A method of optimizing an induction coil of a magnetic resonance imaging system

Abstract

The present application relates to the technical field of magnetic resonance imaging, in particular to a method for optimizing induction coils of a magnetic resonance imaging system, comprising: step S1: closing the radio frequency transmitter of the magnetic resonance imaging system, and then collecting environmental electromagnetic interference signals as calibration data by using the receiving coil channel and the induction coil channel of the magnetic resonance imaging system; step S2: determining the noise estimation error of the receiving coil channel in the estimation process; step S3: determining the induction coil channel which has less effect in the noise estimation process, and then optimizing the coil parameters of the induction coil channel until the measurement requirement is met. The beneficial effect is that the calibration data is introduced in advance, and the estimation error of each induction coil channel to the receiving coil in the noise estimation process is determined, so that the induction coil channel with less effect is screened out. The number of induction coils available in the noise estimation is improved by further optimizing the screened induction coils, so that better noise estimation effect is achieved.

Description

Technical Field

[0001] This invention relates to the field of magnetic resonance imaging technology, and more specifically to a method for optimizing the induction coil of a magnetic resonance imaging system. Background Technology

[0002] Magnetic Resonance Imaging (MRI) is a non-invasive imaging method that can display the structure and properties of human tissues. In recent years, the development of electromagnetically shielded MRI technology has made portable MRI possible. However, without electromagnetic shielding, MRI is inevitably affected by environmental electromagnetic interference. Technically, induction coils can be used to measure and eliminate this interference. Specifically, multiple induction coils are needed to accurately measure environmental electromagnetic interference, which is then used to estimate and eliminate electromagnetic interference affecting the main channel.

[0003] In the prior art, there are already technical solutions for noise cancellation based on induction coils. For example, Chinese patent CN202110477311.7 discloses an interference cancellation method, medium, and device that can eliminate interference signals from measurement signals received from multiple channels to obtain an effective signal, thereby avoiding the influence of interference signals on the effective signal. The method includes: acquiring a measurement signal containing a mixture of an effective signal and a first interference signal from a first type of channel, and acquiring a second interference signal from a second type of channel; estimating the first interference signal in the measurement signal based on the coupling relationship between first calibration data and second calibration data, and based on the second interference signal; removing the first interference signal from the measurement signal to obtain the target effective signal; wherein the first calibration data and the second calibration data are interference signals acquired from the first type of channel and the second type of channel respectively when the electronic device is in a preset state. This method can be specifically used in scenarios where electromagnetic interference signals affect magnetic resonance imaging signals.

[0004] However, in actual implementation, the inventors found that this type of technical solution requires the introduction of a large number of induction coils to estimate noise, but there is no effective method to evaluate the optimization effect of the induction coils. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, a method for optimizing the induction coil of a magnetic resonance imaging system is provided.

[0006] The specific technical solution is as follows:

[0007] A method for optimizing the induction coil of a magnetic resonance imaging system includes:

[0008] Step S1: Turn off the radio frequency transmission of the magnetic resonance imaging system, and then use the receiving coil channel and induction coil channel of the magnetic resonance imaging system to collect the environmental electromagnetic interference signal as calibration data;

[0009] This operation can be achieved by setting the RF pulse flip angle to 0.

[0010] Step S2: Determine the noise estimation error of the receiving coil channel during the estimation process based on the calibration data;

[0011] The estimation error is the error in estimating the electromagnetic interference of the receiving coil channel based on the multi-channel sampling data of the induction coil channel;

[0012] Step S3: Determine the induction coil channel that has a smaller impact on the noise estimation process based on the noise estimation error, and then optimize the coil parameters of the induction coil channel until the measurement requirements are met.

[0013] On the other hand, step S2 includes:

[0014] Step S21: Obtain sampling data for each of the induction coil channels and the receiving coil channels from the calibration data, and construct an intermediate fitting equation for noise estimation of the receiving coil channels based on the sampling data;

[0015] In step S21, the fitting coefficients in the intermediate fitting equation are unknowns;

[0016] Step S22: Establish a system of fitting equations based on the intermediate fitting equations, obtain the fitting coefficients for each receiving coil channel based on the system of fitting equations, and generate the actual fitting equations based on the fitting coefficients and the intermediate fitting equations.

[0017] Step S23: Use the fitting equation to generate the estimation error for each of the receiving coil channels.

