Physiological Signal Detection Method, System and Storage Medium for Magnetic Resonance Scanning

Abstract

This application relates to the technical field of physiological signal detection. This application discloses a physiological signal detection method, system, and storage medium for magnetic resonance scanning, which can achieve precise detection of multi-dimensional physiological parameters in a strong magnetic field environment. The physiological signal detection method for magnetic resonance scanning includes, during the process of a magnetic resonance device in a physiological signal detection system for magnetic resonance scanning scanning a patient, obtaining multi-spectral pulse wave signals of the patient's face by using a time-sequential spectroscopic method; while obtaining the multi-spectral pulse wave signals, obtaining cardiac vibration signals of the patient's chest by using a vibration measurement method; preprocessing the multi-spectral pulse wave signals to obtain a first target signal; preprocessing the cardiac vibration signals to obtain a second target signal; extracting physiological signals with multiple physiological parameters based on the first target signal and the second target signal; and feeding back the physiological signals to medical staff.

Description

Technical Field

[0001] This application relates to the technical field of physiological signal detection. More specifically, this application relates to a physiological signal detection method, system, and storage medium for magnetic resonance scanning. Background Art

[0002] Cardiovascular Magnetic Resonance (CMR), as a high-resolution medical imaging technology, plays a crucial role in evaluating the structure and function of the heart and blood vessels. However, the periodic movements caused by physiological activities such as the heart and respiration will generate motion artifacts in a high-field magnetic resonance (such as 3T and above) environment, thereby affecting the accurate identification of lesions. Although existing gating technologies, such as Electrocardiogram (ECG) and Photoplethysmogram (PPG), can alleviate the artifact problem to a certain extent, ECG and PPG can only detect single physiological signals and cannot achieve precise detection of multi-dimensional physiological parameters, failing to meet the current demand for detecting the multi-dimensional physiological parameters of patients in a magnetic resonance environment. Summary of the Invention

[0003] The objective of the embodiments of this application is to provide a physiological signal detection method, system, and storage medium for magnetic resonance scanning, which can achieve precise detection of multi-dimensional physiological parameters in a strong magnetic field environment. The embodiments of this application are mainly implemented through the following technical solutions:

[0004] In the first aspect of the embodiments of this application, a physiological signal detection method for magnetic resonance scanning is provided, including:

[0005] During the process of a magnetic resonance device in a physiological signal detection system for magnetic resonance scanning scanning a patient, a multi-spectral pulse wave signal of the patient's face is obtained by using a time-sequential spectroscopy method;

[0006] While obtaining the multi-spectral pulse wave signal, a cardiac vibration signal of the patient's chest is obtained by using a vibration measurement method;

[0007] The multi-spectral pulse wave signal is preprocessed to obtain a first target signal;

[0008] The cardiac vibration signal is preprocessed to obtain a second target signal;

[0009] Based on the first target signal and the second target signal, a physiological signal with multiple physiological parameters is extracted;

[0010] The physiological signal is fed back to medical staff.

[0011] According to an embodiment of the present application, the steps of obtaining the multi-spectral pulse wave signal of the patient's face by using the time-sequential spectroscopy method include:

[0012] Receiving a first image sequence transmitted by a first camera of the physiological signal detection system for magnetic resonance scanning, the first image sequence including an image sequence of multiple frame periods, and each image sequence of a frame period being an image sequence obtained by the first camera successively photographing the face according to the spectral order with multiple different wavelengths of spectra;

[0013] In the first image sequence, combining the first images corresponding to the spectra of the same wavelength into a single spectral video to obtain a plurality of spectral videos;

[0014] Performing time-series analysis on all pixels in the target spectral video, and using a skin reflection model to model the change of the reflection value of each pixel in the RGB channels of the target spectral video over time, to obtain a single-spectral pulse wave signal corresponding to the target spectral video, where the target spectral video is any one of the plurality of spectral videos;

[0015] Constructing the multi-spectral pulse wave signal from all the single-spectral pulse wave signals.

[0016] According to an embodiment of the present application, the steps of obtaining the cardiac vibration signal of the patient's chest by using the vibration measurement method include:

[0017] Receiving a second image sequence transmitted by a second camera of the physiological signal detection system for magnetic resonance scanning;

[0018] Calculating a first motion amplitude in the X direction and a second motion amplitude in the Y direction of each second image in the second image sequence by using the optical flow method;

[0019] Combining the first motion amplitudes in the X direction of all the second images into a first motion amplitude sequence;

[0020] Combining the second motion amplitudes in the Y direction of all the second images into a second motion amplitude sequence;

[0021] Extracting the motion angle from the first motion amplitude sequence and the second motion amplitude sequence by using a sliding window method;

[0022] Obtaining the cardiac vibration signal based on the motion angle.

[0023] According to an embodiment of the present application, the steps of preprocessing the multi-spectral pulse wave signal to obtain a first target signal include;

[0024] Performing high-pass filtering on the multi-spectral pulse wave signal to obtain a first signal to be processed;

[0025] Perform motion compensation processing on the first signal to be processed to obtain a second signal to be processed;

[0026] Perform noise reduction processing on the second signal to be processed to obtain the first target signal.

[0027] According to an embodiment of the present application, the steps of preprocessing the cardiac vibration signal to obtain a second target signal include:

[0028] Perform high-pass filtering on the cardiac vibration signal to obtain a third signal to be processed;

[0029] Perform motion compensation processing on the third signal to be processed to obtain a fourth signal to be processed;

[0030] Perform noise reduction processing on the fourth signal to be processed to obtain the second target signal.

[0031] According to an embodiment of the present application, the steps of extracting a physiological signal with multiple physiological parameters based on the first target signal and the second target signal include:

[0032] Extract the blood oxygen saturation of the physiological signal from the first target signal;

[0033] Calculate the heart rate and heart rate variability of the physiological signal based on the second target signal;

[0034] Calculate the respiratory rate of the physiological signal from the second target signal using the optical flow method.

[0035] According to an embodiment of the present application, after the step of extracting a physiological signal with multiple physiological parameters based on the first target signal and the second target signal, the physiological signal detection method for magnetic resonance scanning further includes:

[0036] When at least one physiological parameter of the physiological signal reaches a first preset threshold, activate an alarm mechanism;

[0037] When the trend change state of at least one physiological parameter of the physiological signal reaches a dangerous state, activate a warning mechanism.

[0038] In a second aspect of the embodiments of the present application, there is provided a physiological signal detection system for magnetic resonance scanning, including:

[0039] A multi-spectral pulse wave signal acquisition module, configured to acquire a multi-spectral pulse wave signal of the face of the patient by using a time-sequential spectroscopic method during the process of the magnetic resonance device of the physiological signal detection system for magnetic resonance scanning scanning the patient;

[0040] A cardiac vibration signal acquisition module, configured to acquire a cardiac vibration signal of the patient's chest by using a vibration measurement method while acquiring the multi-spectral pulse wave signal;

[0041] A first target signal acquisition module, configured to preprocess the multi-spectral pulse wave signal to obtain a first target signal;

[0042] A second target signal acquisition module, configured to preprocess the cardiac vibration signal to obtain a second target signal;

[0043] A physiological signal extraction module, configured to extract a physiological signal with multiple physiological parameters based on the first target signal and the second target signal;

[0044] A feedback module, configured to feedback the physiological signal to medical staff.

[0045] In a third aspect of the embodiments of the present application, a physiological signal detection system for magnetic resonance scanning is provided, including: a processor, a memory, a multi-modal non-contact multi-parameter detection optical device, a laser, and a magnetic resonance device, where the multi-modal non-contact multi-parameter detection optical device includes a first camera and a second camera, the memory is used to store a computer program, the processor is used to call and run the computer program stored in the memory, and execute the steps of the physiological signal detection method for magnetic resonance scanning described in the first aspect of the embodiments of the present application, the magnetic resonance device, the first camera, and the second camera are all communicatively connected to the processor, and the laser is disposed on one side of the second camera;

[0046] The first camera is configured to perform an image acquisition operation on the skin of the patient's face to obtain a first image sequence, and transmit the first image sequence to the processor;

[0047] The second camera is configured to perform an image acquisition operation on the patient's chest to obtain a second image sequence, and transmit the second image sequence to the processor;

[0048] When the second camera performs an image acquisition operation on the chest, the laser emits laser light to irradiate the chest.

[0049] In a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, where the computer-readable storage medium is used to store a computer program, and the computer program enables a computer to execute the steps of the physiological signal detection method for magnetic resonance scanning provided in the first aspect of the embodiments of the present application.

