MRI Real-Time Triggering Method, System, Terminal and Medium Based on SCG-PPG Signals

Abstract

The present invention discloses an MRI real-time triggering method, system, terminal and medium based on SCG-PPG signals. The method includes: capturing the periodic motion of the heart to obtain an SCG signal, and detecting the dynamic changes in blood flow to obtain a PPG signal; preprocessing the SCG signal and the PPG signal to determine the periodic changes in the SCG signal and the PPG signal, and calibrating the trigger points for each cardiac cycle; establishing a mathematical model based on the trigger points for each cardiac cycle, and adjusting the optimal MRI triggering timing in real time. The present invention effectively avoids the problems in the traditional MRI triggering method, such as signal anomalies that are likely to occur, affecting the triggering accuracy, and physiological delays that lead to untimely triggering, affecting the imaging quality. The present invention greatly improves the reliability of data acquisition, provides high-quality cardiac image data, and has important application value.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear magnetic resonance imaging, and particularly to an MRI real-time triggering method, system, terminal and medium based on SCG-PPG signals. Background Art

[0002] Magnetic Resonance Imaging (MRI) is a high-resolution and non-invasive medical imaging technology that plays an important role in the diagnosis and evaluation of complex lesions such as Cardiovascular Diseases (CD). However, during the MRI scanning process, the physiological movements of patients (such as cardiac pulsation, blood flow, and respiratory movement) can cause motion artifacts. Especially in a high magnetic field strength magnetic resonance (such as 3T and above) environment, these artifacts have a significant impact on image quality and diagnostic accuracy. To improve the imaging quality, it is usually necessary to synchronize data acquisition during a specific physiological cycle of the patient (such as the cardiac systolic or diastolic phase), achieve precise timing trigger control, thereby effectively reducing artifacts and ensuring the reliability of diagnostic results. Therefore, developing an efficient MRI real-time triggering technology is of great significance for improving the imaging quality.

[0003] Currently, the traditional method widely used for MRI triggering is the triggering technology based on the Electrocardiogram (ECG) signal. ECG realizes the synchronization of data acquisition by detecting the electrical activity of the heart and has a high application maturity. However, this technology has obvious deficiencies in practical applications. On the one hand, pathological conditions such as Arrhythmia and Myocardial Infarction may cause abnormal electrocardiogram signals, affecting the triggering accuracy. In addition, in a high magnetic field strength magnetic resonance environment, the ECG signal is easily affected by electromagnetic interference, resulting in signal fluctuations or losses, further reducing the triggering stability. These problems limit the applicability of the traditional ECG triggering technology in complex clinical environments. To address the problems of ECG gating, the Photoplethysmography (PPG) signal measured at the fingertip is often used as an alternative. However, due to the relatively long distance between the PPG signal and the heart, the generated trigger signal has a physiological delay of up to several hundred milliseconds from the real cardiac activity, which cannot reserve enough time for subsequent MR data acquisition, further limiting the imaging quality. It can be seen that the traditional MRI triggering method is prone to signal abnormalities, affecting the triggering accuracy, and there is a problem of untimely triggering due to physiological delay, affecting the imaging quality. Summary of the Invention

[0004] The present invention provides a real-time triggering method, system, terminal and medium for MRI based on SCG-PPG signals. The technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present invention provides a real-time triggering method for MRI based on SCG-PPG signals, wherein the method includes:

[0006] Capture the periodic motion of the heart to obtain an SCG signal, and detect the dynamic changes in blood flow to obtain a PPG signal;

[0007] Preprocess the SCG signal and the PPG signal to determine the periodic changes in the SCG signal and the PPG signal, and calibrate the trigger points for each cardiac cycle;

[0008] Establish a mathematical model for real-time calculation of the optimal MRI triggering timing based on the trigger points for each cardiac cycle, and adjust the optimal MRI triggering timing in real time.

[0009] In one implementation, capturing the periodic motion of the heart to obtain an SCG signal includes:

[0010] Position the defocus camera at the fourth rib site in the lower left of the chest cavity and cooperate with the laser system for irradiation;

[0011] Adjust the configuration of the defocus camera to collect speckle images;

[0012] Use the optical flow method to deeply analyze the collected speckle images, calculate the motion amplitudes of the speckles in the X-axis and Y-axis directions, and obtain a motion amplitude sequence;

[0013] Divide the motion amplitude sequence to determine the motion angle of each segment, where the motion angle reflects the direction of heart vibration;

[0014] Based on the motion angle of each segment, obtain the SCG signal.

[0015] In one implementation, adjusting the configuration of the defocus camera includes:

[0016] Adjust the defocus camera so that the distance between the focal plane and the chest cavity plane is greater than the distance between the lens and the focal plane.

[0017] In one implementation, detecting the dynamic changes in blood flow to obtain a PPG signal includes:

[0018] Cover the key parts of the face with an optical sensor, emit a synchronous signal of a periodic square wave, and collect images;

[0019] Extract the part containing skin pixels from the collected images to obtain a skin area;

[0020] Based on the skin area, determine the G-channel image sequence, and extract the PPG signal from the G-channel image sequence.

[0021] In one implementation, the preprocessing includes: filtering, detrending, and normalization; wherein, the filtering is used to remove high-frequency noise and / or low-frequency noise in the SCG signal and the PPG signal; the detrending is used to remove the linear trend and / or non-linear trend in the SCG signal and the PPG signal; the normalization is used to map the amplitudes of the SCG signal and the PPG signal to a unified standard range.

