Non-contact myocardial infarction detection method, device and equipment and storage medium

Abstract

The invention provides a non-contact myocardial infarction detection method, device and equipment and a storage medium, and the non-contact myocardial infarction detection method comprises the steps: collecting a radio frequency signal fed back by a target user, and determining the motion state of the target user based on the radio frequency signal; the motion state comprises a standing state and a slight body movement state; according to the motion state of the target user, recovering a heartbeat waveform of the target user, and extracting a heartbeat signal; carrying out segmentation processing on the heartbeat signal of the target user; and extracting physiological features from the segmented heartbeat signals of the target user, and classifying the target user based on the physiological features of the target user to obtain a classification result. According to the method and the device, the detection accuracy under the condition of slight movement is remarkably improved, and the problem of poor myocardial infarction detection accuracy in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the field of medical detection technology, and in particular to a non-contact myocardial infarction detection method, device, equipment and storage medium. Background Art

[0002] Myocardial infarction (MI), commonly known as a heart attack, is an acute disease caused by coronary artery blockage, which leads to insufficient myocardial oxygen supply and myocardial cell necrosis. The high mortality rate and serious sequelae of MI make early detection a critical medical need. Although traditional MI detection methods, such as electrocardiogram (ECG) and blood tests, can effectively identify cardiac abnormalities, their limitations in practical applications are becoming increasingly apparent.

[0003] In the field of smart health monitoring, radio frequency (RF) signals have received more and more attention as an emerging non-contact physiological signal sensing method. RF-based heart health monitoring technology extracts physiological information such as heart rate and breathing rate by transmitting and receiving electromagnetic waves reflected from the human body, and then evaluates health status. Compared with traditional electrocardiogram and blood test methods, RF signal monitoring does not require physical contact and has higher convenience and sustainability. Especially for the elderly or patients with long-term diseases, RF technology can avoid the discomfort and limitations brought by conventional detection equipment, making health monitoring more popular and sustainable.

[0004] However, RF signals face many technical challenges in their applications. First, the processing of RF signals requires more complex algorithm support, including signal acquisition, filtering, demodulation, feature extraction and other steps. Each link may be affected by different factors, resulting in instability in heart rate detection results. For example, although the RF system performs well in static conditions, once the user has slight body movements, such as arm shaking, leg swinging, etc., the noise in the signal will increase significantly, thereby affecting the accuracy of heart rate extraction. Therefore, how to maintain high detection accuracy when the user is performing slight activities is an important technical problem facing the RF health monitoring system.

[0005] Secondly, since the propagation of RF signals is easily affected by physiological phenomena such as human breathing, accurately extracting heart rate signals has become the primary challenge facing RF heart monitoring systems. Traditional heart rate detection methods based on RF signals rely on simple spectrum analysis techniques, but due to the influence of human motion and breathing on RF signals, these methods are difficult to effectively separate heartbeat signals, especially when the user makes slight body movements, the signal will be severely interfered.

[0006] To overcome these problems, researchers have proposed a variety of improvement plans. For example, the more advanced frequency modulated continuous wave (FMCW) radar technology is used to accurately measure the phase changes of the echo to obtain higher resolution physiological signals. This technology can effectively capture tiny chest displacements, thereby reflecting the periodic fluctuations of heartbeat and breathing. However, due to the overlap between the frequency ranges of heartbeat and breathing, these methods still have difficulty in accurately separating heartbeat signals in complex environments, especially when the user makes slight body movements, and the robustness of traditional signal processing methods is poor.

[0007] In addition, the effective application of RF signals also faces the problem of environmental adaptability. Different environmental conditions will have different effects on the propagation of RF signals. For example, walls, furniture and other objects may change the propagation path of the signal or even introduce additional noise. In this case, how to suppress environmental noise and extract effective physiological information through precise signal processing algorithms has become one of the key issues in RF health monitoring technology.

[0008] There is currently no effective solution to the problem of poor accuracy in myocardial infarction detection in existing related technologies. Summary of the invention

[0009] The present invention provides a non-contact myocardial infarction detection method, device, equipment and storage medium, which are used to solve the defect of poor accuracy of myocardial infarction detection in the prior art.

[0010] In a first aspect, the present invention provides a non-contact myocardial infarction detection method, comprising: Collecting the radio frequency signal fed back by the target user, and determining the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; According to the motion state of the target user, restore the heartbeat waveform of the target user and extract the heartbeat signal; Segmenting the heartbeat signal of the target user; Extracting physiological features from the target user's segmented heartbeat signal, and classifying the target user based on the target user's physiological features to obtain a classification result.