[0050] The beneficial effects of the embodiments of the present application include:

[0051] Embodiments of the present application use multi-spectral pulse wave signals and cardiac vibration signals to detect the physiological signals of patients. Specifically, in the process of a magnetic resonance device of a physiological signal detection system for magnetic resonance scanning scanning a patient, embodiments of the present application use a time-sequential spectroscopy method to obtain multi-spectral pulse wave signals of the patient's face; while obtaining the multi-spectral pulse wave signals, a vibration measurement method is used to obtain cardiac vibration signals of the patient's chest; preprocess the multi-spectral pulse wave signals to obtain a first target signal; preprocess the cardiac vibration signals to obtain a second target signal; extract physiological signals with various physiological parameters based on the first target signal and the second target signal; and feedback the physiological signals to medical staff. Thus, embodiments of the present application can achieve accurate detection of multi-dimensional physiological parameters in a strong magnetic field environment, enabling medical staff to monitor various physiological indicators in real time and ensuring the safety of patients' lives. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in 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 application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0053] Figure 1 It is a flowchart of the physiological signal detection method for magnetic resonance scanning according to the present application in some embodiments;

[0054] Figure 2 It is a schematic diagram of the principle of the physiological signal detection method for magnetic resonance scanning according to the present application in some embodiments;

[0055] Figure 3 It is a flowchart block diagram for constructing an early warning model in the present application;

[0056] Figure 4 It is a flowchart of the physiological signal detection method for magnetic resonance scanning according to the present application in other embodiments;

[0057] Figure 5 It is a schematic block diagram of the principle of the physiological signal detection system for magnetic resonance scanning according to the present application in some embodiments;

[0058] Figure 6 It is a schematic block diagram of the principle of the physiological signal detection system for magnetic resonance scanning according to the present application in other embodiments;

[0059] Figure 7 It is a schematic block diagram of the principle of the multi-modal non-contact multi-parameter detection optical device in the present application in some embodiments;

[0060] Figure 8 This is a schematic block diagram of the physiological signal detection system for magnetic resonance scanning in some other embodiments of the present application. Specific Embodiments

[0061] To make the above objects, features, and advantages of the present application more apparent and understandable, the following provides a detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings. Many specific details are set forth in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the spirit of the present application. Therefore, the present application is not limited by the specific embodiments disclosed below.

[0062] It should be noted that the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present application, "a plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0063] The term "exemplary" or "for example" is used to indicate an example, illustration, or explanation. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or designs. Rather, the use of the term "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0064] The term "comprising", "including", or any other variation thereof is intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to such process, method, product, or device.

[0065] Unless otherwise defined, all technical and scientific terms used in the specification of the present application have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The term "and / or" used in the specification of the present application includes any and all combinations of one or more of the related listed items.

[0066] The following further describes the specific embodiments of the present application in conjunction with the accompanying drawings.

[0067] Reference Figure 1As shown in the figure, it is a flowchart of a physiological signal detection method provided in the first aspect of the embodiment of the present application. In Figure 1 the physiological signal detection method for magnetic resonance scanning includes:

[0068] S1. During the process of the magnetic resonance device of the physiological signal detection system for magnetic resonance scanning scanning a patient, a multi-spectral pulse wave signal of the patient's face is obtained by using a time-sequential spectroscopic method.

[0069] The multi-spectral pulse wave signal can be expressed as an rPPG (Remote Photoplethysmography) signal. It should be understood that the rPPG technology detects the heart rate and heart rate variability by capturing the change in subcutaneous vascular volume caused by cardiac pulsation, affecting the reflected light of the skin, and forming a pulsating signal in the video.

[0070] Further, the step of obtaining the multi-spectral pulse wave signal of the patient's face by using the time-sequential spectroscopic method includes:

[0071] S11. Receive a first image sequence transmitted by a first camera of the physiological signal detection system for magnetic resonance scanning. The first image sequence includes image sequences of multiple frame periods. The image sequence of each frame period is an image sequence obtained by the first camera successively shooting the face with multiple different wavelengths of spectra in spectral order.

[0072] The first camera needs to be aligned with the skin area of the patient's face for shooting, while ensuring that the face skin is included in the picture and reducing the possibility of frame loss of the first camera.

[0073] The first camera is a multi-spectral camera.

[0074] The frame period refers to the time required for the first camera to successively shoot the face with multiple different wavelengths of spectra. The frame period can refer to Figure 2 as shown.

[0075] In this embodiment, the multiple different wavelengths of spectra are four different wavelengths of spectra, specifically blue light, green light, red light, and infrared light. Exemplarily, the blue light is 465 nm, the green light is 530 nm, the red light is 680 nm, and the infrared light is 850 nm. During the process of the first camera obtaining the first image sequence, blue light, green light, red light, and infrared light are alternately used for fill light, so that image data of different spectra (that is, the first image sequence) can be captured, providing support for the measurement of heart rate, heart rate variability, and blood oxygen saturation.

[0076] The spectral order can be set by those skilled in the art according to actual needs, and no further limitation is imposed herein.

[0077] S12. In the first image sequence, combine the first images corresponding to the spectra of the same wavelength into a single spectral video to obtain a plurality of spectral videos.

[0078] It should be understood that each image in the first image sequence is a first image.

[0079] In the embodiments of the present application, four spectral videos can be obtained, that is, one spectral video corresponds to blue light, one spectral video corresponds to green light, one spectral video corresponds to red light, and one spectral video corresponds to infrared light.

[0080] S13. Perform time series analysis on all pixels in the target spectral video, and use the skin reflection model to model the change of the reflection value of each pixel in the RGB channels (that is, the three channels of red, green, and blue) in the target spectral video over time, to obtain a single-spectral pulse wave signal corresponding to the target spectral video, where the target spectral video is any one of the plurality of spectral videos.

[0081] The skin reflection model is a mathematical model used to describe how light interacts with the skin. As a multi-layered structure, the reflection characteristics of the skin are determined by various factors, including the oil on the skin surface, melanin in the epidermis, and hemoglobin in the dermis. When light irradiates the skin, a part of the light is reflected by the oil on the skin surface to form specular reflection (or highlight), while another part of the light penetrates the skin and is reflected again after scattering to form diffuse reflection.

[0082] In the embodiments of the present application, the reflection signal of the skin reflection model for each pixel in the RGB channels is mainly composed of light intensity, specular reflection, diffuse reflection, and quantization noise. Among them, specular reflection is mainly composed of the reflected light on the skin surface. Although it does not contain the physiological information of heartbeats, the change of the light angle and intensity caused by the movement of the subject (that is, the patient) may cause noise interference. Diffuse reflection is closely related to the absorption and scattering of light by skin tissues and is the key source of pulse signals. Affected by the periodic change of hemoglobin concentration, a periodic reflection signal consistent with the heartbeat is formed. Although melanin and carotene contribute to the absorption and scattering of light, their reflection intensity does not change with heart activity.

[0083] The single-spectral pulse wave signal can refer to Figure 2The signals corresponding to label 1, label 2, label 3, and label 4, where the signal pointed to by label 1 is the single-spectrum pulse wave signal corresponding to blue light, the signal pointed to by label 2 is the single-spectrum pulse wave signal corresponding to green light, the signal pointed to by label 3 is the single-spectrum pulse wave signal corresponding to red light, and the signal pointed to by label 4 is the single-spectrum pulse wave signal corresponding to infrared light.

[0084] S14. Combine all the single-spectrum pulse wave signals to form the multi-spectrum pulse wave signal.

[0085] By accurately describing the influence of hemoglobin concentration change on the reflected light intensity and combining the pigment distribution and skin color depth of the skin tissue in the embodiments of the present application, the first camera can effectively extract the multi-spectrum rPPG signal (i.e., the multi-spectrum pulse wave signal) containing multi-dimensional vital sign information such as heart rate, heart rate variability, and blood oxygen.

[0086] S2. While acquiring the multi-spectrum pulse wave signal, acquire the heart vibration signal of the patient's chest by using the vibration measurement method.

[0087] The heart vibration signal can be expressed as an SCG (Seismocardiography) signal.

[0088] Based on the mechanical vibration of the heart, the SCG signal can effectively extract parameters such as heart rate, heart rate variability, and respiratory rate, thus supplementing and enriching the detection dimension of the overall physiological parameters.

[0089] Further, the step of acquiring the heart vibration signal of the patient's chest by using the vibration measurement method includes:

[0090] S21. Receive the second image sequence transmitted by the second camera of the physiological signal detection system facing the magnetic resonance scan.