[0022] In one implementation, determining the periodic changes of the SCG signal and the PPG signal includes:

[0023] Perform real-time peak detection processing on the SCG signal and the PPG signal to capture the periodic changes of the SCG signal and the PPG signal;

[0024] The real-time peak detection processing includes: sliding time window processing and peak detection. Among them, the sliding time window processing includes: respectively dividing the long-term continuous SCG signal and PPG signal into multiple segments, and gradually processing each fixed-length data subset in the SCG signal and the PPG signal;

[0025] The peak detection includes: respectively determining the local maxima in the SCG signal and the PPG signal to obtain the potential peak positions of the SCG signal and the PPG signal; obtaining a preset height threshold, and using the height threshold to screen the local maxima to obtain the screened local maxima; determining the minimum peak spacing, and analyzing the screened local maxima based on the minimum peak spacing to obtain the real-time peaks in the SCG signal and the PPG signal and determine the cardiac cycle.

[0026] In one implementation, the real-time adjustment of the optimal MRI triggering timing includes:

[0027] Obtain the physiological state information, and the constructed mathematical model combines the physiological state information to real-time adjust the optimal MRI triggering timing.

[0028] In a second aspect, an embodiment of the present invention further provides an MRI real-time triggering system based on SCG-PPG signals. Among them, the system is used to implement the steps of the above-mentioned MRI real-time triggering method based on SCG-PPG signals. The system includes:

[0029] The SCG-PPG signal acquisition module is used to capture the periodic motion of the heart to obtain the SCG signal, and detect the dynamic changes of blood flow to obtain the PPG signal;

[0030] A trigger point calibration module, which is used to preprocess the SCG signal and the PPG signal, determine the periodic changes of the SCG signal and the PPG signal, and calibrate the trigger points of each cardiac cycle;

[0031] An MRI trigger timing optimization module, which is used to establish a mathematical model for real-time calculation of the optimal MRI trigger timing based on the trigger points of each cardiac cycle, and adjust the optimal MRI trigger timing in real time.

[0032] In a third aspect, an embodiment of the present invention further provides a terminal. The terminal includes a memory, a processor, and an MRI real-time trigger program based on SCG-PPG signals stored in the memory and executable on the processor. When the processor executes the MRI real-time trigger program based on SCG-PPG signals, the steps of the MRI real-time trigger method based on SCG-PPG signals in any one of the above solutions are implemented.

[0033] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium. A MRI real-time trigger program based on SCG-PPG signals is stored on the computer-readable storage medium. When the MRI real-time trigger program based on SCG-PPG signals is executed by a processor, the steps of the MRI real-time trigger method based on SCG-PPG signals in any one of the above solutions are implemented.

[0034] Beneficial effects: The present invention provides a method for real-time triggering of MRI based on SCG-PPG signals. The present invention first captures the periodic movement of the heart to obtain the SCG signal, and detects the dynamic changes of blood flow to obtain the PPG signal. Then, the SCG signal and the PPG signal are preprocessed to determine the periodic changes of the SCG signal and the PPG signal, and the trigger points of each cardiac cycle are calibrated. Finally, a mathematical model for real-time calculation of the optimal MRI trigger timing is established based on the trigger points of each cardiac cycle, and the optimal MRI trigger timing is adjusted in real time. The SCG signal collected in the present invention can accurately capture the minute vibrations caused by cardiac mechanical activities, and the PPG signal can accurately reflect the dynamic changes of blood flow, breaking through the limitations of traditional MRI trigger technologies in pathological states, effectively avoiding the problem that the trigger is not timely due to physiological delay and affecting the imaging quality. Moreover, by fusing the SCG signal and the PPG signal, the present invention can provide more comprehensive and accurate physiological information, effectively avoiding the problem that signal abnormalities are prone to occur in traditional MRI trigger technologies and affecting the trigger accuracy, greatly improving the reliability of data acquisition, providing high-quality cardiac image data, and having important application value. Description of the Drawings

[0035] Figure 1 It is a flowchart of a preferred embodiment of the method for real-time triggering of MRI based on SCG-PPG signals provided by an embodiment of the present invention.

[0036] Figure 2 Schematic diagram of MRI triggered imaging provided by an embodiment of the present invention.

[0037] Figure 3 Schematic diagram of the acquisition of SCG-PPG signals provided by an embodiment of the present invention.

[0038] Figure 4 Schematic diagram of the architecture of an MRI real-time triggering system based on SCG-PPG signals provided by an embodiment of the present invention.

[0039] Figure 5 Principle block diagram of the terminal provided by an embodiment of the present invention. Detailed implementation manners

[0040] To make the objectives, technical solutions and effects of the present invention clearer and more definite, the following further describes the present invention in detail with reference to the accompanying drawings and by way of examples. It should be understood that the specific examples described herein are only used to explain the present invention and are not used to limit the present invention.

[0041] The flowcharts shown in the accompanying drawings are only illustrative examples, and do not necessarily include all the contents, operations or steps, nor do they necessarily need to be executed in the described order. For example, some operations or steps can also be decomposed, combined or partially merged, so the actual execution order may be changed according to the actual situation.

[0042] It should be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0043] It should be understood that, in order to facilitate a clear description of the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. For example, the first control information and the second control information are only used to distinguish different control information, and do not limit their order.

[0044] Those skilled in the art can understand that the terms such as "first" and "second" do not limit the quantity and execution order, and the terms such as "first" and "second" do not necessarily mean different.

