ECG gating signal processing device and method for nuclear magnetic resonance

Abstract

The present invention discloses a device and method for processing electrocardiogram gating signals under nuclear magnetic resonance, belonging to the technical field of nuclear magnetic resonance electrocardiogram gating processing, including: collecting the superimposed signal of the electrocardiogram signal and the nuclear magnetic noise through an integrated circuit board with physical magnetic shielding, and sequentially performing noise filtering and amplification to obtain the original electrocardiogram signal; performing real-time frequency analysis on the original electrocardiogram signal, extracting various electrocardiogram signal features, reconstructing a mask vector to filter noise and reconstruct the time-domain R-wave signal, and reconstructing the electrocardiogram signal based on the reconstructed time-domain R-wave signal and various electrocardiogram signal features, detecting the R-wave trigger point based on the reconstructed electrocardiogram signal and generating a gating signal, and also performing missed trigger and false trigger detection on the R-wave trigger point in real time; outputting the reconstructed electrocardiogram signal, the gating signal, and various electrocardiogram signal features to the nuclear magnetic resonance system, supporting the matching of different gating trigger conditions based on the magnetic resonance scanning sequence, so as to achieve accurate positioning of the R wave.

Description

Technical Field

[0001] The present invention belongs to the technical field of nuclear magnetic resonance electrocardiogram gating processing, and particularly relates to an electrocardiogram gating signal processing device and method for nuclear magnetic resonance. Background Art

[0002] Nuclear magnetic resonance electrocardiogram gating technology (ECG Gated MRI) is a technology that synchronizes an electrocardiogram (ECG) signal with magnetic resonance imaging (MRI), mainly used for the acquisition and analysis of cardiac images. Due to the continuous beating of the heart, motion artifacts will be caused in conventional MRI imaging, affecting the quality of the image and the accuracy of diagnosis. The electrocardiogram gating technology can significantly improve the image quality and diagnostic accuracy by performing MRI imaging at specific moments in the cardiac cycle. The nuclear magnetic resonance electrocardiogram gating technology uses the electrocardiogram signal to coordinate and synchronize the magnetic resonance scanner to ensure that image acquisition is performed at the stage of minimal cardiac motion. Specifically, the R wave in the ECG signal is used as the trigger signal because the R wave represents the starting stage of cardiac contraction and is relatively easy to detect. Based on the R wave signal, image acquisition can be performed at specific moments in the cardiac cycle (such as the ventricular diastolic phase), thereby reducing or eliminating cardiac motion artifacts.

[0003] Existing nuclear magnetic resonance electrocardiogram gating includes prospective gating and retrospective gating. Prospective gating: Image acquisition is performed after each R wave, and usually a fixed delay time is set to ensure that the acquisition is performed at the stage of relatively static heart. This method is applicable to patients with stable heart rate and has a short imaging time, but it cannot be used for the case of arrhythmia. Retrospective gating: The electrocardiogram signal and MRI data are acquired simultaneously, and then the images are sorted and reconstructed according to the electrocardiogram signal in image post-processing. This method can capture images of the entire cardiac cycle, and can obtain cardiac images with high temporal resolution and spatial resolution, which is helpful for comprehensively analyzing cardiac function and is widely used for cardiac function evaluation, myocardial histology examination and pathological diagnosis, but the acquisition time is long. These two electrocardiogram gating technologies can effectively reduce image artifacts caused by heart beating and significantly improve the image quality and clarity. Clear cardiac images are helpful for evaluating the cardiac anatomical structure and functional status, especially applicable to the diagnosis of complex cardiac diseases such as cardiomyopathy and valvular heart disease.

[0004] Although the above nuclear magnetic resonance electrocardiogram gating technology has significant advantages in improving cardiac imaging quality, there are also some technical difficulties and challenges. The main difficulties faced in practical applications specifically include:

[0005] 1. The strong magnetic field and radiofrequency pulses in the MRI environment can cause significant interference to the electrocardiogram (ECG) signal. Especially in high-field MRI (such as 3.0T or higher), the interference is more obvious. This interference may cause the distortion of the ECG signal, affect the accurate detection of the R wave, and further affect the accuracy of gating triggering. Since the ECG electrodes and wires may generate artifacts or interference signals in the magnetic field, special designs and shielding measures are required to reduce the impact of these interferences.

[0006] 2. For patients with arrhythmia or atrial fibrillation, the ECG signal is unstable, and the difficulty of R wave detection increases. In this case, the ECG gating may not be accurately triggered, resulting in imaging failure or reduced quality. In some special cases (such as when the T wave is higher or abnormally prominent), the ECG gating system may misidentify the T wave as the R wave, thus triggering an incorrect trigger and affecting the accuracy of image acquisition.