[0091] The second camera is a defocus camera.

[0092] During the heartbeat of the patient, the minute movement on the surface of the thoracic cavity (i.e., the chest) causes changes in the three-dimensional interference field formed by the reflected laser, resulting in the periodic movement of the interference speckle pattern on the imaging plane of the second camera, which is synchronized with the contraction and relaxation processes of the heart. To achieve precise capture, the physiological signal detection system for magnetic resonance scanning adjusts the angle of the second camera so that the laser irradiates the area of the fourth rib in the lower left part of the thoracic cavity. In the area of the fourth rib, the laser interferes with the surface reflected light to form a periodically changing speckle pattern, and then the second camera records these speckle patterns to obtain the cardiac vibration signal. As the heartbeat progresses, the speckle pattern shows periodic changes. The physiological signal detection system for magnetic resonance scanning can accurately reconstruct the cardiac motion process and extract the SCG signal by analyzing the motion patterns in the images.

[0093] To further enhance the visibility of the minute movement of the thoracic cavity, the physiological signal detection system for magnetic resonance scanning adjusts the lens focal length of the second camera to move the focal plane of the second camera from the cardiac plane to a specific position, thereby significantly magnifying the interference speckle image of the laser reflection. When the focal plane is adjusted to the appropriate position, the speckle image changes periodically with the heartbeat, and the defocus degree of the lens of the second camera is proportional to the motion amplification gain of the heart shock signal (i.e., the cardiac vibration signal).

[0094] The physiological signal detection system for magnetic resonance scanning uses a second camera with a high frame rate of at least 200 fps and a resolution of 400×300 to capture the laser speckle image (i.e., the second image sequence).

[0095] S22. Calculate the first motion amplitude in the X direction and the second motion amplitude in the Y direction of each second image in the second image sequence using the optical flow method.

[0096] S23. Combine the first motion amplitudes of all second images in the X direction into a first motion amplitude sequence.

[0097] S24. Combine the second motion amplitudes of all second images in the Y direction into a second motion amplitude sequence.

[0098] S25. Extract the motion angle from the first motion amplitude sequence and the second motion amplitude sequence using a sliding window method.

[0099] Further, the step S25 includes:

[0100] S251. Set a preset window and a preset step.

[0101] The size of the preset window can be set by those skilled in the art according to actual requirements. The specific value of the preset stride can be set by those skilled in the art according to actual requirements.

[0102] S252. Extract a first subsequence set from the first motion amplitude sequence according to the preset window and the preset stride.

[0103] S253. Extract a second subsequence set from the second motion amplitude sequence according to the preset window and the preset stride.

[0104] S254. Calculate the angle to be processed based on the first subsequence set and the second subsequence set.

[0105] Further, the calculation formula in step S254 is:

[0106] ;

[0107] Wherein, represents the main motion quadrant, which needs to be obtained from the four quadrants using the calculation formula in step S254. The purpose of setting is that there may be noise in the actual motion of the speckle, but the motion direction of the speckle caused by the heart is highly concentrated and located in one of the four quadrants; represents selecting the quadrant with the largest ninety-percentile value from the four quadrants and using it as the main motion quadrant to which it belongs. The ninety-percentile value refers to a statistical index used for analyzing the motion data in the four quadrants; represents the ninety-percentile amplitude value in the corresponding quadrant; represents the th element in the first subsequence set; represents the th element in the second subsequence set; represents the current axis coordinate position; represents the current coordinate position; represents the four quadrants in the Cartesian coordinate system, ; represents the angle to be processed; represents taking the median; represents the tangent function; represents the integral value of the th point in the direction, that is, the integral value of the th point in the th element in the second subsequence set; represents the th point in the The integral value in a certain direction, that is, the integral value of the th element in the first sub-sequence set and the th point; represents the th point at the axis coordinate position; represents the th point at the axis coordinate position.

[0108] When processing the cardiac vibration signal, the physiological signal detection system facing magnetic resonance scanning calculates the ninetieth percentile value of the motion data in each quadrant, that is, calculates that 90% of the data points of all the motion data in each quadrant are lower than or equal to a specific value in that quadrant. By comparing the ninetieth percentile values of the four quadrants, the quadrant with the largest value is selected as the "main motion quadrant". The "main motion quadrant" contains the most significant cardiac motion information because the distribution concentration of its motion amplitude is relatively high, which can effectively exclude noise and interference from other non-cardiac motions.

[0109] During the process of calculating the motion angle by the optical flow method, the physiological signal detection system facing magnetic resonance scanning divides the image plane into four quadrants. The motion data in each quadrant represents the direction and amplitude of the object motion within that quadrant. By analyzing the motion data in each quadrant, the system can determine the motion trend of the object on the image plane, thereby helping to extract the cardiac vibration signal.

[0110] S255. Calculate the average value of all the angles to be processed to obtain the motion angle.

[0111] S26. Obtain the cardiac vibration signal based on the motion angle.

[0112] Furthermore, the calculation formula in step S26 is:

[0113] ;

[0114] where represents the cardiac vibration signal; represents the signal sequence, that is, the first motion amplitude sequence; represents the motion angle; represents the signal sequence, that is, the second motion amplitude sequence.

[0115] Step S2 provides a non-contact, non-invasive and high-precision means for cardiac health detection, which can support cardiac magnetic resonance (CMR) image acquisition or accurately record the cardiac motion position.

[0116] S3. Preprocess the multi-spectral pulse wave signal to obtain a first target signal.

[0117] Further, the step S3 includes:

[0118] S31. Perform high-pass filtering on the multi-spectral pulse wave signal to obtain a first signal to be processed.

[0119] Since low-frequency signals usually originate from physiological activities such as breathing or body movement, and these physiological activities are all irrelevant to the pulse wave signal, these low-frequency noises will affect the accuracy and stability of the multi-spectral pulse wave signal. In practical applications, low-frequency noises often appear as fluctuations with a long period, usually below 0.1 Hz. These fluctuations mainly come from the respiratory rhythm, head movement, or slight jitter of the first camera. To effectively remove these irrelevant low-frequency interferences, the physiological signal detection system for magnetic resonance scanning uses a high-pass filter to perform high-pass filtering on the multi-spectral pulse wave signal, thereby obtaining the first signal to be processed. The high-pass filter can allow signals higher than a preset cut-off frequency to pass through, while filtering out low-frequency components lower than the preset cut-off frequency.

[0120] Through the above method, the physiological signal detection system for magnetic resonance scanning can effectively remove low-frequency noises caused by breathing or body movement, retain high-frequency components related to the pulse wave signal, thereby improving the accuracy of the multi-spectral pulse wave signal (i.e., rPPG signal), and providing a clearer basis for subsequent extraction of heart rate and other physiological parameters.

[0121] S32. Perform motion compensation processing on the first signal to be processed to obtain a second signal to be processed.

[0122] Further, the step S32 includes:

[0123] S321. Divide the first signal to be processed into multiple first signal segments.

[0124] S322. Detect first feature points in each first signal segment.

[0125] The first feature points are the peaks, valleys, or inflection points in the corresponding first signal segments.

[0126] S323. Use the optical flow method to combine all first feature points and all first signal segments, and calculate the first optical flow field between adjacent first signal segments in all first signal segments. The first optical flow field describes the movement direction and speed of each first feature point in adjacent first signal segments.

[0127] The optical flow method is the Lucas-Kanade method based on gradients or the Horn-Schunck method based on global energy optimization (this method is a classic dense optical flow algorithm).

[0128] S324. Calculate the first compensation amount for each first feature point based on all the first optical flow fields.

[0129] The first compensation amount includes motion parameters such as translation and rotation, and is used to correct the signal distortion caused by motion.

[0130] S325. Perform motion compensation processing on the first signal to be processed based on the compensation amount of each first feature point, and obtain the second signal to be processed.

[0131] In the embodiment of the present application, the optical flow method is used to eliminate the noise generated by motion in the first signal to be processed, so as to obtain the second signal to be processed.

[0132] The motion compensation processing in the S32 step aims to correct the interference of the patient's motion on the first signal to be processed, and the result is to obtain a corrected signal, that is, the second signal to be processed. The second signal to be processed can more accurately reflect the physiological state of the patient. After compensation, the waveform of the second signal to be processed is more stable and the fluctuation is more regular. This not only facilitates the extraction of subsequent physiological parameters, but also effectively reduces the noise and interference components in the signal, thus significantly improving the signal-to-noise ratio. More importantly, the signal processed by motion compensation can more truly reflect the physiological changes of the patient, provide a reliable diagnosis basis for medical staff, and thus ensure the life safety of the patient.