[0045] It should also be understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0046] The MRI real-time triggering method based on SCG-PPG signals of this embodiment is applied to a terminal, which may be a computer, a smart TV, a mobile phone or other intelligent terminal products. Figure 1 As shown in , the MRI real-time triggering method based on SCG-PPG signal of this embodiment includes the following steps:

[0047] Step S100: Capture the periodic movement of the heart to obtain an SCG signal, and detect the dynamic changes of blood flow to obtain a PPG signal.

[0048] In this embodiment, combined with Figure 2 As shown, it is necessary to obtain the seismocardiogram (SCG) signal and the photoplethysmography (PPG) signal in real time to provide high-quality raw data to support subsequent in-depth analysis and precise processing. To this end, this embodiment is equipped with a high-precision sensor to ensure the accuracy and real-time performance of signal acquisition, and its sampling frequency is usually set high enough to capture rapidly changing physiological signals. In this embodiment, the acquisition of SCG signals mainly relies on the optical model of the defocus camera and the laser system to significantly amplify the tiny vibrations caused by the activity of the heart machine, and these vibrations reflect detailed information about the mechanical cycle of the heart. The acquisition of PPG signals can detect the dynamic changes of blood flow with the help of optical sensors (such as cameras), thereby providing pulse wave characteristics and relevant information about hemodynamics. Specifically, the acquisition of SCG signals is based on the principle of optical interference. The beating of the heart will cause a tiny movement of the chest cavity, which will cause the interference field of the laser reflected on the chest surface to change. This change is manifested as a speckle displacement on the camera imaging plane. By capturing the changes in these speckle displacements with a high-speed camera, the SCG signal can be extracted. The acquisition of PPG signals is based on the principle of optical reflection. The light source is used to illuminate the skin surface and detect changes in reflected light to extract the pulse wave signal. When the light source is irradiated to the skin, part of the light is absorbed by the skin tissue, and part of the light is reflected back to the optical sensor. The changes in light absorption caused by blood flow will be detected by changes in the intensity of the reflected light. By extracting the alternating current (AC) signal of the reflected light, the characteristics of the pulse wave can be obtained. In this embodiment, the real-time acquisition of SCG signals and PPG signals provides a solid foundation for subsequent precise physiological data analysis triggered by MRI gating.

[0049] In practical applications, the process of SCG signal extraction first requires precise adjustment of the configuration of the defocus camera and the laser system. Figure 3As shown, the camera 2 is precisely positioned at the fourth rib site in the lower left of the chest cavity and cooperates with the laser system to irradiate this area. The camera 2 in this embodiment is a defocus camera. To improve the detection accuracy of the heart motion signal, the configuration of the defocus camera is adjusted so that the distance L1 between the focal plane and the chest cavity plane is greater than the distance L2 between the lens and the focal plane. This configuration of the defocus camera significantly magnifies the laser speckle image caused by heart vibrations, enabling tiny heart motions to be clearly presented. The ratio of L1 / L2 can be used to represent the defocus degree of the camera, and a higher defocus degree corresponds to a more significant signal amplification effect. This configuration enables the defocus camera to continuously acquire speckle images at a frame rate of at least 200 fps and a resolution of 400×300, providing high-quality data for subsequent signal processing. After the speckle images are acquired, in this embodiment, the optical flow method can be used to deeply analyze the captured speckle images, accurately calculate the motion amplitudes of the speckles in the X-axis and Y-axis directions, and obtain the motion amplitude sequence. Then, the motion amplitude sequence is reasonably divided by the sliding window method, and the motion angle of each segment is calculated. . This motion angle represents the direction of heart vibration and can be used to synthesize the final SCG signal. This SCG signal can accurately reflect the periodic motion of the heart and becomes the key to triggering the synchronous imaging of the Cardiovascular Magnetic Resonance (CMR) device. Through this high-precision signal extraction and analysis method, the SCG signal can not only achieve the accurate reconstruction of heart motion but also provide stable and reliable data information for clinical applications, with broad application prospects.

[0050] In practical applications, the extraction of the PPG signal first requires ensuring the precise adjustment of the position of the optical sensor (i.e., the camera) to clearly capture the facial skin area of the subject. For this purpose, in this embodiment, the camera must be precisely positioned to ensure that it covers key parts of the face, such as the forehead, cheeks, or chin, etc., and these areas can effectively reflect the changes in blood flow. The role of the synchronization signal is particularly crucial. In this embodiment, a synchronization signal in the form of a periodic square wave can be emitted to ensure that the camera and other devices can achieve precise temporal synchronization during image acquisition. The frame rate of this synchronization signal is generally set to not less than 200 fps to ensure the efficiency and accuracy of image acquisition and avoid signal loss or distortion caused by a low frame rate. After completing image acquisition, first, the part containing skin pixels needs to be extracted from the acquired images to obtain the skin area. The accuracy of this step directly affects the quality of subsequent signal extraction, so efficient image processing techniques must be adopted. After successfully extracting the skin area, in this embodiment, a G-channel image sequence with stronger pulsatility is obtained, and the final PPG signal is extracted by analyzing the G-channel image sequence. This PPG signal reflects the changes in blood flow, is closely related to the activities of the heart, and is widely used in the monitoring of physiological parameters such as heart rate and blood oxygen saturation. Through this series of precise image processing and signal extraction steps, the PPG signal provides reliable and stable data support for physiological monitoring. This technology has high precision and robustness, can meet various medical monitoring needs, and provides important technical guarantees in health management and disease early warning.