[0007] 3. The ECG gating technology needs to perform real-time signal processing while collecting the ECG signal to accurately identify the R wave and trigger. The signal processing algorithm must complete the analysis within a short time; otherwise, it will cause a trigger delay, thereby affecting the image quality. For retrospective gating, the amount of MRI data collected is large, and the computational load of subsequent data reconstruction and processing increases significantly, posing higher requirements on the system hardware performance. Summary of the Invention

[0008] In view of the above, the object of the present invention is to provide an ECG gating signal processing device and method for nuclear magnetic resonance to solve the problems in the related art that the ECG signal waveform is affected by magnetic resonance and generates a large amount of noise, making it impossible to accurately locate the R wave, and the difficulty of locating the R wave due to the large differences in the individual physiological conditions of patients. The device and method of the present invention can be used for both prospective and retrospective ECG gating and adapt to the ECG signal processing schemes of various sequence scans.

[0009] To achieve the above object of the invention, an ECG gating signal processing device for nuclear magnetic resonance provided by an embodiment includes an acquisition module, a calculation module, and a transmission module, all of which are integrated on an integrated circuit board and are physically magnetically shielded;

[0010] The acquisition module is used to collect the superimposed signal of the ECG signal and the nuclear magnetic noise under the condition of physical magnetic shielding, and sequentially perform noise filtering and amplification on the superimposed signal to obtain the original ECG signal;

[0011] The calculation module is used to perform real-time frequency analysis on the original ECG signal, extract various ECG signal features in the frequency domain, reconstruct the mask vector after the real-time frequency analysis to filter the noise and reconstruct the time-domain R-wave signal, reconstruct the ECG signal based on the reconstructed time-domain R-wave signal and various ECG signal features, detect the R-wave trigger point based on the reconstructed ECG signal and generate a gating signal, and also perform missed triggering and false triggering detection on the R-wave trigger point in real time;

[0012] The transmission module is used to output the reconstructed ECG signal, gating signal and various ECG signal features to the nuclear magnetic resonance system, and supports matching different gating trigger conditions based on the magnetic resonance scanning sequence.

[0013] Preferably, the superimposed signal is subjected to noise filtering and amplification in sequence to obtain the original ECG signal, including:

[0014] The superimposed signal collected by the integrated circuit board is first subjected to front-end RF and anti-aliasing hardware low-pass filtering to filter out RF and static magnetic field coupling interference, and then the filtered signal is enhanced by the back-end modulation and amplification circuit to obtain the original ECG signal and output it.

[0015] Preferably, the original ECG signal is subjected to real-time frequency analysis to extract various ECG signal features in the frequency domain, including:

[0016] Perform real-time frequency analysis on the original ECG signal, convert the original ECG signal into the frequency domain to obtain a frequency domain signal, identify the main frequency of the ECG signal and the noise frequency of the nuclear magnetic noise based on the frequency domain signal in the frequency domain, and then retain the main frequency of the ECG signal to obtain a filtered signal;

[0017] Multiple numerical calculations are performed on the original ECG signal and the filtered signal in the frequency domain to extract various ECG signal features, including ECG morphology features, heart rate features, signal amplitude features, ECG signal power spectrum density features, and QRS wave peak energy features.

[0018] Preferably, reconstructing the mask vector after real-time frequency analysis to filter noise and reconstruct the time domain R-wave signal comprises:

[0019] After real-time frequency analysis, in the frequency domain, according to the main frequency of the identified ECG signal and the noise frequency of the nuclear magnetic noise, by comparing the clean ECG signal under normal ECG and the interference ECG signal under nuclear magnetic noise, the R wave frequency with the highest amplitude and the most significant ECG signal characteristics in the clean ECG signal is screened out as the R wave frequency center, and a mask vector is constructed based on the R wave frequency center through the similarity criterion and under the constraint condition, wherein the constraint condition is: retain the ECG-related bands as much as possible, and remove the noise bands in a targeted manner;

[0020] Perform a linear integration operation on the frequency-domain signal in the frequency domain according to the mask vector, remove the noise frequency band, and then convert it to the time domain to reconstruct the time-domain R-wave signal.

