real time, and dynamically adjusts to minimize the electromagnetic noise finally detected.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A feedback circuit for electromagnetic interference signals of a non-electromagnetic shielded magnetic resonance imaging device, characterized in that: include: Electromagnetic noise detection sensor, used to collect electromagnetic noise signals on the human body surface in real time; An impedance matching circuit, for matching human skin impedance, with an input end electrically connected to the electromagnetic noise detection sensor and an output end electrically connected to the preamplifier circuit; A preamplifier circuit amplifies the filtered electromagnetic noise signal data, wherein the input end of the preamplifier circuit is electrically connected to the impedance matching circuit, and the output end of the preamplifier circuit is electrically connected to the gain adjustment circuit; A gain adjustment circuit includes a gain amplifier circuit and a control circuit. The gain amplifier circuit is used to perform gain amplification, and its input end is electrically connected to the preamplifier circuit, and its output end is electrically connected to the inverter. The control circuit is used to calculate the gain multiple of the gain amplifier circuit in real time based on the magnitude of the collected noise voltage, and control the dynamic gain adjustment of the gain amplifier circuit. The input end is electrically connected to the preamplifier circuit, and the output end is connected to the gain amplifier circuit. The inverter, whose input end is electrically connected to the gain adjustment circuit, converts the amplified signal into a voltage signal of equal magnitude and opposite phase, and outputs it to the human body surface for noise reduction; The signal source has an input end electrically connected to the gain adjustment circuit and an output end connected to the human body surface, and is used to inject a signal of specific frequency and amplitude into each branch at the injection point and perform detection at other detection points.

2. The feedback circuit for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device according to claim 1, characterized in that: The impedance matching circuit is composed of resistors R4 and R8 and capacitors C2 and C3. One end of the resistors R4 and R8 is used to receive the electromagnetic signal of the electromagnetic noise detection sensor, and the other end thereof is electrically connected to the operational amplifier in the preamplifier circuit.

3. The feedback circuit for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device according to claim 1, characterized in that: The preamplifier circuit is composed of resistors R2 and R9 and an operational amplifier U3. One end of the resistor R2 is connected to the positive electrode of the operational amplifier U3, and the other end is connected to the output end of the operational amplifier U3; one end of the resistor R9 is connected to the negative electrode of the operational amplifier U3, and the other end is grounded; the signal collected by the electromagnetic noise detection sensor is converted into a voltage through the resistor, and the voltage is amplified by the operational amplifier circuit.

4. The feedback circuit for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device according to claim 1, characterized in that: The control circuit is composed of a multi-channel analog-to-digital converter (ADC), a microprocessor, and a multi-channel digital-to-analog converter (DAC); the gain amplifier circuit is composed of a resistor and a gain amplifier; the voltage detected by the electromagnetic noise detection sensor is amplified by a preamplifier and then output to the gain amplifier and the multi-channel analog-to-digital converter; the multi-channel analog-to-digital converter (ADC) collects the voltage signal and sends it to the microprocessor; the microprocessor calculates the gain amplifier multiple in real time based on the size of the collected noise voltage, and then controls the output DC voltage value of the digital-to-analog converter (DAC), thereby realizing dynamic adjustment of the gain of the gain amplifier.

5. The feedback circuit for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device according to claim 1, characterized in that: The inverter is composed of resistors R6 and R10 and an operational amplifier U2. One end of the resistor R6 is electrically connected to the gain amplifier, and the other end is connected to the reverse input terminal of the operational amplifier U2; one end of the resistor R10 is connected to the resistor R6, and one end is connected to the output terminal of the operational amplifier U2. The amplified signal is processed to obtain a voltage signal of equal magnitude and opposite phase, which is output to the human body surface through the reverse injection point to suppress or offset the amplitude of the original noise.

6. The feedback circuit for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device according to claim 1, characterized in that: The signal source consists of a DDS chip, capacitors C1, C4, C5, C6, C7, C8, C9, C10, C11, inductors L2, L3 and a resistor R5. The DDS chip provides signals with specific frequencies and amplitudes for each branch. Capacitors C9 and C10 are connected in parallel to form a power supply filtering capacitor, with one end of C9 grounded. Capacitors C1, C4, C6, C7, C8 and inductors L2, L3 form the output filtering circuit of the signal source.