[0011] According to a non-contact myocardial infarction detection method provided by the present invention, a radio frequency signal fed back by the target user is collected, and a motion state of the target user is determined based on the radio frequency signal, including: Collecting and transmitting a radio frequency signal modulated by the breathing and heartbeat of the target user through a pulse coherent radar; the radio frequency signal includes a plurality of continuous IQ data; determining the amplitude of motion between two pulses based on a phase difference between two consecutive IQ data; The amplitude variance of the motion amplitude is compared with a preset detection threshold to determine the motion state of the target user.

[0012] According to a non-contact myocardial infarction detection method provided by the present invention, when the target user is in a stationary state, the heartbeat waveform of the target user is restored according to the motion state of the target user, and the heartbeat signal is extracted, including: Combining the phase differences of the IQ data of the target user at each time to obtain a phase difference sequence signal; Performing a fast Fourier transform on the phase difference sequence signal, and constructing a bandpass filter to filter the radio frequency signal; The filtered radio frequency signal is subjected to inverse Fourier transform to obtain a heartbeat waveform of the target user, and a heartbeat signal is extracted based on the heartbeat waveform.

[0013] According to a non-contact myocardial infarction detection method provided by the present invention, when the target user is in a state of slight body movement, the heartbeat waveform of the target user is restored according to the movement state of the target user, and the heartbeat signal is extracted, including: Combining the phase differences of the IQ data of the target user at each time to obtain a phase difference sequence signal; Adding white noise to the phase difference sequence signal, and performing empirical mode decomposition to obtain an intrinsic mode function sequence; Convert each of the intrinsic mode function sequences to the frequency domain to determine the main frequency of each of the intrinsic mode function sequences; The heartbeat frequency range of the target user is determined based on the main frequency of the intrinsic mode function sequence, and the heartbeat signal of the target user is extracted.

[0014] According to a non-contact myocardial infarction detection method provided by the present invention, when the target user is in a static state, segmenting the heartbeat signal of the target user includes: Reversely process the phase difference sequence signal, traverse the phase difference sequence signal, determine the peak function of each point, and generate a peak set; The peak value set is traversed in order of index, points in the peak value set are sorted and filtered, and divided into multiple segments to form a heartbeat signal segment set.

[0015] According to a non-contact myocardial infarction detection method provided by the present invention, when the target user is in a static state, segmenting the heartbeat signal of the target user includes: Set the standard template for heartbeat signal; By means of a dynamic programming method, the standard template is iteratively updated with the goal of minimizing the error between the signal segment of the heartbeat signal and the standard template until the iteration converges, thereby obtaining the segmentation result of the heartbeat signal and the target template.

[0016] According to a non-contact myocardial infarction detection method provided by the present invention, physiological features are extracted from the heartbeat signal of the target user after segmentation processing, and the target user is classified based on the physiological features of the target user to obtain a classification result, including: Extracting physiological features from the segmented heartbeat signal; The pre-trained classification model is called to classify the target user according to the extracted physiological features to obtain a detection result.

[0017] In a second aspect, the present invention further provides a non-contact myocardial infarction detection device, comprising: A collection module, used for collecting the radio frequency signal fed back by the target user, and determining the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; An extraction module, used to restore the heartbeat waveform of the target user and extract the heartbeat signal according to the motion state of the target user; A segmentation module, used for segmenting the heartbeat signal of the target user; The detection module is used to extract physiological features from the heartbeat signal of the target user after segmentation processing, and classify the target user based on the physiological features of the target user to obtain a classification result.

[0018] In a third aspect, the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the non-contact myocardial infarction detection method as described in the first aspect above is implemented.

[0019] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the non-contact myocardial infarction detection method as described in the first aspect above.

[0020] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the non-contact myocardial infarction detection method as described in the first aspect above.

[0021] Compared with the prior art, the present invention has the following beneficial effects: The non-contact myocardial infarction detection method provided by the present invention not only enhances the stability of the radio frequency signal, but also significantly improves the detection accuracy under mild exercise conditions by processing the radio frequency signal differently according to the target user's motion state, making it closer to real life scenarios. For example, when the user is watching TV or working, he can still monitor his heart rate changes in real time. This adaptability provides great convenience for family and personal health monitoring, and solves the problem of poor accuracy of myocardial infarction detection in existing related technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a flow chart of the non-contact myocardial infarction detection method provided by the present invention; Figure 2 It is a structural block diagram of the non-contact myocardial infarction detection device provided by the present invention; Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The present invention provides a non-contact myocardial infarction detection method. Figure 1 is a flow chart of the non-contact myocardial infarction detection method provided by the present invention, such as Figure 1 As shown, the method comprises the following steps: Step S101, collecting radio frequency signals fed back by a target user, and determining the target user's motion state based on the radio frequency signals; the motion state includes a static state and a slight body movement state; Step S102, according to the motion state of the target user, restore the heartbeat waveform of the target user and extract the heartbeat signal; Step S103, segmenting the heartbeat signal of the target user; Step S104, extracting physiological features from the segmented heartbeat signal of the target user, and classifying the target user based on the physiological features of the target user to obtain a classification result.