[0021] Preferably, detecting the R-wave trigger point based on the reconstructed electrocardiogram signal and generating a gating signal includes:

[0022] Automatically record the relevant features of the R-wave signal based on the reconstructed electrocardiogram signal, where the relevant features of the R-wave signal include the R-wave peak value, R-wave morphology, R-wave frequency, RR interval, R-wave slope, R-wave abundance, relevant spectrum, and relevant power spectrum;

[0023] Detecting the R-wave trigger point based on the relevant features of the R-wave signal and generating a gating signal includes using a sliding window to perform a weighted average on the relevant features of the R-wave signal within the window as the R-wave trigger point threshold, and taking the R-wave signal that exceeds the R-wave trigger point threshold as the R-wave trigger point and generating a gating signal.

[0024] Preferably, perform real-time detection of missed triggering and false triggering of the R-wave trigger point, including:

[0025] During the detection process, dynamically judge the phenomenon of missed triggering or false triggering of the R-wave trigger point in real time. When missed triggering or false triggering occurs, start the error correction program, backtrack and correct the reconstructed electrocardiogram signal data, match the R-wave peak on the stored data domain through a higher R-wave trigger point threshold, judge the interference noise and the normal R-wave peak, and update the relevant features of the R-wave signal with a certain weight. According to the summarized relevant features of the R-wave signal, especially the relevant features of the R-wave signal affected by noise, find the next correct R-wave trigger point.

[0026] Preferably, when transmitting data, carry the gating trigger method, output the reconstructed electrocardiogram signal and gating signal according to the predetermined sampling rate to implement the trigger gating signal; at the same time, the supported magnetic resonance scanning sequences include a custom prospective gating sequence and a retrospective gating sequence, and these two gating sequences match the custom gating trigger conditions to trigger the gating signal.

[0027] Preferably, in the nuclear magnetic resonance environment, use three-layer magnetic shielding for the front-end integrated circuit board to achieve physical magnetic shielding, including:

[0028] Cover the integrated circuit board with a non-magnetic shielding cover, wrap the integrated circuit board covered with the non-magnetic shielding cover with an anti-magnetic shielding film for shielding, and then place the integrated circuit board wrapped with the anti-magnetic shielding film in a shielding box, so that the collected original electrocardiogram signal and the calculation process and transmission process of the electrocardiogram signal are all protected by three-layer magnetic shielding.

[0029] To achieve the above-mentioned invention purpose, the embodiment also provides a method for processing an electrocardiogram gating signal under nuclear magnetic resonance. The method uses the above-mentioned device and includes the following steps:

[0030] The acquisition module is used to acquire the superimposed signal of the electrocardiogram signal and the nuclear magnetic noise under the condition of physical magnetic shielding, and the superimposed signal is sequentially subjected to noise filtering and amplification and then used as the original electrocardiogram signal;

[0031] The calculation module is used to perform real-time frequency analysis on the original electrocardiogram signal, extract various electrocardiogram signal features in the frequency domain, reconstruct a mask vector to filter noise and reconstruct the time-domain R-wave signal after real-time frequency analysis, reconstruct the electrocardiogram signal based on the reconstructed time-domain R-wave signal and various electrocardiogram signal features, detect the R-wave trigger point based on the reconstructed electrocardiogram signal and generate a gating signal, and at the same time, perform missed trigger and false trigger detection on the R-wave trigger point in real time;

[0032] The transmission module is used to output the reconstructed electrocardiogram signal, the gating signal and various electrocardiogram signal features to the nuclear magnetic resonance system, and at the same time support matching different gating trigger conditions based on the magnetic resonance scanning sequence.

[0033] To achieve the above invention purpose, the embodiment also provides a computing device, including a memory and one or more processors, wherein executable code is stored in the memory, and when the one or more processors execute the executable code, it is used to implement the above electrocardiogram gating signal processing method for nuclear magnetic resonance.

[0034] Compared with the prior art, the beneficial effects of the present invention at least include:

[0035] In the present invention, the acquisition module, the calculation module, and the transmission module integrated on the circuit board are all physically magnetically shielded to reduce the influence of nuclear magnetic resonance at the physical level. On this basis, the original electrocardiogram signal is converted into the frequency domain, and a mask vector is reconstructed based on the electrocardiogram signal in the frequency domain to filter noise and reconstruct the time-domain R-wave signal, reducing the noise influence at the algorithm level. Then, a complete electrocardiogram signal is reconstructed based on the electrocardiogram signal features and the reconstructed time-domain R-wave signal, and the R-wave positioning and detection are realized based on this to trigger the electrocardiogram gating, which can improve the accuracy of R-wave positioning, reduce the influence of magnetic resonance, and since the whole process is based on the patient's original electrocardiogram signal, it can realize accurate R-wave positioning for individual differences, and then generate an accurate electrocardiogram gating signal.