[0133] S33. Perform noise reduction processing on the second signal to be processed to obtain the first target signal.

[0134] Specifically, in the embodiment of the present application, independent component analysis (ICA) is used to perform noise reduction processing on the second signal to be processed.

[0135] Independent component analysis is a blind source separation technique that can extract independent components from multiple mixed signals. Independent component analysis is widely used in denoising complex signals. In the physiological signal detection system for magnetic resonance scanning, the signals are often disturbed by factors such as facial movement, environmental noise, light changes, and equipment errors, resulting in signal chaos and affecting the accurate analysis of physiological parameters. Independent component analysis separates independent components from the mixed signals by assuming that the signal sources are independent and non-Gaussian distributed. During the processing, the ICA optimization algorithm restores the mixed signals to multiple independent components by maximizing the independence of the signals, thereby removing noise components unrelated to blood flow, such as facial movement and environmental interference. This technique can significantly improve the purity and reliability of the signals, ensure that the physiological signals extracted from the physiological signal detection system for magnetic resonance scanning are more accurate, provide more accurate data support for the analysis of parameters such as heart rate and blood oxygen saturation, and thus improve the performance of the non-contact health detection system (i.e., the physiological signal detection system for magnetic resonance scanning).

[0136] Before performing high-pass filtering on the multi-spectral pulse wave signal, the multi-spectral pulse wave signal can be normalized first.

[0137] The step S3 can ensure the accuracy and stability of the first target signal.

[0138] S4. Preprocess the cardiac vibration signal to obtain a second target signal.

[0139] Further, the step S4 includes:

[0140] S41. Perform high-pass filtering on the cardiac vibration signal to obtain a third signal to be processed.

[0141] In the embodiment of the present application, a high-pass filter is used to perform high-pass filtering on the cardiac vibration signal.

[0142] S42. Perform motion compensation processing on the third signal to be processed to obtain a fourth signal to be processed.

[0143] Further, the step S42 includes:

[0144] S421. Divide the third signal to be processed into multiple second signal segments.

[0145] S422. Detect second feature points in each second signal segment.

[0146] The second feature points are the peaks, valleys, or inflection points in the corresponding second signal segments.

[0147] S423. Use the optical flow method in combination with all the second feature points and all the second signal segments to calculate the second optical flow field between adjacent second signal segments in all the second signal segments, where the second optical flow field describes the movement direction and speed of each second feature point in adjacent second signal segments.

[0148] The optical flow method is the gradient-based Lucas-Kanade method or the Horn-Schunck method based on global energy optimization (this method is a classic dense optical flow algorithm).

[0149] S424. Calculate the second compensation amount for each second feature point based on all the second optical flow fields.

[0150] The second compensation amount includes motion parameters such as translation and rotation, which are used to correct signal distortion caused by motion.

[0151] S425. Perform motion compensation processing on the third signal to be processed based on the compensation amount of each second feature point to obtain the fourth signal to be processed.

[0152] In the embodiment of the present application, the optical flow method is used to eliminate the noise generated by motion in the third signal to be processed, thereby obtaining the fourth signal to be processed. The motion compensation processing in step S42 aims to correct the interference of the patient's motion on the third signal to be processed, and the result is to obtain a corrected signal, that is, the fourth signal to be processed. The fourth signal to be processed can more accurately reflect the physiological state of the patient. After compensation, the waveform of the fourth signal to be processed is more stable and the fluctuation is more regular, which not only facilitates the extraction of subsequent physiological parameters, but also effectively reduces the noise and interference components in the fourth signal to be processed, thereby significantly improving the signal-to-noise ratio of the fourth signal to be processed.

[0153] S43. Perform noise reduction processing on the fourth signal to be processed to obtain the second target signal.

[0154] In the embodiment of the present application, independent component analysis is used to perform noise reduction processing on the fourth signal to be processed.

[0155] S5. Extract physiological signals with multiple physiological parameters based on the first target signal and the second target signal.

[0156] Further, the S5 step includes:

[0157] S51. Extract the blood oxygen saturation of the physiological signal from the first target signal. The S51 step can be understood as using a blood oxygen saturation detection model to detect the blood oxygen saturation.

[0158] Specifically, the step S51 (i.e., the implementation step of the blood oxygen saturation detection model) includes:

[0159] S511. Perform detrending processing on the first target signal to obtain a third target signal.

[0160] The step S511 can remove the long-term trend component in the first target signal.

[0161] S512. Perform band-pass filtering on the third target signal to obtain a fourth target signal.

[0162] In the embodiment of the present application, a band-pass filter is used to implement the step S512. The band-pass filter can selectively pass signals within a specific frequency range and attenuate or block signals of other frequencies. For adults, the specific frequency range is 0.7 Hz to 3 Hz; for infants, the specific frequency range is 1.5 Hz to 5 Hz.

[0163] The step S512 can retain the pulse wave signal within the heart rate frequency band.

[0164] Further, the step S512 can be expressed as a formula: PPG filtered (t) = BandPass(PPG raw (t)); where PPG filtered (t) is the fourth target signal, BandPass() is the band-pass filter, PPG raw (t) is the third target signal, and t is the current moment.

[0165] S513. Detect the peak corresponding to the red light channel (i.e., the R channel), the valley corresponding to the red light channel, the peak corresponding to the green light channel (i.e., the G channel), and the valley corresponding to the green light channel in the fourth target signal.

[0166] S514. Calculate the amplitude difference corresponding to the red light channel by using the difference between the peak corresponding to the red light channel and the valley corresponding to the red light channel.

[0167] The calculation formula of the step S514 is:

[0168] ;

[0169] where is the amplitude difference corresponding to the red light channel, is the peak corresponding to the red light channel, is the valley corresponding to the red light channel.

[0170] S515. Calculate the amplitude difference corresponding to the green light channel by using the difference between the peak and the valley corresponding to the green light channel.

[0171] The calculation formula for step S515 is:

[0172] ;

[0173] Wherein, is the amplitude difference corresponding to the green light channel; is the peak corresponding to the green light channel; is the valley corresponding to the green light channel.

[0174] S516. Calculate and obtain the relative amplitude feature by dividing the amplitude difference corresponding to the green light channel by the amplitude difference corresponding to the red light channel.

[0175] The calculation formula for step S516 is:

[0176] ;

[0177] Wherein, is the relative amplitude feature, is the amplitude difference corresponding to the green light channel, is the amplitude difference corresponding to the red light channel.

[0178] In other embodiments, the relative amplitude feature may also be calculated by calculating the amplitude of the heart rate frequency corresponding to the red light channel and the amplitude of the heart rate frequency corresponding to the green light channel through Fourier transform. The specific calculation formula is as follows:

[0179] ;

[0180] Wherein, is the relative amplitude feature, is the amplitude of the heart rate frequency corresponding to the green light channel, is the amplitude of the heart rate frequency corresponding to the red light channel, which is also the heart rate frequency.

[0181] S517. Calculate and obtain the blood oxygen saturation based on the relative amplitude feature.

[0182] In the embodiments of the present application, the regression model established by the polynomial regression method is used to implement step S517.

[0183] Specifically, the calculation formula for step S517 is:

[0184] ;

[0185] Wherein, SpO2 is the blood oxygen saturation, k is the cross - band absorption difference correction factor, RRs is the relative amplitude feature, and b is the reference value of the red band.

[0186] The two parameters k and b reflect the correction relationship of the absorption difference of hemoglobin in the red light channel and the green light channel, and both are obtained through pre - training with a large amount of experimental data.

[0187] S52. Calculate the heart rate and heart rate variability of the physiological signal based on the second target signal. The step S52 can be understood as detecting the heart rate and the heart rate variability through a heart rate detection model.

[0188] Exemplarily, the step S52 includes (that is, the implementation steps of the heart rate detection model):

[0189] S521. Identify the R - wave position of the second target signal, and calculate the RR interval based on the R - wave position (that is, the time interval between two adjacent R - wave positions).

[0190] S522. Calculate the heart rate and heart rate variability based on the RR interval.

[0191] In other embodiments, the heart rate detection model combines multi - spectral rPPG signals and SCG signals. The CNN is used to extract time - domain features and frequency - domain features, while the long short - term memory network (LSTM) or Transformer model is used to capture long - term dependencies in the time series and calculate the heart rate value in real - time.

[0192] S53. Calculate the respiratory rate of the physiological signal from the second target signal by using the optical flow method. The step S53 can be understood as detecting the respiratory rate through a respiratory detection model.