7. A feedback noise reduction method for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device, characterized by: It includes the following steps: S1: Set each electromagnetic noise detection sensor at different positions on the human body surface; S2: After the electromagnetic noise detection sensor detects a noise signal, it obtains a voltage signal through impedance matching. The voltage signal is amplified by a preamplifier and then output to a gain adjustment circuit; S3: After receiving the signal from the preamplifier, the gain adjustment circuit analyzes the amplitude of the voltage signal, sets a weighted average formula and an objective function. For the electromagnetic noise detection sensors according to their positions on the human body, starting from a certain position, the n electromagnetic noise detection sensors are numbered counterclockwise from 1 to n, and the m injection points are numbered counterclockwise from 1 to m. First, determine the noise weights and gain magnitudes of the signal strengths of each injection point on each measurement sensor through actual measurement, and adjust the signal through a gain amplifier; S4: The adjusted signal passes through an inverter to obtain a voltage with the same magnitude and opposite phase, and is injected back into the human body through an electrode in contact with the human body; S6: Repeat steps S1 - S5 until the noise voltage signal collected by the analog-to-digital converter is less than a pre-set threshold, and the cancellation of the noise interference signal is completed.

8. The feedback noise reduction method for electromagnetic interference signals of a non-electromagnetic shielding magnetic resonance imaging device according to claim 7, characterized in that: The calculation of the noise weights and gain magnitudes in step S3 specifically includes: S31: The electromagnetic noise detection sensors are numbered counterclockwise from 1 to n starting from a certain position according to their positions on the human body, and the m injection points are numbered counterclockwise from 1 to m. First, the noise weight of each injection point on the signal strength of each measurement sensor is determined through actual measurement. The signal source in the circuit applies a signal of a specific amplitude to injection point No. 1, and detection is performed at detection point No. 1, detection point No. 2, ... detection point No. n respectively. The sensitivity relationship between the injection point No. 1 of the signal source and detection point No. 1, detection point No. 2, ... detection point No. n is obtained, which is defined as the weight and recorded as w 11 、w 12 、w 13 、…w 1m Similarly, the signal source applies a signal of a specific amplitude to the injection point i, and detects it at other detection points. The weight obtained is recorded as w i1 、w i 、w i3 、…w im , and so on to get other weights w n1 、w n2 、w n3 、…w nm , w 11 …w nm Construct a symmetric weight matrix: S32: Construct a weighted formula for the detection signal: Where s represents the actual output voltage signal of the noise sensor at the detection point, s1 represents the actual output voltage signal of the noise sensor at the first detection point, and s n represents the actual output voltage signal of the noise sensor at the nth detection point, n represents the total number of noise sensors, u represents the human body coupled noise voltage signal at the detection point, u1 represents the human body coupled noise voltage signal at the first detection point, w represents the weight, x represents the gain of the gain circuit, v represents the reverse voltage signal output by the reverse circuit in each circuit, m represents the number of reverse injection points, and both n and m are integers not less than 1; The matrix expression of formula (1) is: Rewrite the matrix expression as the following formula: S = A*X + U (3) Where A is an m*1 column vector, X is the gain matrix to be solved, and U is an n*1 column vector, which are respectively represented as follows: U=(u1 u2 … u n ) T S33: Solve for the unknown parameters in the gain matrix (x1 x2 … x n ), by minimizing the sum of squares of errors, the gap between the actual measured value and the calculated value is used to evaluate the fit, so that the linear fit is achieved to the best possible value, and the function target value is established: min||Calculation-Measurement|| 2 Immediately: min||AX+U|| 2 (4) Where X∈R n ,A∈R m×n ,U∈R m ; When m = n, solve the optimal solution of equation (4), solve its partial derivative with respect to parameter X, and set it equal to zero, which is represented by the following set of equations: The simultaneous equations (5) have a unique solution X = A -1 U, find x1, x2, x3..x m , then we get the gain matrix X; When m < n, use the least squares method to evaluate the gap between the actual measurement values and the calculated values of equation (1) to achieve the best fit; The objective function min||calculation-measurement|| 2 , converted into mathematical expression: min||AX+b|| 2 Where X is the vector composed of the gain parameters to be solved, (AX + b) is the residual vector. Assuming the calculation function is AX and the measurement sample is b (b = -U), and the number of equations is greater than the number of unknowns, then there is: AX ≈ b Converted into the least squares expression: min||AX-b|| 2 , change ||AX-b|| 2 Expand to get: ||AX-b|| 2 =(AX-b) T (AX-b)=X T A T AX-2X T A T b+b T b solves its partial derivative with respect to the parameter X and sets it equal to zero: After simplification, we get: X=(A T A) -1 A T b, that is, X=(A T A) -1 A T (-U), expressed as the pseudo-inverse form of the matrix: X = A + b; If the long matrix A is full rank, in the overdetermined case, the pseudo-inverse of A is represented as: A + =(A T A) -1 A T In the underdetermined case, the pseudo-inverse of A is represented as: A + =A T (CHALLENGE ACCEPTED T ) -1 In summary, find the optimal solutions of w and x, that is, find the noise weights and the gain of the gain circuit.

Images