[0036] The present invention can be used not only for prospective electrocardiogram gating but also for retrospective electrocardiogram gating, reducing the time required for relevant practitioners to mark the R wave, adopting a combined filtering method to remove interference signals in the electrocardiogram signal and analyze the electrocardiogram signal noise to reconstruct a clean electrocardiogram signal, and at the same time summarizing electrocardiogram-related features, continuously detecting false triggers and missed triggers, and improving the trigger accuracy of nuclear magnetic resonance imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0038] Figure 1 It is a schematic structural diagram of an electrocardiogram gating signal processing device for nuclear magnetic resonance provided by an embodiment;

[0039] Figure 2 It is a schematic diagram of physically magnetically shielding an integrated circuit board provided by an embodiment;

[0040] Figure 3 It is a schematic diagram of an electrocardiogram signal with nuclear magnetic resonance scanning noise provided by an embodiment;

[0041] Figure 4 It is the extracted and reconstructed time-domain R-wave signal provided by an embodiment;

[0042] Figure 5 It is the separated noise signal provided by an embodiment;

[0043] Figure 6 It is an example diagram of the original electrocardiogram signal and the reconstructed electrocardiogram signal, including the trigger signal, provided by an embodiment;

[0044] Figure 7 It is an example diagram of R-wave error detection and correction provided by an embodiment;

[0045] Figure 8 It is a flowchart of an electrocardiogram gating signal processing method for nuclear magnetic resonance provided by an embodiment. Detailed implementation manners

[0046] To make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific implementation manners described herein are only used to explain the present invention and do not limit the protection scope of the present invention.

[0047] The inventive concept of the present invention is as follows: To solve the problems in the related art that the electrocardiogram signal waveform is affected by magnetic resonance and generates a large amount of noise, making it impossible to accurately locate the R wave, and the R wave is difficult to locate due to the large differences in the individual physiological conditions of patients. The embodiments of the present invention provide an electrocardiogram gating signal processing solution for nuclear magnetic resonance, which can accurately locate the R wave in the electrocardiogram signal under the nuclear magnetic resonance interference scenario, and thus improve the accuracy of generating the electrocardiogram gating signal based on the R wave.

[0048] Such as Figure 1As shown in the figure, an electrocardiogram gating signal processing device provided by an embodiment includes an acquisition module, a calculation module, and a transmission module. The device collects the electrocardiogram signals of the subject in a three-lead manner. The acquisition module and the calculation module are placed beside the subject, and the transmission module is connected to an optical fiber to extend out of the MRI detection room. Finally, the output port of the transmission module is connected to the computer connected to the MRI device to trigger the MRI gating sequence and transmit and record relevant data. When the device is turned on, due to hardware power-on restrictions, the data of a preset time needs to be discarded, and then the electrocardiogram signals and gating signals can be output.

[0049] In the embodiment, the acquisition module, the calculation module, and the transmission module are all integrated on an integrated circuit board and are physically magnetically shielded. Specifically, as Figure 2 shown, in the nuclear magnetic resonance environment, the integrated circuit board at the front end uses three-layer magnetic shielding to achieve physical magnetic shielding, including: covering the integrated circuit board with a non-magnetic shielding cover, wrapping the integrated circuit board covered with the non-magnetic shielding cover with an anti-magnetic shielding film for shielding, and then placing the integrated circuit board wrapped with the anti-magnetic shielding film in a shielding box, so that the superimposed signal of the electrocardiogram signal and nuclear magnetic noise used by the integrated circuit board is protected by the three-layer magnetic shielding. At the same time, the subsequent data analysis and processing process and the transmission process are all protected by the three-layer magnetic shielding, reducing the partial interference of nuclear magnetic noise at the physical level.

[0050] In the embodiment, the acquisition module is used to collect the superimposed signal of the electrocardiogram signal and nuclear magnetic noise in the nuclear magnetic resonance environment according to a preset sampling rate under the condition of physical magnetic shielding, and perform noise filtering and amplification on the superimposed signal in sequence as the original electrocardiogram signal. In this way, the initial noise reduction and enhancement functions of the signal are completed, and the magnetic shielding and enhancement functions of the electrocardiogram signal at the hardware front end in the nuclear magnetic resonance environment are realized, improving the signal-to-noise ratio.