[0193] In the embodiments of the present application, by using the optical flow method, the chest movement is converted into optical flow information and encoded. Then, the pixel brightness sequence is extracted to obtain respiratory waveform information. This method can effectively compensate for image offset and blur caused by facial or body movement, thereby improving the accuracy and stability of physiological signal extraction.

[0194] In other embodiments, for respiratory detection, a convolutional neural network (CNN) can also be used to extract chest or abdominal movement features from video images. These features are converted into frequency - domain features through a fast Fourier transform (FFT) to calculate the respiratory rate.

[0195] It should be understood that the respiratory rate can also be estimated through the movement change signal of the patient's abdomen. The respiratory rate is the breathing frequency.

[0196] S6. Feed back the physiological signal to the medical staff.

[0197] In the embodiments of the present application, the physiological signal may be displayed in real time on a display screen outside the MRI room, so that the medical staff can monitor the physiological signal data in real time.

[0198] The embodiments of the present application use multi-spectral pulse wave signals and cardiac vibration signals to detect the physiological signals of patients. Thus, the embodiments of the present application can achieve accurate detection of multi-dimensional physiological parameters in a strong magnetic field environment, enabling medical staff to monitor various physiological indicators in real time and ensuring the safety of patients' lives.

[0199] The embodiments of the present application integrate multi-spectral optical signals (i.e., the multi-spectral pulse wave signals) and cardiac mechanical vibration signals (i.e., the cardiac vibration signals). This method realizes high-precision detection of various physiological signals and improves the reliability and accuracy of data through signal fusion. This design provides technical support for the application of a multi-parameter detection system (i.e., a physiological signal detection system for magnetic resonance scanning), and also lays a foundation for health management, clinical diagnosis, and early screening of cardiovascular diseases.

[0200] In addition, the embodiments of the present application also calculate the pulse wave transmission time difference between the multi-spectral rPPG signal and the SCG signal, as well as the time difference between different spectral rPPG signals, to achieve deeper physiological information mining. These time difference parameters, combined with multi-dimensional physiological features extracted from the signals, can be used to construct a blood pressure estimation model based on machine learning, providing a new method for non-contact blood pressure measurement.

[0201] In some embodiments, the embodiments of the present application adopt a multi-dimensional pulse transit time (MD-PTT) blood pressure detection technology, combined with signal trajectories at different positions and different wavelengths at the same position, to calculate the multi-point pulse wave conduction time and the multi-spectral pulse wave conduction time, and use traditional machine learning algorithms to establish a mapping relationship with blood pressure.

[0202] Specifically, by calculating the peak time interval between the SCG signal (the cardiac vibration signal) and the single-spectral pulse wave signal under any spectrum within the same cardiac cycle, the multi-point pulse wave conduction time (i.e., multi-point PTT) is obtained; by calculating the peak time intervals of the single-spectral pulse wave signals corresponding to all wavelengths of the spectrum within the same cardiac cycle (i.e., calculating the peak time intervals of the rPPG signals extracted under different spectra within the same cardiac cycle), the multi-spectral pulse wave conduction time is obtained.

[0203] The embodiments of the present application can extract the heart rate and heart rate variability based on MD-PTT, and use traditional machine learning algorithms to establish a mapping relationship with blood pressure.

[0204] Furthermore, the steps of establishing the mapping relationship with blood pressure by using traditional machine learning algorithms include: calculating the first systolic peak time difference (PPG-PPW-PTT) between the forehead PPG (G channel) signal and the neck pressure pulse wave (PPW) signal, calculating the second systolic peak time difference (PPG-PPG-PTT) between the forehead PPG and the neck PPG signal, and extracting key features related to blood pressure changes based on the first systolic peak time difference and the second systolic peak time difference. In the embodiments of the present application, the forehead PPG signal can be obtained through the multi-spectral pulse wave signal. The neck pressure pulse wave can be detected by a pressure sensor detection method or a Doppler ultrasound detection method.

[0205] To ensure the accuracy of peak detection, the findpeaks (i.e., peak seeking) function in MATLAB (Matrix Laboratory, an advanced numerical calculation and programming environment) is used, and the main peaks in the signal are accurately located by setting appropriate parameters to ensure the time alignment of each signal segment; the time stamps of the simultaneously collected blood pressure values are used to accurately calculate the average PTT value (i.e., the first systolic peak time difference and the second systolic peak time difference) of each segment. In addition, the embodiments of the present application also extract the augmentation index (AIx) feature. By detecting the peaks, valleys and dicrotic notches in the signal, and dividing multiple signal segments according to the blood pressure value time stamps, the augmentation index of each signal segment is calculated. Subsequently, a polynomial regression model (i.e., ) is used to establish the mapping relationship between PTT (i.e., the first systolic peak time difference and the second systolic peak time difference) and blood pressure, and the polyfit (polynomial fitting) function and polyval (polynomial evaluation) function in MATLAB are used to optimize the regression coefficients by the least squares method to obtain a blood pressure detection model (i.e., a blood pressure estimation model).

[0206] In other embodiments, machine learning algorithms such as support vector machine (SVM) and random forest (RF) can also be used to train the blood pressure detection model.

[0207] In the embodiments of the present application, during real-time detection, the preprocessed multi-spectral pulse wave signal and the cardiac vibration signal (i.e., the first target signal and the second target signal) are input into the trained blood pressure estimation model, and the blood pressure estimation model calculates and outputs the systolic and diastolic blood pressure values of the patient in real time according to the trained mapping relationship. Thereby, the accuracy and real-time performance of blood pressure detection can be ensured.

[0208] The blood pressure estimation model can accurately extract and calculate key physiological parameters from non-contact acquired physiological signals, providing comprehensive, real-time and accurate physiological data support for the detection of patients' physiological signals during magnetic resonance scanning.

[0209] In some embodiments, features such as the heart rate interval (i.e., the RR interval), heart rate variability (HRV), and the changing trend of heart rate are extracted from the heart rate signal (i.e., the cardiac vibration signal), and these features can effectively reflect the stability of the cardiac rhythm and the activity status of the autonomic nervous system.

[0210] Embodiments of the present application extract the respiratory cycle change and the value of respiratory rate from the respiratory frequency signal, detect too fast or too slow breathing conditions (such as less than 10 breaths per minute or more than 25 breaths per minute), and use the Fast Fourier Transform (FFT) to extract frequency domain features (such as the low-frequency and high-frequency components of the power spectral density).

[0211] Embodiments of the present application achieve real-time monitoring of the change in blood oxygen level by extracting waveform features and actual values from the blood oxygen saturation (SpO2) signal. As one of the most sensitive physiological parameters, the blood oxygen saturation can capture minute changes through trend analysis every 10 seconds or 20 seconds. Once it is detected that the blood oxygen saturation is lower than the set threshold (usually 95%) or the downward trend exceeds 3%, the physiological signal detection system for magnetic resonance scanning will immediately trigger an alarm.

[0212] Embodiments of the present application extract the real-time values of systolic blood pressure and diastolic blood pressure from the blood pressure signal, and combine with the fluctuation of blood pressure to timely detect abnormalities (such as systolic blood pressure exceeding 180 mmHg or diastolic blood pressure lower than 60 mmHg). A peak detection algorithm is used in blood pressure waveform analysis to capture the key change points of blood pressure fluctuation.

[0213] In terms of personalized modeling, the physiological signal detection system for magnetic resonance scanning combines the normal range standard of physiological parameters released by authoritative medical institutions as the basis for threshold setting, and integrates machine learning algorithms with the experience of clinicians to refine the threshold setting, ensuring the personalization and accuracy of the detection results.

[0214] Due to the particularity of the magnetic resonance scanning process, patients may be in different physiological states, and the physiological signal detection system for magnetic resonance scanning can dynamically adjust the threshold to meet the needs of different populations and special cases. For this purpose, refer to Figure 3 As shown, the physiological signal detection system for magnetic resonance scanning introduces deep learning methods, such as convolutional neural network (CNN), long short-term memory network (LSTM), and Transformer (i.e., Figure 3("converter model" in it), further optimize personalized modeling. Specifically, CNN can effectively extract spatio-temporal features (i.e., temporal features such as heart rate intervals or respiratory cycles) in physiological signals and dynamically adjust according to individual differences; LSTM can capture long-term dependencies in time series, providing accurate predictions and anomaly detections; through the Transformer model, the system can process and learn multi-dimensional physiological signals from different patient groups and perform more accurate modeling and predictions. For the elderly, high-risk patients or special disease groups, the physiological signal detection system for magnetic resonance scanning can set precise thresholds according to individual differences and establish personalized normal physiological ranges in combination with clustering analysis (such as the K-means algorithm, that is, the K-means clustering algorithm) to ensure the accuracy of real-time detection. In other embodiments, the physiological signal detection system for magnetic resonance scanning also introduces decision trees to further optimize personalized modeling.