[0026] In this method, first, the radio frequency signal fed back by the target user is collected, and then the target user is judged whether he is in a static state or a slight body movement state according to the radio frequency signal. Among them, the slight body movement state is a slight physical activity that can be tolerated during the detection process, including reading, eating, surfing the Internet, etc., and the target user will have a slight forward leaning, backward leaning or small movement of the limbs in these activities. Then, according to the movement state of the target user, the corresponding means are used to restore the heartbeat waveform of the target user and extract the heartbeat signal. On this basis, the extracted heartbeat signal is segmented to extract the specific information of each heartbeat cycle. Finally, the physiological characteristics of the target user are extracted from the segmented heartbeat signal, and the target user is classified based on the physiological characteristics to determine whether the target user is a myocardial infarction population. In the existing detection technology, only when the target user is allowed to remain static can a high accuracy rate be achieved. In the above process, the radio frequency signal is processed differently according to the movement state of the target user, which not only enhances the stability of the radio frequency signal, but also significantly improves the detection accuracy under slight movement conditions, making it closer to the actual life scene. For example, when users are watching TV or working, they can still monitor their heart rate changes in real time. This adaptability provides great convenience for family and personal health monitoring and solves the problem of poor accuracy of myocardial infarction detection in existing related technologies.

[0027] In some of the embodiments, step S101, collecting the radio frequency signal fed back by the target user, and determining the motion state of the target user based on the radio frequency signal, includes: collecting and transmitting the radio frequency signal modulated by the breathing and heartbeat of the target user through a pulse coherent radar; the radio frequency signal includes a plurality of continuous IQ data; determining the motion amplitude between two pulses based on the phase difference between two continuous IQ data; comparing the amplitude variance of the motion amplitude with a preset detection threshold to determine the motion state of the target user.

[0028] For example, a pulse coherent radar (PCR) is first used to transmit an RF signal and capture the reflection modulated by breathing and heartbeat. PCR can achieve high accuracy over a relatively long distance. The transmission signal of PCR is as follows:

[0029] in, is a rectangular pulse, is the frequency, is the time index, is the phase, It is a pulse signal that alternates between 0 and 1. The signal received by PCR is as follows:

[0030] in, represents the channel response, It's a delay. It is noise. is the time index. The output of PCR is IQ data, which can be considered as an extension of the envelope data. IQ data is usually represented as a complex number, where the real part represents the amplitude of the radar signal and the imaginary part represents the phase of the radar signal. represents IQ data, where a and b is a real number, is the imaginary unit. For two consecutive IQ data samples, the motion between the two pulses is calculated based on their phase difference:

[0031]

[0032] in, is the phase of the signal, is the wavelength, It's in time t The displacement distance at is the real part of the IQ data, indicating the amplitude information of the signal, It is the imaginary part of the IQ data, indicating the phase information of the signal. The chest movement caused by breathing and heartbeat is a complex movement, including the expansion, contraction and vibration of the chest, and the phase change of the signal is relative to these complex movements.

[0033] The chest movement caused by breathing in adults ranges from 4 to 12 mm, while the movement caused by heartbeat is about 0.5 mm. Slight chair movement (including leaning forward and backward, shaking legs, and moving arms) will not exceed 30 cm. Set the two detection thresholds to 2 and 30 cm. If the amplitude variance is less than 2 cm, the user is considered to be stationary and the stationary heartbeat segmentation algorithm is used; if the amplitude variance is greater than 2 cm and less than 30 cm, the user is considered to be performing sitting movements and the motion heartbeat segmentation algorithm is used; if the amplitude change is greater than 30 cm, the user is considered to be performing significant movements and the RF signal is discarded.

[0034] In some of the embodiments, when the target user is in a stationary state, a bandpass filter is used to process the radio frequency signal. The bandpass filter filters out breathing and other low-frequency noise signals, retaining only the heartbeat frequency component, making the heartbeat waveform clearer. Specifically, step S102, according to the motion state of the target user, restores the heartbeat waveform of the target user and extracts the heartbeat signal, including: combining the phase difference of the IQ data of the target user at each time to obtain a phase difference sequence signal; performing a fast Fourier transform on the phase difference sequence signal, constructing a bandpass filter to filter the radio frequency signal; performing an inverse Fourier transform on the filtered radio frequency signal to obtain the heartbeat waveform of the target user, and extracting the heartbeat signal based on the heartbeat waveform.