[0051] It can be seen that this embodiment adopts a camera-based SCG signal and PPG signal acquisition method. The SCG signal is based on the principle of optical interference and can accurately capture the minute vibrations caused by the mechanical activities of the heart, while the PPG signal, relying on the principle of optical reflection, keenly monitors the dynamic changes in blood flow. This acquisition method breaks through the limitations of traditional ECG triggering technology in pathological conditions such as arrhythmia, as well as the interference and patient discomfort that may be caused by contact monitoring in the MRI environment. By fusing the SCG signal and the PPG signal, the system can provide more comprehensive and accurate physiological information, provide solid data support for high-quality MRI imaging, and promote the development of cardiac imaging in clinical diagnosis and disease screening, which has important clinical significance and application prospects.

[0052] It should be noted that although this embodiment mainly focuses on the fusion of SCG signals and PPG signals, it is possible to consider further fusing other physiological signals to form a more comprehensive multimodal signal monitoring system, so as to provide richer physiological information and enhance the accuracy and robustness of MRI trigger control. With the continuous progress of sensor technology, cameras or other optical sensors with higher precision and higher frame rate can also be used. For example, physiological signal acquisition devices based on electromagnetic induction, microelectromechanical systems or other non-optical principles can be developed to replace the current SCG signal and PPG signal acquisition devices, so as to further improve the quality and real-time performance of signal acquisition, thereby improving the performance of the entire system.

[0053] In addition, although the optical interference principle is the main method for current SCG signal acquisition, other high-precision vibration detection technologies can also be adopted, such as contact vibration detection based on piezoelectric sensors (without affecting the patient's comfort), or using ultrasonic technology to detect minute chest movements as a supplement or alternative to SCG signal acquisition. In addition to the optical reflection principle, a PPG signal acquisition method based on the optical transmission principle can also be explored, especially for detecting transmitted light at locations such as fingers or earlobes to obtain more stable pulse wave signals. In addition, combined with wearable optoelectronic devices such as smart bracelets or rings, continuous monitoring and transmission of PPG signals can be achieved to provide a more flexible signal source for MRI triggering.

[0054] Step S200: Preprocess the SCG signal and the PPG signal, determine the periodic changes of the SCG signal and the PPG signal, and calibrate the trigger points of each cardiac cycle.

[0055] Preprocessing is an important step in data analysis. Especially when dealing with cardiac motion signals, the quality of preprocessing directly affects the accuracy and reliability of subsequent analysis. The main purpose of preprocessing is to perform operations such as filtering, detrending, and normalization on the SCG signal and the PPG signal to remove unnecessary noise and trends, and enhance the characteristics of the effective signals, so as to provide clean and standardized data for subsequent analysis.

[0056] During the signal acquisition process, it is usually interfered by high-frequency and low-frequency noises. High-frequency noise may come from electrical interference of the device or rapid changes in the environment, while low-frequency noise may be caused by device vibration or baseline drift, etc. To improve the signal quality, this embodiment performs filtering on the acquired SCG signal and PPG signal to remove high-frequency noise and / or low-frequency noise in the SCG signal and the PPG signal. Band-pass filtering is a commonly used preprocessing method. The band-pass filter sets a frequency range , and only retains the signal components within this frequency range, removing the interference of other frequencies. Specifically, the input signal will be transformed to the frequency domain through Fourier transform and then multiplied by the frequency response of the band - pass filter to filter out the components other than the target frequency. The signal after inverse Fourier transform is the signal that has been filtered, that is, the redundant frequency components have been removed:

[0057] (1)

[0058] This processing can effectively suppress the noise components in the signal and retain the useful physiological signals.

[0059] In many physiological signals, especially those generated by physiological phenomena (such as heart movement), there may be long - term linear and / or non - linear trends. These trends do not represent the effective information of the target signal but may be caused by device biases, environmental factors, or other external interferences. To make the signal more stable and unaffected by these trends, in this embodiment, detrending processing is performed on the SCG signal and the PPG signal to remove the linear or non - linear trends in the SCG signal and the PPG signal to obtain a purer signal. Specifically, in this embodiment, a linear or non - linear function can be fitted to represent the trend component and subtracted from the original signal. Assume that the signal obtained after the above - mentioned filtering processing contains a trend component (linear trend), then the detrended signal is:

[0060] (2)

[0061] represents the slope of the linear trend, that is, the rate at which the trend changes with time. If is positive, it means the signal increases with time; if is negative, it means the signal decreases with time; if is zero, it means the signal has no linear trend. represents the intercept of the linear trend, that is, when t = 0, the value of the trend component. It represents the value of the trend component at the starting point of time. If the signal has a non - linear trend, a higher - order polynomial fitting or a more complex model can be used to ensure the stationarity and accuracy of the signal. The purpose of detrending is to remove those irrelevant components and only retain the part related to the target signal.

[0062] Normalization is to map the amplitudes of SCG signals and PPG signals into a unified standard range, usually [0, 1], to facilitate subsequent signal analysis. Especially in tasks such as multimodal signal fusion and feature extraction, the amplitudes of different signals may vary greatly, and normalization becomes particularly important at this time. Normalization helps to eliminate the amplitude differences between signals from different sources, enabling signals to be compared and processed under the same standard. Assume the signal is the signal after filtering and detrending, and its minimum value is , and the maximum value is , then the normalized signal can be expressed as:

[0063] (3)

[0064] After preprocessing the SCG signal and PPG signal in this embodiment, the periodic changes of the SCG signal and PPG signal are further analyzed to calibrate the trigger points of each cardiac cycle, so as to analyze the optimal trigger timing of each cardiac cycle and achieve a more intelligent and adaptive MRI trigger control.