[0051] Specifically, performing noise filtering and amplification on the superimposed signal in sequence includes: first performing front-end radio frequency and anti-aliasing hardware low-pass filtering on the superimposed signal collected through the integrated circuit board to filter out radio frequency and static magnetic field coupling interference, and then enhancing the filtered signal through a rear-end modulation amplification circuit to obtain the original electrocardiogram signal (i.e., the collected ECG signal) and output it, as Figure 3 shown. Specifically, when filtering, the superimposed signal is made hierarchical and combined, and multiple filters are applied through different weight parameters for multiple filtering. The main purposes include but are not limited to: restricting the amplitude of the noise; filtering out some high-frequency noise without losing the electrocardiogram signal.

[0052] In an embodiment, the calculation module is used to perform real-time frequency analysis on the original ECG signal and extract various ECG signal features in the frequency domain. Specifically, the original ECG signal is subjected to real-time frequency analysis, that is, the original ECG signal can be converted to the frequency domain to obtain a frequency domain signal through Fourier transform, and the main frequency of the ECG signal and the noise frequency of the nuclear magnetic noise are identified in the frequency domain based on the frequency domain signal. The energy of the ECG QRS wave group is mainly concentrated in 0~40hz, and the peak energy is concentrated in 8~16hz, which is regarded as the main frequency of the ECG signal. Then, the filtered signal is obtained with the goal of retaining the main frequency of the ECG signal. Multiple numerical calculations are performed in the frequency domain based on the filtered signal and the original ECG signal to extract various ECG signal features, wherein various ECG signal features include but are not limited to ECG morphology features, heartbeat frequency features, signal amplitude features, ECG signal power spectrum density features, QRS wave peak energy features, etc.

[0053] The calculation module reconstructs the mask vector after real-time frequency analysis to filter the noise and reconstruct the time domain R wave signal, specifically including: after real-time frequency analysis, in the frequency domain, according to the main frequency of the identified ECG signal and the noise frequency of the nuclear magnetic noise, by comparing the clean ECG signal under normal ECG and the interference ECG signal under nuclear magnetic noise, the R wave frequency with the highest amplitude and the most significant ECG signal characteristics in the clean ECG signal is selected as the R wave frequency center, and an adaptive mask vector is constructed based on the R wave frequency center by similarity criteria and under constraints, wherein the constraints are: retain ECG related bands as much as possible, and specifically remove the noise bands. When constructing the mask vector, according to the similarity criterion, the ECG R wave band where the ECG frequency based on the R wave frequency center mainly exists is selected, and the mask vector is designed according to the clean ECG signal, and the ECG R wave band is assigned a value of 1, the noise band is assigned a value of 0, and the aliasing band transitioning between the ECG R wave band and the noise band is assigned a value in the interval of 0.1-0.9 to obtain the mask vector.

[0054] After obtaining the mask vector, a linear integration operation is performed on the frequency domain signal in the frequency domain according to the mask vector to mask the noise part to remove the noise frequency band, and then the signal is converted to the time domain through inverse Fourier transform to reconstruct the time domain R wave signal. Figure 4 To extract the reconstructed time domain R wave signal, Figure 5 is the noise signal separated based on the mask vector.

[0055] The calculation module also reconstructs the ECG signal based on the reconstructed time-domain R-wave signal and various ECG signal features. Specifically, the extracted time-domain R-wave signal is reversely calculated, and the complete ECG signal is reconstructed using various ECG signal features. Figure 6 The lower part, Figure 6 The upper part is the collected original ECG signal with noise.

[0056] The calculation module also detects the R-wave trigger point based on the reconstructed electrocardiogram signal and generates a gating signal. Different from the common Pan-Tompkins algorithm, the R-wave detection algorithm adopted in the embodiments of the present invention uses a multi-threshold method combining electrocardiogram features to generate a gating signal. Specifically, it automatically records the features related to the R-wave signal based on the reconstructed electrocardiogram signal, including but not limited to the R-wave peak value, R-wave morphology, R-wave frequency, RR interval, R-wave slope, R-wave abundance, correlation spectrum, correlation power spectrum, etc. Then, based on the features related to the R-wave signal, the R-wave trigger point is detected. Specifically, a sliding window is used to perform weighted averaging on the features related to the R-wave signal within the window as the next R-wave trigger point threshold, and the R-wave signal that detects exceeding the R-wave trigger point threshold is used as the R-wave trigger point and a gating signal is generated. As Figure 6 shown, the gating signal is represented by a red mark, and it can be seen that Figure 6 there is obvious magnetic field noise generated by the MRI scan, and the electrocardiogram is almost invisible due to being covered. During the entire acquisition process, the device of the present invention can continuously send the gating signal and the clean electrocardiogram signal. It should be noted that in this embodiment, a sampling rate of 500 hz is adopted, and all processing processes are completed within 2 ms and the output is completed.