[0018] On the other hand, in step S21, the original data of the receiving coil of each receiving coil channel and the original data of the induction coil of each induction coil channel in the calibration data are sampled in advance using a convolution kernel to obtain the sampled data.

[0019] On the other hand, the fitting equation includes:

[0020]

[0021] In the formula, s r,t To estimate the noise, c i,jThe fitting coefficients for the j-th sampled data on the i-th induction coil channel, i = (1, 2, ..., M), s i,j For the i-th induction coil channel, the sampled value of the j-th sampling data;

[0022] During the execution of step S21, for the t-th sample, a convolution kernel of size 2K+1 is defined, which contains the tK-th to t+K-th sample data of all the induction coil channels.

[0023] On the other hand, the estimation error is:

[0024]

[0025] In the formula, E is the estimation error, and s r,j For the r-th receiving coil channel, the sampled value of the j-th sampled data.

[0026] On the other hand, step S3 includes:

[0027] Step S31: Filter the induction coil channels according to the estimated error to obtain the coil to be optimized;

[0028] Step S32: Adjust the coil parameters of the coil to be optimized, then re-acquire the calibration data and calculate the estimated error;

[0029] Step S33: Repeat step S32 until the iteration condition is met and then output.

[0030] On the other hand, the coil parameters include the position of the induction coil, the direction of the induction coil, and the coil tuning parameters.

[0031] On the other hand, the iteration conditions include pre-determining the minimum number of induction coils required based on noise estimation needs;

[0032] After performing step S33, the following is also included:

[0033] Determine whether the currently adjusted induction coil channel can estimate electromagnetic interference noise, i.e., whether the estimation error is lower than a preset threshold. If not, increase the minimum number of induction coils. If, after removing the data from the least effective induction coil, the estimation error is still less than the preset threshold, then that induction coil can be removed, and the minimum number of induction coils can be reduced.

[0034] The above technical solution has the following advantages or beneficial effects:

[0035] To address the issue that existing noise cancellation methods may suffer from poor noise estimation results due to suboptimal induction coil placement, this solution pre-introduces calibration data and identifies the estimation error of each induction coil channel on the receiving channel coil during noise estimation. This process filters out induction coil channels with minimal impact. Further optimization of these filtered induction coils increases the number of usable induction coils for noise estimation, thereby achieving better noise estimation results. Attached Figure Description

[0036] Embodiments of the invention will be described more fully with reference to the accompanying drawings. However, the drawings are for illustration and explanation only and do not constitute a limitation on the scope of the invention.

[0037] Figure 1 This is an overall schematic diagram of an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of sub-step S2 in an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of the convolution kernel in an embodiment of the present invention;

[0040] Figure 4 This is a schematic diagram of step S3 in an embodiment of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0043] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0044] This invention includes:

[0045] A method for optimizing the induction coils of a magnetic resonance imaging system is provided. The magnetic resonance system includes at least one set of receiving coils for actual imaging and multiple sets of induction coils for noise estimation and noise cancellation of the receiving coils. The receiving coils and induction coils have different placement, orientation, coil parameters, etc., depending on the magnetic resonance system. The actual algorithm for noise cancellation may also change depending on the embodiment.

[0046] like Figure 1 As shown, the induction coil optimization method includes:

[0047] Step S1: Turn off the radio frequency transmission of the magnetic resonance imaging system, and then use the receiving coil channel and induction coil channel of the magnetic resonance imaging system to collect the environmental electromagnetic interference signal as calibration data;

[0048] This operation can be achieved by setting the RF pulse flip angle to 0.

[0049] Step S2: Determine the noise estimation error of the receiving coil channel during the estimation process based on the calibration data;

[0050] The estimation error is the error in estimating the electromagnetic interference of the receiving coil channel based on the multi-channel sampling data of the induction coil channel;

[0051] Step S3: Determine the induction coil channel that has a smaller impact on the noise estimation process based on the noise estimation error, and then optimize the coil parameters of the induction coil channel until the measurement requirements are met.