[0065] In one implementation, this embodiment performs real-time peak detection on the SCG signal and PPG signal to capture the periodic changes of the SCG signal and PPG signal. Real-time peak detection is a key technology used to analyze the periodic fluctuations in continuous signals, especially important when dealing with dynamic signals. In this process, the sliding time window processing and peak detection cooperate closely to ensure that the system can accurately capture and provide real-time feedback on the periodic changes of the signal. Especially in the MR real-time trigger system based on SCG signals and PPG signals, this method is crucial for improving the diagnostic accuracy and image quality.

[0066] Specifically, the sliding time window is the first step in this process, and its main purpose is to simulate the display of real-time signals. By dividing the long continuous signal into multiple small segments, the sliding time window can gradually process each fixed-length data subset in the SCG signal and PPG signal. In each sliding process, the window processes the signal segment between the current time point t and its left boundary (that is the signal within the window), and preprocess, smooth, and detect the peaks of this segment of the signal. In this way, the sliding window simulates the continuous display and real-time processing of the signal. In dynamic signal analysis, the length of the sliding time window is crucial for the real-time performance and accuracy of detection. If the window is too short, it may over-respond to the instantaneous fluctuations in the SCG signal and PPG signal, while if the window is too long, it may lead to detection delays. Selecting an appropriate sliding window length can ensure that the periodic fluctuations in the SCG signal and PPG signal can be captured and fed back in a timely and accurate manner.

[0067] Preferably, when performing real-time peak detection in this embodiment, signal smoothing processing can also be carried out. Its main purpose is to further remove high-frequency noise in the signal, improve the denoising accuracy, and at the same time retain the main trend and periodic fluctuations of the signal. In practical applications, the smoothing processing method in this embodiment is local polynomial fitting, especially the Savitzky-Golay filter. This filter effectively eliminates high-frequency noise and sudden fluctuations by performing local polynomial fitting on the signal, while retaining the periodic characteristics of the signal. Specifically, the Savitzky-Golay filter realizes the smoothing process by performing polynomial fitting within a local window of the signal. Suppose we have an original signal , and the filter generates a smoothed signal by performing polynomial fitting within each sliding window of the signal, so as to replace the local fluctuations in the original signal. The smoothing filter can be expressed by polynomial fitting as:

[0068] (4)

[0069] where is the value of the smoothed signal, is the value at the i + k position in the original signal, are the polynomial coefficients obtained by least squares fitting, M is the window size of the filter, that is, the neighborhood length used for fitting in signal processing. Through this local polynomial fitting, the Savitzky-Golay filter can effectively remove short-term sudden fluctuations and noise while maintaining the main trend of the signal. Especially when dealing with high-frequency noise and rapidly changing signals, the polynomial obtained by fitting can better maintain the smooth characteristics of the signal.

[0070] The core purpose of peak detection is to accurately extract effective peaks from SCG signals and PPG signals, so as to provide reliable data for further analysis and real-time feedback. Peaks are usually local maxima that reflect significant changes in the signal, and peak detection ensures that the identified peaks are meaningful by performing certain screening and judgment on the signal. This process generally includes three key steps: finding local maxima, setting height thresholds, and minimum peak spacings to ensure the accuracy of peak identification.

[0071] Specifically, first determine the local maxima in the SCG signal and PPG signal to obtain the potential peak positions of the SCG signal and PPG signal. In the SCG signal and PPG signal, the point n of the local maximum satisfies the following conditions:

[0072] and (5)

[0073] That is, the local maximum is greater than its adjacent points and values. In this way, the core characteristics of the peak can be identified through the local maximum, providing preliminary peak candidates for subsequent analysis.

[0074] Next, set the height threshold and use the height threshold to screen the local maxima to obtain the screened local maxima. Relying solely on local maximum detection may cause some noise or small-amplitude fluctuations to be misidentified as peaks. Therefore, setting a height threshold is crucial. This height threshold is used to filter out the peaks with small amplitudes in the SCG signal and PPG signal, only retaining the peaks with practical significance. When the local maxima in the SCG signal and PPG signal meet the following conditions, they will be regarded as valid peaks:

[0075] (6)

[0076] For example, if the height threshold is set to , only when the local maximum is greater than 0.5, the point corresponding to the local maximum will be identified as a peak. This process can effectively exclude noise and small fluctuations, improving the reliability of peak detection.

[0077] Finally, determine the minimum peak spacing, analyze the screened local maxima based on the minimum peak spacing, obtain the real-time peaks in the SCG signal and PPG signal, and determine the cardiac cycle. To prevent peaks from being too close and being misjudged as the same peak, it is necessary to introduce the minimum peak spacing This parameter specifies the minimum distance between adjacent wave peaks, thus avoiding misjudging multiple wave peaks as the same wave peak. Specifically, if the time positions of two wave peaks are and , respectively, then the spacing between them must satisfy the following condition:

[0078] (7)

[0079] To improve the accuracy of wave peak detection, in this embodiment, the time interval between adjacent two wave peaks can be calculated based on historical signals (such as the signals within the previous sliding time window), and the minimum spacing between adjacent wave peaks can be dynamically adjusted . By calculating the interval between two wave peaks in the historical signal, a reasonable reference can be provided for the current wave peak detection, and then the minimum peak spacing can be adaptively adjusted according to the individual's heart rate change. Based on this, in this embodiment, based on the minimum peak spacing the selected local maxima are analyzed, which can not only avoid misjudging wave peaks as the same wave peak, but also improve the detection sensitivity and ensure the accurate identification of effective wave peaks. This method can not only effectively exclude noise and invalid fluctuations, but also adapt to changes under different physiological states, providing reliable data support for real-time signal processing and feedback.