[0057] While generating the gating signal, that is, during the real-time operation of the nuclear magnetic resonance device, the calculation module also performs real-time detection of missed triggering and false triggering of the R-wave trigger point. Specifically, it includes: during the detection process, dynamically judging the missed triggering or false triggering phenomenon of the R-wave trigger point in real time. When a missed triggering or false triggering occurs, an error correction program is started to backtrack and correct the reconstructed electrocardiogram signal data. On the stored data domain, the R-wave peak is matched through a higher R-wave trigger point threshold to judge the interference noise and the normal R-wave peak, and the features related to the R-wave signal are updated with a certain weight. According to the summarized features related to the R-wave signal, especially the features related to the R-wave signal affected by noise, the next correct R-wave trigger point is found.

[0058] The ability of the calculation module to provide missed triggering and false triggering backtracking correction is shown by Figure 7 shown. Figure 7 In the upper figure above is the original electrocardiogram signal affected by the nuclear magnetic resonance scan artifact collected, and in the lower figure is the filtered electrocardiogram signal. The output gating signal is marked with a red line. There is an incorrect judgment of the R-wave around 5 s. After detecting the error, it is corrected, and accurate electrocardiogram judgment is restored within two heartbeat cycles subsequently.

[0059] In an embodiment, the transmission module is used to output the reconstructed electrocardiogram (ECG) signal, gating signal, and various ECG signal features to the nuclear magnetic resonance (NMR) system, and at the same time supports matching different gating trigger conditions based on the magnetic resonance scanning sequence. Specifically, during data transmission, a gating trigger method is carried, and the reconstructed ECG signal and gating signal are output according to a predetermined sampling rate to trigger the gating signal; at the same time, the supported magnetic resonance scanning sequences include a custom prospective gating sequence and a retrospective gating sequence, and these two gating sequences match the custom gating trigger conditions to trigger the gating signal.

[0060] The device of the present invention can be used for prospective ECG gating, and at the same time can replace relevant practitioners to mark the R wave of the ECG signal. It is also applicable to retrospective ECG gating. During the test of this embodiment, a considerable number of different gating sequences were used for testing, and it can be set to adapt to a variety of ECG gating sequences.

[0061] The device of the present invention outputs the processed clean ECG signal and gating signal according to the required sampling rate, and at the same time supports matching the gating trigger conditions based on the magnetic resonance scanning sequence. By the present invention, the problems in the related art that the ECG signal is interfered by the NMR magnetic field noise, resulting in the inability to accurately locate the cardiac rhythm, and further affecting the triggering accuracy in magnetic resonance imaging are solved. At the same time, accurate R wave markings can be made in all collected ECG signals. It can be used not only for prospective ECG gating but also for retrospective ECG gating, reducing the time required for relevant practitioners to mark the R wave. It realizes the removal of interference signals in the ECG signal and the analysis of ECG signal noise by using a combined filtering method to reconstruct a clean ECG signal, summarizes ECG-related features, continuously detects false triggers and missed triggers, and improves the triggering accuracy of magnetic resonance imaging.

[0062] As Figure 8 shown, the embodiment also provides a method for processing an ECG gating signal under nuclear magnetic resonance, which uses the above-mentioned ECG gating signal processing device and includes the following steps:

[0063] S1, using the acquisition module to acquire the superimposed signal of the ECG signal and nuclear magnetic noise under the physical magnetic shielding condition, and sequentially performing noise filtering and amplification on the superimposed signal as the original ECG signal;

[0064] S2, using the calculation module to perform real-time frequency analysis on the original ECG signal, extract various ECG signal features in the frequency domain, reconstruct a mask vector to filter noise and reconstruct the time-domain R wave signal after real-time frequency analysis, reconstruct the ECG signal based on the reconstructed time-domain R wave signal and various ECG signal features, detect the R wave trigger point based on the reconstructed ECG signal and generate a gating signal, and at the same time, perform missed trigger and false trigger detection on the R wave trigger point in real time;

[0065] S3. Use the transmission module to output the reconstructed electrocardiogram (ECG) signal, gating signal, and various ECG signal characteristics to the nuclear magnetic resonance (NMR) system, and at the same time support matching different gating trigger conditions based on the magnetic resonance scanning sequence.

[0066] Based on the same inventive concept, the embodiment also provides a computing device, including a memory and one or more processors. An executable code is stored in the memory. When the one or more processors execute the executable code, it is used to implement the above-mentioned ECG gating signal processing method for NMR, specifically including the following steps:

[0067] S1. Use the acquisition module to acquire the superimposed signal of the ECG signal and nuclear magnetic noise under the physical magnetic shielding condition, and sequentially perform noise filtering and amplification on the superimposed signal as the original ECG signal.