[0052] Specifically, to achieve better coil optimization, in this embodiment, the radio frequency transmission channel of the magnetic resonance imaging system is shut down beforehand. Then, the receiving coil channel and induction coil channel of the magnetic resonance imaging system are used to collect environmental electromagnetic interference signals to form corresponding data sequences as calibration data, including multiple sets of original data sequences from the induction coil channels and the receiving coil channel. Subsequently, noise estimation is performed on the receiving coil channel based on the calibration data to determine the estimation error. This estimation error includes the noise estimation result error of the receiving coil channel caused by the sampling data of each induction coil channel during the noise estimation process. Analyzing the induction coil channels based on this estimation error allows us to determine the magnitude of the influence of each induction coil channel on the receiving coil channel during the noise estimation process. If the influence is large, it indicates that the environmental electromagnetic interference signal component collected on that induction coil channel plays a significant role in the noise estimation process; if the influence is small, it indicates that the environmental electromagnetic interference signal component collected on that induction coil channel plays a small role in the noise estimation process, possibly due to improper coil settings leading to the formation of invalid coils. In this case, it is necessary to optimize and adjust the coil parameters of the induction coil to reduce the introduction of invalid induction coil channels during the noise estimation process, thereby improving processing efficiency and the accuracy of noise estimation.

[0053] In one embodiment, such as Figure 2 As shown, step S2 includes:

[0054] Step S21: Obtain sampling data for each induction coil channel and receiving coil channel from the self-calibration data, and construct an intermediate fitting equation for noise estimation of the receiving coil channel based on the sampling data;

[0055] In step S21, the fitting coefficients in the intermediate fitting equation are unknowns;

[0056] Step S22: Establish a system of fitting equations based on multiple intermediate fitting equations, obtain the fitting coefficients for each receiving coil channel based on the system of fitting equations, and generate the actual fitting equations based on the fitting coefficients and intermediate fitting equations.

[0057] Step S23: Use the fitted equation to generate the estimation error for each receiving coil channel.

[0058] Specifically, to achieve a better assessment of the estimation error, in this embodiment, sampling data is first obtained by sampling each receiving coil channel and induction coil channel in the calibration data. For example... Figure 3 The sampling method in step S21 includes: pre-sampling the original receiving coil data of each receiving coil channel and the original induction coil data of each induction coil channel in the calibration data using a convolution kernel to obtain sampled data. The original receiving coil data and the original induction coil data of each induction coil channel are time-series data. After aligning the original receiving coil data and the original induction coil data, a convolution kernel of size 2K+1 is used to extract the above sequences in chronological order to form a set of sampled data. The sampled data includes sampled values ​​from multiple channels distributed across adjacent time windows. Assuming there are a total of M induction coil channels and R receiving coil channels, the t-th set of sampled data includes the tK-th to t+K-th samples from all induction coil channels, totaling M×(2K+1) elements.

[0059] Based on the above sampled data, the fitting equations for noise estimation include:

[0060]

[0061] In the formula, s r,t To estimate the noise, c i,j Let be the fitting coefficients for the j-th sampled data on the i-th induction coil channel, i = (1, 2, ..., M), and s i,j For the i-th induction coil channel, the sampled value of the j-th sampled data;

[0062] In step S21, the fitting coefficient is a time-invariant unknown.

[0063] To obtain the fitting coefficients, the aforementioned intermediate fitting equations containing unknowns are constructed for each sampled data point from the r-th receiving coil channel. Then, a system of linear fitting equations is formed by simultaneously solving these intermediate fitting equations to obtain the fitting coefficients. Subsequently, based on the above fitting equations, the estimate s' of the electromagnetic interference experienced by the r-th receiving coil channel, fitted from the induction coil channel data, can be easily obtained. r,t

[0064] Based on the above process, the estimation error can be defined as:

[0065]

[0066] In the formula, E is the estimation error, and s r,j For the r-th receiving coil channel, the sampled value of the j-th sampled data.

[0067] Where * denotes the conjugate operation, and N is the total number of samples. In the noise estimation stage, the smaller the estimation error, the more accurate the estimation result.

[0068] In one embodiment, such as Figure 4 As shown, step S3 includes:

[0069] Step S31: Filter the induction coil channels according to the estimation error to obtain the coil to be optimized;

[0070] Step S32: Adjust the coil parameters of the coil to be optimized, then reacquire calibration data and calculate the estimated error;

[0071] Step S33: Repeat step S32 until the iteration condition is met and then output.

[0072] Specifically, to achieve better optimization results, in this embodiment, the induction coil channels are first screened according to the estimated error to obtain the coil to be optimized. Then, the coil parameters, such as the induction coil position, induction coil direction, and coil tuning parameters, are adjusted to improve the induction coil's ability to collect environmental electromagnetic interference signals. In step S31, one of the M induction coil channels (let's say the i-th one) is pre-selected and the remaining error obtainable by the remaining M-1 channels is calculated. The smaller the remaining error, the less effective the induction coil channel is, and the more it needs to be individually screened and optimized. It should be noted that after adjusting the coil parameters of the coil to be optimized, the calibration data needs to be re-acquired and the fitting coefficients recalculated according to the above process.