[0080] As the time window slides, the new signal segments of the SCG signal and the PPG signal will be subjected to smoothing processing and then wave peak detection, ensuring that the signals in each time period can be analyzed in a timely and accurate manner, obtaining the real-time wave peaks in the SCG-PPG signal, that is, capturing the periodic changes of the SCG signal and the PPG signal, so the cardiac cycle can be determined. Generally speaking, the close cooperation of the sliding time window, signal smoothing processing and wave peak detection can not only display and analyze the periodic fluctuations in the signal in real time, but also ensure the rapid response and accurate analysis of the system. In the MRI real-time triggering system based on the SCG-PPG signal, this method effectively improves the image quality and the real-time performance of diagnosis, making the signal analysis more efficient and reliable.

[0081] It should be noted that in the wave peak detection link, in addition to the existing sliding time window, signal smoothing and local maximum judgment methods, in other implementation manners, in this embodiment, wavelet transform technology can also be introduced to perform multi-scale analysis on the signal to more accurately identify the wave peak characteristics in the signal. At the same time, combined with an adaptive filter, the filtering parameters are dynamically adjusted according to the real-time changes of the signal to further improve the accuracy and anti-noise ability of wave peak detection.

[0082] In cardiac imaging, motion artifacts caused by heart movement can significantly affect the quality of MRI images. Therefore, precise trigger control technology is crucial. In this embodiment, after capturing the periodic changes of the SCG signal and the PPG signal, key physiological events of the cardiac cycle (such as systolic and diastolic phases and blood flow fluctuation characteristics) are extracted from the SCG signal and the PPG signal, and then the trigger point is calibrated using threshold setting. For example, it is determined whether the signal value of the key physiological event exceeds the threshold. If it exceeds the threshold, the trigger point can be calibrated, providing a precise start timing for MRI scanning and reducing the interference of motion artifacts.

[0083] In other implementation manners, in addition to the trigger point calibration method based on peak detection in this embodiment, this embodiment can also explore a trigger point prediction model based on deep learning, training a neural network using a large amount of physiological signal data to directly predict the optimal trigger timing for each cardiac cycle, so as to achieve a more intelligent and adaptive MRI trigger control. In addition, in another implementation manner, this embodiment can also design a distributed signal processing architecture, dispersing the signal processing tasks of this embodiment to multiple small processors or sensor nodes. Each node is responsible for processing local signals, and then aggregating the results through wireless communication or other means to achieve distributed detection and decision-making of the trigger point. In addition, this embodiment can also adopt a prediction algorithm based on the patient's physiological model and historical data to predict the phase change of the cardiac cycle in advance, thereby determining the MRI trigger timing. This method does not directly rely on the peak detection of real-time signals, but realizes trigger control in a model-driven manner.

[0084] Step S300: Establish a mathematical model for real-time calculation of the optimal MRI trigger timing based on the trigger points of each cardiac cycle, and adjust the optimal MRI trigger timing in real time.

[0085] To further optimize the trigger timing, the system introduces a dynamic modeling method (such as Kalman filtering) to construct a mathematical model for real-time calculation of the optimal MRI trigger timing. This mathematical model takes the time series of the real-time monitored SCG signal and PPG signal as inputs, combines physiological state information (such as heart rate, body position, and breathing pattern, etc.), and dynamically adjusts the optimal MRI trigger timing. Specifically, Kalman filtering can achieve time synchronization with MRI scanning by updating and predicting the cardiac activity state in real time, thereby effectively improving the image quality and diagnostic accuracy. As a dynamic optimization algorithm, Kalman filtering achieves precise time synchronization through the following steps:

[0086] ① Prior Estimate: Before receiving the measurement data of the current time step, predict the current cardiac state based on the previous cardiac activity state and control input. Specifically, the prior estimate is through the state transition model and the posterior estimate of the previous time step , it is calculated that:

[0087] (8)

[0088] wherein, is the state transition matrix, which describes how the state transfers from one time step to the next; is the posterior estimate of the previous time step; is the control input matrix; is the control input.

[0089] ② Covariance Prediction: Meanwhile, predict the covariance matrix of the prior estimate , which reflects the uncertainty of the prior estimate:

[0090] (9)

[0091] wherein, is the covariance matrix of the posterior estimate of the previous time step; is the process noise covariance matrix, which represents the uncertainty of the system model.

[0092] ③ Measurement: At the current time step, obtain the measurement value from the physiological signal. These measurement values reflect the key physiological events of the current cardiac cycle.

[0093] ④ Kalman Gain: Calculate the Kalman gain , which is used to weigh the reliability of the prior estimate and the measurement data:

[0094] (10)

[0095] wherein, is the measurement matrix, which describes how the state is mapped to the measurement data; is the measurement noise covariance matrix, which represents the uncertainty of the measurement data.

[0096] ⑤ Current State Estimate: Combine the prior estimate and the measurement value through the Kalman gain to calculate the estimate of the current state .

[0097] (11)

[0098] ⑥ Posterior Covariance Update: Update the covariance matrix of the posterior estimate , reflecting the uncertainty of the current state estimate:

[0099] (12)

[0100] wherein, is the identity matrix.