[0068] S2. Use the calculation module to perform real-time frequency analysis on the original ECG signal, extract various ECG signal characteristics in the frequency domain, reconstruct a mask vector to filter noise and reconstruct the time-domain R-wave signal after real-time frequency analysis, reconstruct the ECG signal based on the reconstructed time-domain R-wave signal and various ECG signal characteristics, detect the R-wave trigger point based on the reconstructed ECG signal and generate a gating signal, and at the same time perform missed trigger and false trigger detection on the R-wave trigger point in real time.

[0069] S3. Use the transmission module to output the reconstructed electrocardiogram (ECG) signal, gating signal, and various ECG signal characteristics to the nuclear magnetic resonance (NMR) system, and at the same time support matching different gating trigger conditions based on the magnetic resonance scanning sequence.

[0070] For the computing device provided by the embodiment, at the hardware level, in addition to including a processor and a memory, it also includes an internal bus, a network interface, memory, and other hardware required for other services. The memory is a non-volatile memory. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the above-mentioned ECG gating signal processing method described in S1 - S3. Of course, in addition to the software implementation method, the present invention does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, and can also be hardware or a logic device.

[0071] In the technical solution of the present invention, a denoising and restoration operation is performed on the original ECG signal, and multiple numerical calculations are carried out. The characteristics of various ECG signals are adaptively summarized, and multiple threshold values for triggering the next R-wave signal are adaptively summarized according to the characteristics, and then the R-wave trigger point is detected to trigger the ECG gating. The present invention does not require human-computer interaction to set thresholds, eliminating the influence of inaccurate triggering caused by artificially setting thresholds when signal fluctuations occur.

[0072] The specific embodiments described above have elaborated in detail the technical solutions and beneficial effects of the present invention. It should be understood that the above is only the most preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, supplements, equivalent substitutions, etc. made within the principle scope of the present invention shall be included within the protection scope of the present invention.

Claims

1. A device for processing ECG gating signals under nuclear magnetic resonance, characterized in that: The acquisition module, calculation module, and transmission module are all integrated on the integrated circuit board and are physically magnetically shielded; The acquisition module is used to acquire the superposition signal of the ECG signal and the nuclear magnetic noise under the physical magnetic shielding condition, and perform noise filtering and amplification on the superposition signal in sequence as the original ECG signal; The calculation module is used to perform real-time frequency analysis on the original ECG signal, extract various ECG signal features in the frequency domain, reconstruct the mask vector after the real-time frequency analysis to filter the noise and reconstruct the time-domain R-wave signal, reconstruct the ECG signal based on the reconstructed time-domain R-wave signal and various ECG signal features, detect the R-wave trigger point based on the reconstructed ECG signal and generate a gating signal, and also perform missed triggering and false triggering detection on the R-wave trigger point in real time; The transmission module is used to output the reconstructed ECG signal, gating signal and various ECG signal features to the nuclear magnetic resonance system, and supports matching different gating trigger conditions based on the magnetic resonance scanning sequence.

2. The ECG gating signal processing device for nuclear magnetic resonance according to claim 1, characterized in that: The superimposed signal is noise filtered and amplified in sequence to obtain the original ECG signal, including: The superimposed signal collected by the integrated circuit board is first subjected to front-end RF and anti-aliasing hardware low-pass filtering to filter out RF and static magnetic field coupling interference, and then the filtered signal is enhanced by the back-end modulation and amplification circuit to obtain the original ECG signal and output it.

3. The ECG gating signal processing device for nuclear magnetic resonance according to claim 1, characterized in that: Perform real-time frequency analysis on the original ECG signal and extract various ECG signal features in the frequency domain, including: Perform real-time frequency analysis on the original ECG signal, convert the original ECG signal into the frequency domain to obtain a frequency domain signal, identify the main frequency of the ECG signal and the noise frequency of the nuclear magnetic noise based on the frequency domain signal in the frequency domain, and then retain the main frequency of the ECG signal to obtain a filtered signal; Multiple numerical calculations are performed on the original ECG signal and the filtered signal in the frequency domain to extract various ECG signal features, including ECG morphology features, heart rate features, signal amplitude features, ECG signal power spectrum density features, and QRS wave peak energy features.