[0215] It should be understood that when processing physiological signals, CNN can effectively extract the spatio-temporal features. CNN slides the convolutional kernel in the time dimension through the convolutional layer to extract local time features, and at the same time performs convolution in the spatial dimension of multi-channel signals to extract spatial features. The pooling layer further reduces the dimensions of the time features and spatial features, enhancing the feature robustness. And the fully connected layer integrates these features and outputs the final spatio-temporal features.

[0216] CNN can dynamically adjust its internal parameters, such as weights and biases, according to individual differences to adapt to the physiological characteristics of different individuals. This adjustment not only includes the optimization of model parameters, but also involves the automatic selection and emphasis of feature extraction, as well as the dynamic setting of the normal range threshold of physiological parameters. For example, for individuals with generally higher heart rates, the model will automatically adjust the normal range of heart rate to avoid false alarms. In this way, CNN can not only extract spatio-temporal features, but also dynamically adjust according to individual differences, providing more personalized and accurate physiological detection and early warning.

[0217] LSTM captures long-term dependencies in time series through its unique gating mechanism (input gate, forget gate, and output gate). The input gate determines how much of the current input information is written into the cell state, the forget gate determines how much information in the cell state is retained or forgotten, and the output gate determines how much information in the cell state is output. These gating mechanisms enable LSTM to dynamically update the cell state at each time step, thus retaining long-term context information. For example, when detecting heart rate, LSTM can remember the heart rate changes at previous time points. Even if they are far from the current time point, they can still affect the current output. This ability enables LSTM to capture long-term dependencies in time series and provide support for accurate physiological parameter prediction and anomaly detection.

[0218] Regarding long-term dependencies in time series, LSTM can more accurately predict future physiological parameters such as heart rate, respiratory rate, blood oxygen saturation, and blood pressure, which is crucial for early warning and intervention of potential health problems. Secondly, long-term dependencies help in understanding the context of the signals. For example, understanding the trend of heart rate changes, even if these changes occur over a long time span, can help the model better identify anomalies. Finally, by capturing normal patterns, LSTM can more easily identify anomalies such as sudden changes in heart rate, abnormal fluctuations in respiratory rate, etc., which may indicate potential health risks.

[0219] In addition, the anomaly detection of LSTM mainly focuses on aspects such as heart rate anomalies, respiratory anomalies, blood oxygen saturation anomalies, and blood pressure anomalies. Specifically, LSTM can detect sudden changes in heart rate, such as arrhythmia, tachycardia, or bradycardia; detect abnormal fluctuations in respiratory rate, such as apnea or hyperventilation; detect a decrease in blood oxygen saturation to timely identify hypoxemia; detect sudden increases or decreases in blood pressure to warn of hypertension or hypotension. By comprehensively analyzing multiple physiological parameters, LSTM can also detect potential comprehensive anomalies, such as multi-parameter changes before cardiac arrest. These anomaly detection functions enable LSTM to detect the physiological state of patients in real time and immediately issue an alarm when an anomaly is detected, reminding doctors to take timely intervention measures to ensure the safety and health of patients.

[0220] The steps of the clustering analysis to establish the personalized normal physiological range include: First, remove the missing values and outliers of the physiological signals to ensure the integrity and accuracy of the signals, and perform standardization or normalization on the physiological signals to eliminate the dimensional differences between different physiological parameters (such as heart rate, respiratory rate, blood oxygen saturation, etc.), ensuring that the influence of each parameter on the clustering analysis is balanced. Next, select the K-means clustering algorithm for data analysis, and preset the number of clusters K. The selection of this K value can be determined by methods such as the elbow method or the silhouette coefficient to find the optimal number of clusters. The process of the K-means algorithm includes randomly selecting K cluster centers and assigning the data points to the nearest cluster by calculating the distance from each data point to each cluster center. Then, recalculate the center of each cluster (i.e., the mean of all data points within the cluster), and repeat this process until the cluster centers no longer change significantly. Finally, divide all data points into K clusters, and each cluster represents a physiological state. After completing the clustering, it is necessary to conduct a detailed analysis of the clustering results to identify the cluster representing the normal physiological range, which usually contains the data of healthy individuals. According to the statistical characteristics (such as mean and standard deviation) of this cluster, the range of normal physiological parameters can be set. For example, the normal range of heart rate can be set as the mean of the heart rate data within this cluster plus or minus one standard deviation. Finally, verify the rationality of this normal range by combining historical data to ensure that it can accurately reflect the physiological state of healthy individuals, and make dynamic adjustments through real-time detection data to adapt to the differences of different individuals.

[0221] In some embodiments, referring to Figure 4 as shown, after the step of extracting the physiological signal with multiple physiological parameters based on the first target signal and the second target signal, the physiological signal detection method for magnetic resonance scanning further includes:

[0222] S7. When at least one physiological parameter of the physiological signal reaches the first preset threshold, activate the alarm mechanism.

[0223] The heart rate, the heart rate variability, the blood oxygen saturation, and the respiratory rate are all physiological parameters.

[0224] The first preset threshold can be a preset normal fluctuation range, and specifically can be set by those skilled in the art according to actual needs.

[0225] The alarm mechanism reminds medical staff or the patient himself through visual (such as flashing red light) signals (i.e., optical signals) and auditory (such as alarm sound) signals (i.e., acoustic signals) to ensure that corresponding measures can be taken in a timely manner in different environments. The alarm mechanism has instantaneity and can quickly notify relevant personnel to handle the occurred abnormal situation.

[0226] S8. When the trending change state of at least one physiological parameter of the physiological signal reaches a dangerous state, start the warning mechanism.

[0227] The trending change state of a physiological parameter refers to the regular or continuous change pattern presented by the physiological parameter over time within a period. Such a change pattern can be ascending, descending, fluctuating, or remaining stable, etc.

[0228] The warning mechanism is a trend prediction.

[0229] The warning mechanism dynamically detects the change trend of physiological parameters (such as by combining the Dynamic Time Warping (DTW) algorithm) to predict potential risks in advance. When the physiological signal detection system for magnetic resonance scanning detects that the trending change of a key physiological parameter may reach a dangerous state (i.e., reach the danger threshold), the warning mechanism will issue a prompt to gain more response time for medical staff and patients and intervene in potential safety problems of patients in advance.

[0230] Through precise health detection and risk prediction, and combining the dual mechanisms of alarm and warning, the physiological signal detection system for magnetic resonance scanning in this application embodiment can significantly reduce health risks. Especially during the CMR scanning of patients with cardiovascular diseases, this dual mechanism combining alarm and warning significantly improves the safety of patients and the accuracy of diagnosis by medical staff, provides more sufficient response time for clinical intervention, and further optimizes the intelligent and precise level of health management.

[0231] In some embodiments, the physiological signal detection method for magnetic resonance scanning further includes saving the physiological signal. In some embodiments, the first preset threshold is generated by a health detection model.

[0232] The training method of the health detection model is as follows: Obtain the historical record data and the true label set of the patient, and each data in the historical record data has a one-to-one correspondence with one of the true labels in the true label set; input the target training data into the Transformer model to obtain a prediction threshold, where the target training data is any one of the historical record data; calculate the loss function based on the prediction threshold and the true label corresponding to the target training data; adjust the network parameters of the Transformer model based on the loss function to obtain the health detection model.

[0233] The historical record data refers to physiological signals obtained from real-time detections in the past. The historical record data may also include initial data with annotations. Physiological signals obtained from real-time detections in the past may include heart rate (HR), heart rate variability (HRV), respiratory rate (RR), blood oxygen saturation (SpO2), heart beat interval (RR interval), the changing trend of heart rate, the change of respiratory cycle, the value of respiratory rate, frequency domain characteristics, waveform trend changes, and peak points. The specific mechanisms of the Transformer model include multi-head self-attention, feed forward network, residual connection, and layer normalization. Among them, multi-head self-attention simultaneously focuses on features at different positions through multi-head self-attention to capture long-range dependencies in the sequence; the feed forward network performs non-linear transformations on the features at each position to extract higher-level features; the residual connection avoids the problems of gradient disappearance and gradient explosion and improves the training stability of the model; layer normalization further improves the training stability and performance of the model. These mechanisms enable the Transformer model to effectively process multi-dimensional physiological signals, provide accurate prediction results, and help doctors timely detect potential health problems and take intervention measures.