[0101] In cardiac imaging, Kalman filtering combined with MRI triggering control technology plays a key role in reducing motion artifacts by accurately calculating the triggering timing and ensuring the high quality and stability of MRI images. Facing the dynamic changes of the cardiac cycle and the interference of physiological factors such as respiration and body position changes, this embodiment utilizes the adaptive adjustment mechanism of Kalman filtering to optimize the triggering timing in real time according to the patient's physiological state, correct the trigger point, and maintain the high quality of image acquisition. Kalman filtering can recursively estimate the dynamic changes of physiological signals, compensate for system noise and uncertainty, accurately predict and adjust the triggering timing, and reduce image distortion. Therefore, this embodiment adopts dynamic modeling methods such as Kalman filtering to track and adjust the optimal MRI triggering timing in real time, considering the interaction of multiple factors such as the respiratory cycle and body position changes, and accurately respond to the changes to ensure the accuracy and reliability of MRI scanning. In this embodiment, the MRI triggering control and trigger point calibration steps cooperate seamlessly, continuously monitor and optimize the real-time physiological state data, ensure image acquisition at a specific phase of the cardiac cycle, and minimize the impact of motion artifacts. This technology not only improves the adaptability of the triggering control system and promotes the application progress of cardiac imaging research, but also provides reliable cardiac image data for clinical practice to assist in the formulation of personalized treatment plans.

[0102] In summary, this embodiment preferably adopts a camera-based SCG-PPG signal acquisition method. The SCG signal is based on the principle of optical interference and can accurately capture the minute vibrations caused by cardiac mechanical activities, while the PPG signal, relying on the principle of optical reflection, keenly monitors the dynamic changes in blood flow. This acquisition method breaks through the limitations of traditional ECG triggering techniques in pathological conditions such as arrhythmia, as well as the interference and patient discomfort that contact monitoring may cause in the MRI environment. Of course, in other implementation manners, this embodiment can also adopt a contact-based SCG-PPG signal acquisition method, and this embodiment does not make any limitations thereto. In addition, by fusing the SCG signal and the PPG signal, the system can provide more comprehensive and accurate physiological information, providing solid data support for high-quality MRI imaging, promoting the development of cardiac imaging in clinical diagnosis and disease screening, and having important clinical significance and application prospects. In addition, the real-time peak detection technology in this embodiment can accurately capture the periodic changes of the signal and accurately calibrate the trigger points of the cardiac cycle. Based on these trigger points, this embodiment establishes a mathematical model, incorporates individual differences, and uses an optimization algorithm to adjust the MRI trigger timing in real time, flexibly adapting to the dynamic changes of the cardiac cycle and the interference of various physiological factors such as respiration and body position changes. This mechanism effectively reduces motion artifacts, improves the stability of image quality, provides more accurate and reliable cardiac images for clinical diagnosis, helps to scientifically formulate personalized treatment plans, optimizes the intelligence and accuracy levels of medical decisions, and significantly improves the clinical value and application effect of MRI imaging.

[0103] Based on the above embodiments, the present invention further provides an MRI real-time triggering system based on SCG-PPG signals, and the system is used to implement the steps of the above-mentioned MRI real-time triggering method based on SCG-PPG signals, as Figure 4 shown. The system includes: an SCG-PPG signal acquisition module 10, a trigger point calibration module 20, and an MRI trigger timing optimization module 30. Specifically, the SCG-PPG signal acquisition module 10 is used to capture the periodic motion of the heart to obtain the SCG signal, and to detect the dynamic changes in blood flow to obtain the PPG signal. The trigger point calibration module 20 is used to preprocess the SCG signal and the PPG signal, determine the periodic changes of the SCG signal and the PPG signal, and calibrate the trigger points of each cardiac cycle. The MRI trigger timing optimization module 30 is used to establish a mathematical model for real-time calculating the optimal MRI trigger timing based on the trigger points of each cardiac cycle, and to adjust the optimal MRI trigger timing in real time.

[0104] The working principles of the various modules in the MRI real-time triggering system based on SCG signals and PPG signals in this embodiment are the same as those of the respective steps in the above method embodiment, and will not be elaborated here.

[0105] Each module in the above MRI real-time triggering system based on SCG signals and PPG signals can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in or independent of the processor in the terminal in the form of hardware, or stored in the memory in the terminal in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0106] Based on the above embodiments, the present invention also provides a terminal, and the principle block diagram of the terminal can be as Figure 5 shown. The terminal may include one or more processors 100 ( Figure 5 only one is shown in the figure), a memory 101, and a computer program 102 stored in the memory 101 and executable on one or more processors 100. For example, an MRI real-time triggering program based on SCG signals and PPG signals. When one or more processors 100 execute the computer program 102, each step in the embodiment of the MRI real-time triggering method based on SCG signals and PPG signals can be implemented. Alternatively, when one or more processors 100 execute the computer program 102, the functions of each module / unit in the embodiment of the MRI real-time triggering system based on SCG signals and PPG signals can be implemented, which is not limited herein.

[0107] In one embodiment, the so-called processor 100 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0108] In one embodiment, the memory 101 may be an internal storage unit of the electronic device, such as the hard disk or memory of the electronic device. The memory 101 may also be an external storage device of the electronic device, such as a plug-in hard disk equipped on the electronic device, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash card, etc. Further, the memory 101 may also include both the internal storage unit and the external storage device of the electronic device. The memory 101 is used to store computer programs and other programs and data required by the terminal. The memory 101 may also be used to temporarily store the data that has been output or will be output.