[0073] In one embodiment, the iteration condition includes pre-determining the minimum number of induction coils required based on noise estimation needs;

[0074] After step S33, the following is also included:

[0075] Determine whether the currently adjusted induction coil channels can estimate electromagnetic interference noise, and if not, increase the minimum number of induction coils.

[0076] Specifically, in order to achieve a good iterative effect on the induction coils, in this embodiment, after step S33 is performed, R' (R'<R) channels are further selected and the estimation error is calculated, it is determined whether the estimation error is less than a preset error threshold; if it is less, it indicates that the adjusted induction coil channels can estimate the electromagnetic interference noise, and if it is greater, it is necessary to increase the number of induction coil channels, and perform the aforementioned processes of residual error calculation, optimization and determination until the requirement is met.

[0077] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation manners and protection scope of the present invention. For those skilled in the art, it should be appreciated that all schemes obtained by equivalent replacements and obvious changes made using the content of the specification and the drawings of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for optimizing the induction coil of a magnetic resonance imaging system, characterized in that, include: Step S1: Turn off the radio frequency transmission of the magnetic resonance imaging system, and then use the receiving coil channel and induction coil channel of the magnetic resonance imaging system to collect the environmental electromagnetic interference signal as calibration data; Step S2: Determine the noise estimation error of the receiving coil channel during the estimation process based on the calibration data; The estimation error is the error in estimating the electromagnetic interference of the receiving coil channel based on the multi-channel sampling data of the induction coil channel; Step S3: Determine the induction coil channel that has a small effect in the noise estimation process based on the noise estimation error, and then optimize the coil parameters of the induction coil channel until the measurement requirements are met; Step S3 includes: Step S31: Filter the induction coil channels according to the estimated error to obtain the coil to be optimized; Step S32: Adjust the coil parameters of the coil to be optimized, then re-acquire the calibration data and calculate the estimated error; Step S33: Repeat step S32 until the iteration condition is met and then output the result; In step S31, one of the induction coil channels is pre-removed and the remaining error of the remaining channels is calculated. Based on the remaining error, the function of the induction coil channel is determined, and the induction coil channel with the smallest function is selected as the coil to be optimized.

2. The induction coil optimization method according to claim 1, characterized in that, Step S2 includes: Step S21: Obtain sampling data for each of the induction coil channels and the receiving coil channels from the calibration data, and construct an intermediate fitting equation for noise estimation of the receiving coil channels based on the sampling data; In step S21, the fitting coefficients in the intermediate fitting equation are unknowns; Step S22: Establish a system of fitting equations based on the intermediate fitting equations, obtain the fitting coefficients for each receiving coil channel based on the system of fitting equations, and generate the actual fitting equations based on the fitting coefficients and the intermediate fitting equations. Step S23: Use the fitting equation to generate the estimation error for each of the receiving coil channels.

3. The induction coil optimization method according to claim 2, characterized in that, In step S21, the original data of the receiving coil of each receiving coil channel and the original data of the induction coil of each induction coil channel in the calibration data are sampled in advance using a convolution kernel to obtain the sampled data.

4. The induction coil optimization method according to claim 3, characterized in that, The fitting equation includes: ； In the formula, To estimate the noise, For the first The first induction coil channel The fitting coefficients of the sampled data. , For the first The induction coil channel, the first The sampled values ​​of the sampled data; During the execution of step S21, for the first Each sample, defined as size. The convolution kernel, containing the first of all the induction coil channels. To the One sampled data.

5. The induction coil optimization method according to claim 3, characterized in that, The estimation error is: ； In the formula, The estimation error is... For the first The receiving coil channel, the first The sampled values ​​of each sampled data point.

6. The induction coil optimization method according to claim 1, characterized in that, The coil parameters include the position of the induction coil, the direction of the induction coil, and the coil tuning parameters.

7. The induction coil optimization method according to claim 1, characterized in that, The iteration conditions include pre-determining the minimum number of induction coils required based on noise estimation needs; After performing step S33, the following is also included: Determine whether the currently adjusted induction coil channel can estimate electromagnetic interference noise. If not, increase the minimum number of induction coils.

Images