[0234] Further, the calculation formula of the loss function is ; where is the loss function, is the total number of the historical record data, is the true label corresponding to the target training data (the prediction threshold corresponding to the th data in the historical record data), is the prediction threshold.

[0235] Further, the embodiments of the present application can also optimize the performance of the health detection model by means of cross-validation and hyperparameter tuning.

[0236] It should be understood that the health detection model in the embodiments of the present application is updated regularly according to real-time data to ensure that it can adapt to changes in the physiological state of the patient. If the value of the physiological parameter is detected to exceed the personalized set normal range (i.e., the first preset threshold) or the change trend of the physiological parameter reaches a dangerous state, the health detection model will automatically adjust its internal parameters and immediately activate the alarm mechanism or the early warning mechanism. This adjustment of internal parameters may involve changing feature weights, updating thresholds, or adjusting the decision boundary of the model to better adapt to the current physiological state of the patient. Exemplarily, if the patient's blood pressure continues to be high, the health detection model may increase the weight of the features related to blood pressure and accordingly adjust the blood pressure threshold to more sensitively detect blood pressure changes.

[0237] In other embodiments, the Transformer model can be replaced by a convolutional neural network (CNN), or a long short-term memory network (LSTM), linear regression, support vector machine, or random forest.

[0238] It should also be mentioned that considering the need for non-contact detection, wireless sensor technology or a combination of PPG, SCG, and other physiological information detection technologies can be explored to replace the existing optical sensing technology. These technologies can further reduce the interference to the patient while providing real-time detection of physiological parameters.

[0239] In addition, bioelectrical signals (such as electrocardiogram ECG) can be considered to replace the existing optical sensing technology. Although the acquisition of bioelectrical signals requires the use of contact sensors, the discomfort of the patient can be minimized by improving the design of the sensors.

[0240] Reference Figure 5 shown is a schematic block diagram of a physiological signal detection system for magnetic resonance scanning provided in the second aspect of the embodiments of the present application. In Figure 5 it, the physiological signal detection system 100 for magnetic resonance scanning includes:

[0241] A multi-spectral pulse wave signal acquisition module 101, configured to acquire a multi-spectral pulse wave signal of the face of the patient by using a time-sequential spectroscopic method during the process of the magnetic resonance device of the physiological signal detection system for magnetic resonance scanning scanning the patient;

[0242] A cardiac vibration signal acquisition module 102, configured to acquire a cardiac vibration signal of the chest of the patient by using a vibration measurement method while acquiring the multi-spectral pulse wave signal;

[0243] A first target signal obtaining module 103, configured to preprocess the multi-spectral pulse wave signal to obtain a first target signal;

[0244] The second target signal acquisition module 104 is configured to preprocess the cardiac vibration signal to obtain a second target signal;

[0245] The physiological signal extraction module 105 is configured to extract physiological signals with multiple physiological parameters based on the first target signal and the second target signal;

[0246] The feedback module 106 is configured to feedback the physiological signal to medical staff.

[0247] Reference Figure 6 As shown, it is a schematic block diagram of a physiological signal detection system for magnetic resonance scanning provided in the third aspect of the embodiments of the present application. In Figure 6 it, the physiological signal detection system 200 for magnetic resonance scanning includes: a processor 201, a memory 202, a multimodal non-contact multi-parameter detection optical device 203, a laser 204 (which can refer to Figure 2 the laser lamp in), and a magnetic resonance device 205. Among them, as shown in Figure 7 the multimodal non-contact multi-parameter detection optical device 203 includes a first camera 2031 and a second camera 2032. The memory 202 is used to store computer programs, and the processor 201 is used to call and run the computer programs stored in the memory 202, and execute the steps of the physiological signal detection method for magnetic resonance scanning described in the first aspect of the embodiments of the present application. The magnetic resonance device 205, the first camera 2031, and the second camera 2032 are all communicatively connected to the processor 201, and the laser 204 is disposed on one side of the second camera 2032;

[0248] The first camera 2031 is configured to perform image acquisition operations on the skin of the patient's face to obtain a first image sequence, and transmit the first image sequence to the processor 201;

[0249] The second camera 2032 is configured to perform image acquisition operations on the chest of the patient to obtain a second image sequence, and transmit the second image sequence to the processor 201;

[0250] When the second camera 2032 performs image acquisition operations on the chest, the laser 204 emits laser light to irradiate the chest.

[0251] In some embodiments, refer to Figure 8As shown, the physiological signal detection system 200 for magnetic resonance scanning also includes an early warning module 206. The early warning module 206 combines the dual mechanisms of alarm and early warning, which can not only quickly notify relevant personnel to deal with abnormal situations that have occurred, but also predict potential risks in advance based on trend changes in physiological characteristics, thereby gaining more response time for medical staff and patients, intervening in potential problems in advance, significantly reducing health risks, providing personalized decision-making support, and further ensuring the life safety of patients.

[0252] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium is used to store a computer program, wherein the computer program enables a computer to execute the steps of the physiological signal detection method for magnetic resonance scanning provided in the first aspect of the embodiment of the present application.

[0253] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and 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-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an 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 (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0254] The technical features of the above embodiments can be combined without changing the basic principles of the present application. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0255] The above embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the scope of the patented application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the scope of patent protection of the present application shall be subject to the appended claims.

Claims

1. A physiological signal detection method for magnetic resonance scanning, characterized in that: include: In the process of scanning a patient by a magnetic resonance device of a physiological signal detection system for magnetic resonance scanning, a multi-spectral pulse wave signal of the patient's face is obtained by using a time-series spectroscopic method; While acquiring the multi-spectral pulse wave signal, using a vibration measurement method to acquire a heart vibration signal of the patient's chest; Preprocessing the multi-spectral pulse wave signal to obtain a first target signal; Preprocessing the heart vibration signal to obtain a second target signal; extracting a physiological signal having a plurality of physiological parameters based on the first target signal and the second target signal; Feeding back the physiological signal to medical personnel; The extracting of the physiological signal having multiple physiological parameters based on the first target signal and the second target signal includes: extracting the blood oxygen saturation of the physiological signal from the first target signal; The step of extracting the blood oxygen saturation of the physiological signal from the first target signal includes: performing detrending processing on the first target signal to obtain a third target signal; performing bandpass filtering processing on the third target signal to obtain a fourth target signal; detecting the peak corresponding to the red light channel, the trough corresponding to the red light channel, the peak corresponding to the green light channel, and the trough corresponding to the green light channel in the fourth target signal; using the difference between the peak corresponding to the red light channel and the trough corresponding to the red light channel to calculate the amplitude difference corresponding to the red light channel; using the difference between the peak corresponding to the green light channel and the trough corresponding to the green light channel to calculate the amplitude difference corresponding to the green light channel; dividing the amplitude difference corresponding to the green light channel by the amplitude difference corresponding to the red light channel to calculate a relative amplitude feature; and calculating the blood oxygen saturation based on the relative amplitude feature; The calculation formula of the step of calculating the amplitude difference corresponding to the red light channel by using the difference between the peak corresponding to the red light channel and the trough corresponding to the red light channel is: ;in, is the amplitude difference corresponding to the red light channel; is the peak corresponding to the red light channel; is the trough corresponding to the red light channel; The calculation formula of the step of calculating the amplitude difference corresponding to the green light channel by using the difference between the peak corresponding to the green light channel and the trough corresponding to the green light channel is: ; is the amplitude difference corresponding to the green light channel; is the peak corresponding to the green light channel; is the trough corresponding to the green light channel; The calculation formula for the step of calculating the relative amplitude feature by dividing the amplitude difference corresponding to the green light channel by the amplitude difference corresponding to the red light channel is: ; is the relative amplitude characteristic; The calculation formula of the step of calculating and obtaining the blood oxygen saturation based on the relative amplitude feature is: ; is the blood oxygen saturation; is the correction factor for cross-band absorption differences, is the red band reference value; and The two parameters reflect the correction relationship between the red and green light channels for the difference in hemoglobin absorption; The step of preprocessing the multi-spectral pulse wave signal to obtain the first target signal includes: performing high-pass filtering on the multi-spectral pulse wave signal to obtain a first signal to be processed; performing motion compensation on the first signal to be processed to obtain a second signal to be processed; performing noise reduction on the second signal to be processed to obtain the first target signal; The steps of performing motion compensation processing on the first signal to be processed to obtain the second signal to be processed include: dividing the first signal to be processed into multiple first signal segments; detecting a first feature point in each first signal segment, wherein the first feature point is a peak, a trough or an inflection point in the corresponding first signal segment; using an optical flow method to combine all the first feature points and all the first signal segments, and calculating a first optical flow field between adjacent first signal segments in all the first signal segments, wherein the first optical flow field describes the movement direction and speed of each first feature point in adjacent first signal segments; calculating a first compensation amount for each first feature point based on all the first optical flow fields, wherein the first compensation amount includes translation and rotation motion parameters, and the first compensation amount is used to correct signal distortion caused by motion; performing motion compensation processing on the first signal to be processed based on the first compensation amount of each first feature point, and obtaining the second signal to be processed.