[0109] Those skilled in the art can understand that Figure 5 the principle block diagram shown in is only the block diagram of some structures related to the solution of the present invention, and does not constitute a limitation on the terminal to which the solution of the present invention is applied. The specific terminal may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0110] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to the memory, storage, operation database or other media used in the embodiments provided by the present invention may include non-volatile and / or volatile memories. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time triggering method for MRI based on SCG-PPG signals, characterized in that, The method includes: capturing the periodic motion of the heart to obtain an SCG signal, and detecting the dynamic changes in blood flow to obtain a PPG signal; preprocessing the SCG signal and the PPG signal to determine the periodic changes in the SCG signal and the PPG signal, and calibrating the trigger points of each cardiac cycle; establishing a mathematical model for real-time calculation of the optimal MRI trigger timing based on the trigger points of each cardiac cycle, and adjusting the optimal MRI trigger timing in real time; The determining the periodic changes in the SCG signal and the PPG signal includes: performing real-time peak detection processing on the SCG signal and the PPG signal to capture the periodic changes in the SCG signal and the PPG signal; The real-time peak detection processing includes: sliding time window processing, signal smoothing processing, and peak detection. Among them, the sliding time window processing includes: respectively dividing the long-time continuous SCG signal and PPG signal into multiple segments, and gradually processing each fixed-length data subset in the SCG signal and the PPG signal; The signal smoothing processing includes: using a Savitzky-Golay filter to perform smoothing processing by polynomial fitting within a local window of the signal; The peak detection includes: respectively determining the local maxima in the SCG signal and the PPG signal to obtain the potential peak positions of the SCG signal and the PPG signal; obtaining a preset height threshold, and using the height threshold to screen the local maxima to obtain the screened local maxima; determining the minimum peak spacing, and analyzing the screened local maxima based on the minimum peak spacing to obtain the real-time peaks in the SCG signal and the PPG signal and determine the cardiac cycle; Establishing a mathematical model for real-time calculation of the optimal MRI trigger timing based on the trigger points of each cardiac cycle, and adjusting the optimal MRI trigger timing in real time, includes: When establishing the mathematical model, introducing a Kalman filter to construct a mathematical model for real-time calculation of the optimal MRI trigger timing. The mathematical model takes the time series of the real-time monitored SCG signal and PPG signal as inputs, and combines physiological state information to dynamically adjust the optimal MRI trigger timing.

2. The MRI real-time triggering method based on SCG-PPG signals according to claim 1, wherein The capturing the periodic motion of the heart to obtain an SCG signal includes: positioning a defocused camera at the fourth rib site in the lower left of the chest cavity and irradiating it in cooperation with a laser system; adjusting the configuration of the defocused camera to collect a speckle image; using an optical flow method to deeply analyze the collected speckle image, calculating the motion amplitudes of the speckles in the X-axis and Y-axis directions to obtain a motion amplitude sequence; dividing the motion amplitude sequence to determine the motion angle of each segment, where the motion angle reflects the direction of heart vibration; obtaining the SCG signal based on the motion angle of each segment.

3. The MRI real-time triggering method based on SCG-PPG signals according to claim 2, wherein The adjusting the configuration of the defocused camera includes: adjusting the defocused camera so that the distance between the focal plane and the chest cavity plane is greater than the distance between the lens and the focal plane.

4. The MRI real-time triggering method based on SCG-PPG signals according to claim 1, characterized in that, Detecting the dynamic changes in blood flow to obtain a PPG signal includes: Cover the key parts of the face with an optical sensor, emit a synchronous signal of a periodic square wave, and collect images; Extract the part containing skin pixels from the collected images to obtain the skin area; Based on the skin area, determine the G-channel image sequence, and extract the PPG signal from the G-channel image sequence.

5. The MRI real-time triggering method based on SCG-PPG signals according to claim 1, characterized in that The preprocessing includes: filtering, detrending, and normalization; wherein, the filtering is used to remove high-frequency noise and / or low-frequency noise in the SCG signal and the PPG signal; the detrending is used to remove the linear trend and / or non-linear trend in the SCG signal and the PPG signal; the normalization is used to map the amplitudes of the SCG signal and the PPG signal to a unified standard range.

6. An MRI real-time triggering system based on SCG-PPG signals, characterized in that, The system is used to implement the steps of the MRI real-time triggering method based on the SCG-PPG signal according to any one of claims 1-5 above. The system includes: An SCG-PPG signal acquisition module, configured to capture the periodic motion of the heart to obtain the SCG signal, and detect the dynamic changes of blood flow to obtain the PPG signal; A trigger point calibration module, configured to preprocess the SCG signal and the PPG signal, determine the periodic changes of the SCG signal and the PPG signal, and calibrate the trigger points of each cardiac cycle; An MRI trigger timing optimization module, configured to establish a mathematical model for real-time calculating the optimal MRI trigger timing based on the trigger points of each cardiac cycle, and adjust the optimal MRI trigger timing in real time.

7. A terminal, characterized in that, The terminal includes a memory, a processor, and an MRI real-time trigger program based on the SCG-PPG signal stored in the memory and executable on the processor. When the processor executes the MRI real-time trigger program based on the SCG-PPG signal, the steps of the MRI real-time triggering method based on the SCG-PPG signal according to any one of claims 1-5 above are implemented.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an MRI real-time trigger program based on the SCG-PPG signal. When the MRI real-time trigger program based on the SCG-PPG signal is executed by a processor, the steps of the MRI real-time triggering method based on the SCG-PPG signal according to any one of claims 1-5 above are implemented.

Images