4. The ECG gating signal processing device for nuclear magnetic resonance according to claim 3, characterized in that: Reconstruct the mask vector after real-time frequency analysis to filter the noise and reconstruct the time-domain R-wave signal, including: After real-time frequency analysis, in the frequency domain, according to the main frequency of the identified ECG signal and the noise frequency of the nuclear magnetic noise, by comparing the clean ECG signal under normal ECG and the interference ECG signal under nuclear magnetic noise, the R wave frequency with the highest amplitude and the most significant ECG signal characteristics in the clean ECG signal is screened out as the R wave frequency center, and a mask vector is constructed based on the R wave frequency center through the similarity criterion and under the constraint condition, wherein the constraint condition is: retain the ECG-related bands as much as possible, and remove the noise bands in a targeted manner; According to the mask vector, a linear integration operation is performed on the frequency domain signal in the frequency domain, and after the noise frequency band is removed, it is converted to the time domain and reconstructed into the time domain R wave signal.

5. The ECG gating signal processing device for nuclear magnetic resonance according to claim 3, characterized in that: Detect R wave trigger points based on reconstructed ECG signals and generate gating signals, including: Automatically record R wave signal related features based on the reconstructed ECG signal, wherein the R wave signal related features include R wave peak, R wave morphology, R wave frequency, RR interval, R wave slope, R wave abundance, related spectrum, and related power spectrum; Based on the relevant characteristics of the R wave signal, the R wave trigger point is detected and the gating signal is generated, including using a sliding window to perform weighted averaging on the relevant characteristics of the R wave signal in the window as the R wave trigger point threshold, and detecting the R wave signal that exceeds the R wave trigger point threshold as the R wave trigger point and generating a gating signal.

6. The ECG gating signal processing device for nuclear magnetic resonance according to claim 1, characterized in that: Real-time detection of missed triggering and false triggering of R wave trigger points, including: During the detection process, the missed triggering or false triggering of the R wave trigger point is judged in real time and dynamically. When a missed trigger or false trigger occurs, the error correction program is started to retrospectively correct the reconstructed ECG signal data. The R wave peak is matched with a higher R wave trigger point threshold in the stored data domain, the interference noise and the normal R wave peak are judged, and the R wave signal related features are updated with a certain weight. According to the summarized R wave signal related features, the next correct R wave trigger point is found.

7. The ECG gating signal processing device for nuclear magnetic resonance according to claim 6, characterized in that: The method of finding the next correct R wave trigger point based on the summarized R wave signal related features includes: According to the relevant characteristics of the interfered R wave signal under noise, find the next correct R wave trigger point.

8. The electrocardiogram gating signal processing device for nuclear magnetic resonance according to claim 1, characterized in that: During data transmission, it is equipped with a gated trigger method to output the reconstructed ECG signal and gating signal according to the predetermined sampling rate to trigger the gating signal; the supported magnetic resonance scanning sequences include customized prospective gating sequences and retrospective gating sequences. These two gating sequences match the customized gating trigger conditions to trigger the gating signal.

9. The ECG gating signal processing device for nuclear magnetic resonance according to claim 1, characterized in that: In the nuclear magnetic resonance environment, three layers of magnetic shielding are used to achieve physical magnetic shielding for the front-end integrated circuit board, including: The integrated circuit board is covered with a non-magnetic shielding cover, and the integrated circuit board covered with the non-magnetic shielding cover is wrapped and shielded by an anti-magnetic shielding film, and then the integrated circuit board wrapped by the anti-magnetic shielding film is placed in a shielding box, so that the collected original electrocardiogram signal and the calculation and transmission processes of the electrocardiogram signal are protected by three layers of magnetic shielding.

10. A method for processing ECG gating signals under nuclear magnetic resonance, characterized in that: The method uses the device according to any one of claims 1 to 9, comprising the following steps: The acquisition module is used to collect the superposition signal of the ECG signal and the nuclear magnetic noise under the condition of physical magnetic shielding, and the superposition signal is successively subjected to noise filtering and amplification to be used as the original ECG signal; The computing module is used to perform real-time frequency analysis on the original ECG signal, extract various ECG signal features in the frequency domain, reconstruct the mask vector after the real-time frequency analysis to filter the noise and reconstruct the time-domain R-wave signal, reconstruct the ECG signal based on the reconstructed time-domain R-wave signal and various ECG signal features, detect the R-wave trigger point based on the reconstructed ECG signal and generate a gating signal, and also perform missed triggering and false triggering detection on the R-wave trigger point in real time; The transmission module is used to output the reconstructed ECG signal, gating signal and various ECG signal features to the MRI system, while supporting the matching of different gating trigger conditions based on the MRI scanning sequence.

11. A computing device comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the one or more processors execute the executable code, they are used to implement the method for processing ECG gating signals under nuclear magnetic resonance as described in claim 10.

Images