2. The physiological signal detection method for magnetic resonance scanning according to claim 1, characterized in that: The steps of acquiring the multi-spectral pulse wave signal of the patient's face by using the time-series spectroscopic method include: receiving a first image sequence transmitted by a first camera of the physiological signal detection system for magnetic resonance scanning, wherein the first image sequence comprises an image sequence of multiple frame periods, and the image sequence of each frame period is an image sequence obtained by the first camera using multiple spectra of different wavelengths to sequentially photograph the face in a spectral order; In the first image sequence, first images corresponding to spectrums of the same wavelength are combined into a single spectrum video to obtain a plurality of spectrum videos; Performing a time series analysis on all pixels in a target spectral video, and using a skin reflectance model to model the change in the reflectance value of each pixel in the RGB channel of the target spectral video over time, to obtain a single spectrum pulse wave signal corresponding to the target spectral video, wherein the target spectral video is any one of the multiple spectral videos; All single-spectrum pulse wave signals constitute the multi-spectrum pulse wave signal.

3. The physiological signal detection method for magnetic resonance scanning according to claim 1, characterized in that: The steps of obtaining the cardiac vibration signal of the patient's chest by using the vibration measurement method include: receiving a second image sequence transmitted by a second camera of the physiological signal detection system for magnetic resonance scanning; Calculating a first motion amplitude in the X direction and a second motion amplitude in the Y direction of each second image in the second image sequence by using an optical flow method; combining the first motion amplitudes of all the second images in the X direction into a first motion amplitude sequence; combining the second motion amplitudes of all the second images in the Y direction into a second motion amplitude sequence; extracting motion angles from the first motion amplitude sequence and the second motion amplitude sequence using a sliding window method; The heart vibration signal is obtained based on the movement angle.

4. The physiological signal detection method for magnetic resonance scanning according to claim 1, characterized in that: The step of preprocessing the heart vibration signal to obtain a second target signal comprises: Performing high-pass filtering on the heart vibration signal to obtain a third signal to be processed; Performing motion compensation processing on the third signal to be processed to obtain a fourth signal to be processed; Perform noise reduction processing on the fourth signal to be processed to obtain the second target signal.

5. The physiological signal detection method for magnetic resonance scanning according to claim 1, characterized in that: The step of extracting a physiological signal having a plurality of physiological parameters based on the first target signal and the second target signal further includes: Calculating the heart rate and heart rate variability of the physiological signal based on the second target signal; The respiratory rate of the physiological signal is calculated from the second target signal using an optical flow method.

6. The physiological signal detection method for magnetic resonance scanning according to claim 1, characterized in that: After the step of extracting a physiological signal having a plurality of physiological parameters based on the first target signal and the second target signal, the physiological signal detection method for magnetic resonance scanning further includes: When at least one physiological parameter of the physiological signal reaches a first preset threshold, an alarm mechanism is activated; When the trend change state of at least one physiological parameter of the physiological signal reaches a dangerous state, the early warning mechanism is activated.

7. A physiological signal detection system for magnetic resonance scanning, characterized in that: include: A multi-spectral pulse wave signal acquisition module, used for acquiring a multi-spectral pulse wave signal of the face of the patient by using a time-series spectroscopic method during the process of scanning the patient by a magnetic resonance device of a physiological signal detection system for magnetic resonance scanning; A cardiac vibration signal acquisition module, used to acquire the cardiac vibration signal of the patient's chest by using a vibration measurement method while acquiring the multi-spectral pulse wave signal; A first target signal acquisition module, used for preprocessing the multi-spectral pulse wave signal to obtain a first target signal; A second target signal acquisition module, used for preprocessing the heart vibration signal to obtain a second target signal; A physiological signal extraction module, configured to extract a physiological signal having a plurality of physiological parameters based on the first target signal and the second target signal; A feedback module, used for feeding back the physiological signal to medical staff; The physiological signal extraction module is also used to extract the blood oxygen saturation of the physiological signal from the first target signal; The physiological signal extraction module is also used to perform detrending processing on the first target signal to obtain a third target signal; perform bandpass filtering processing on the third target signal to obtain a fourth target signal; detect the peak corresponding to the red light channel, the trough corresponding to the red light channel, the peak corresponding to the green light channel, and the trough corresponding to the green light channel in the fourth target signal; use the difference between the peak corresponding to the red light channel and the trough corresponding to the red light channel to calculate the amplitude difference corresponding to the red light channel; use the difference between the peak corresponding to the green light channel and the trough corresponding to the green light channel to calculate the amplitude difference corresponding to the green light channel; divide the amplitude difference corresponding to the green light channel by the amplitude difference corresponding to the red light channel to calculate the relative amplitude feature; Calculate and obtain the blood oxygen saturation based on the relative amplitude feature; The calculation formula of the step of calculating the amplitude difference corresponding to the red light channel by using the difference between the peak corresponding to the red light channel and the trough corresponding to the red light channel is: ;in, is the amplitude difference corresponding to the red light channel; is the peak corresponding to the red light channel; is the trough corresponding to the red light channel; The calculation formula of the step of calculating the amplitude difference corresponding to the green light channel by using the difference between the peak corresponding to the green light channel and the trough corresponding to the green light channel is: ; is the amplitude difference corresponding to the green light channel; is the peak corresponding to the green light channel; is the trough corresponding to the green light channel; The calculation formula for the step of calculating the relative amplitude feature by dividing the amplitude difference corresponding to the green light channel by the amplitude difference corresponding to the red light channel is: ; is the relative amplitude characteristic; The calculation formula of the step of calculating and obtaining the blood oxygen saturation based on the relative amplitude feature is: ; is the blood oxygen saturation; is the correction factor for cross-band absorption differences, is the red band reference value; and The two parameters reflect the correction relationship between the red and green light channels for the difference in hemoglobin absorption; The first target signal acquisition module is further used to perform high-pass filtering on the multi-spectral pulse wave signal to obtain a first signal to be processed; perform motion compensation on the first signal to be processed to obtain a second signal to be processed; perform noise reduction on the second signal to be processed to obtain the first target signal; The first target signal acquisition module is also used to divide the first signal to be processed into multiple first signal segments; detect a first feature point in each first signal segment, the first feature point is a peak, a trough or an inflection point in the corresponding first signal segment; use the optical flow method to combine all the first feature points and all the first signal segments, and calculate the first optical flow field between adjacent first signal segments in all the first signal segments, the first optical flow field describes the movement direction and speed of each first feature point in adjacent first signal segments; calculate the first compensation amount of each first feature point based on all the first optical flow fields, the first compensation amount includes translation and rotation motion parameters, and the first compensation amount is used to correct signal distortion caused by motion; based on the first compensation amount of each first feature point, motion compensation processing is performed on the first signal to be processed to obtain the second signal to be processed.

8. A physiological signal detection system for magnetic resonance scanning, characterized in that: include: A processor, a memory, a multimodal non-contact multi-parameter detection optical device, a laser and a magnetic resonance device, wherein the multimodal non-contact multi-parameter detection optical device comprises a first camera and a second camera, the memory is used to store a computer program, the processor is used to call and run the computer program stored in the memory, and execute the steps of the physiological signal detection method for magnetic resonance scanning according to any one of claims 1 to 6, the magnetic resonance device, the first camera and the second camera are all connected to the processor in communication, and the laser is arranged on one side of the second camera; The first camera is used to perform an image acquisition operation on the skin on the patient's face to obtain a first image sequence, and transmit the first image sequence to the processor; The second camera is used to perform an image acquisition operation on the patient's chest to obtain a second image sequence, and transmit the second image sequence to the processor; When the second camera performs an image acquisition operation on the chest, the laser emits laser light to irradiate the chest.

9. A computer-readable storage medium, characterized in that: Used to store a computer program, wherein the computer program enables a computer to execute the steps of the physiological signal detection method for magnetic resonance scanning as described in any one of claims 1 to 6.

Images