[0035] Exemplarily, first, each time phase difference in the IQ data is Combine to get the phase difference sequence signal Then, the phase difference sequence signal Perform a fast Fourier transform (FFT), select the 0.8-3.6 Hz frequency band, and determine the frequency set within this band and the frequency at which it has its maximum amplitude . Select the maximum amplitude and two adjacent frequency points to create a custom narrowband bandpass filter to remove other frequency interference. Finally, perform an inverse Fourier transform (IFFT) on the filtered frequency domain data to obtain the filtered time domain data, which is the heartbeat waveform.

[0036] Based on this embodiment, step S103, the heartbeat signal of the target user is segmented, including: reverse processing the phase difference sequence signal, traversing the phase difference sequence signal, determining the peak function of each point, and generating a peak set; traversing the peak set in sequence index, sorting and filtering the points in the peak set, and dividing it into multiple segments to form a heartbeat signal segment set.

[0037] Exemplarily, first, initialize the peak set is empty and the phase difference sequence signal Reverse. Ergodic phase difference sequence signal , calculate the peak function at each point = , used to judge the point Is it The window of points is a peak value. The condition for whether it is a true peak is ,in and Separate arrays The mean and standard deviation of h is a constant parameter, set to 3. Select significant peaks and add them to the peak set T , for distances less than k The adjacent peaks are retained, the larger value is retained, and the smaller value is removed. Finally, the sorted peak signal is divided into multiple segments to form a heartbeat signal segment set S .

[0038] In some of the embodiments, when the target user is in a state of slight body movement, white noise is added to the radio frequency signal to improve the extreme value distribution of the signal, thereby reducing the occurrence of mode aliasing. Then, by performing multiple empirical mode decompositions (EMD) on the radio frequency signal and calculating the mean, a stable intrinsic mode function (IMF) is obtained, and the heartbeat signal is extracted by filtering. Specifically, step S102, according to the motion state of the target user, restores the heartbeat waveform of the target user and extracts the heartbeat signal, including: combining the phase difference of the IQ data of the target user at each time to obtain a phase difference sequence signal; adding white noise to the phase difference sequence signal, and performing empirical mode decomposition to obtain an intrinsic mode function sequence; converting each intrinsic mode function sequence to the frequency domain, determining the main frequency of each intrinsic mode function sequence; determining the heartbeat frequency range of the target user based on the main frequency of the intrinsic mode function sequence, and extracting the heartbeat signal of the target user.

[0039] Exemplarily, first, each time phase difference in the IQ data is Combine to get the phase difference sequence signal Then, the initial white noise in the time domain Perform fast Fourier transform to get its spectrum .Will Converted into a spectrum with sine wave amplitude , the specific formula is:

[0040] in, represents the sampling rate. This formula ensures that the amplitude of the spectrum changes in the form of a sine wave, making the generated noise regular in frequency and helping to smooth the distribution of signal extreme values ​​during the decomposition process. Then, the adjusted spectrum Perform an inverse Fourier transform to obtain the noise in the time domain , and added to the phase difference sequence signal Then perform empirical mode decomposition on the signal with noise added to obtain the first m Secondary decomposition result: Intrinsic mode function Repeat the above steps. M Each time, different white noise is added to the decomposition. M = 1000. Calculate the ensemble mean of IMFs for each intrinsic mode function conduct M The average of the experimental results is used to obtain the final IMF:

[0041] This step ensures that each IMF can reflect the true frequency characteristics of the signal and avoid mode aliasing problems. Finally, perform FFT analysis on each IMF to find the IMF whose frequency mainly falls within the range of 0.8 Hz - 3.6 Hz (i.e., the heartbeat frequency range), and record this IMF as , which is the heartbeat signal.

[0042] On the basis of this embodiment, step S103, the heartbeat signal of the target user is segmented, including: setting a standard template for the heartbeat signal; iteratively updating the standard template through a dynamic programming method with the goal of minimizing the error between the signal segment of the heartbeat signal and the standard template until the iteration converges, thereby obtaining the segmentation result of the heartbeat signal and the target template.

[0043] Exemplarily, first, initialize the standard template and segmentation parameters. Set to zero vector, set iteration counter Then, update the segmentation result . Using dynamic programming method, according to the current template Heartbeat sequence Perform segmentation and obtain segmentation results . Update the template again , based on the latest segmentation results, recalculate the template Repeat the above steps until the segmentation result is S and templates u Convergence. When converged, the final segmentation result is returned S and templates u .

[0044] In some of the embodiments, step S104, extracting physiological features from the target user's segmented heartbeat signal, and classifying the target user based on the target user's physiological features to obtain a classification result, includes: extracting physiological features from the segmented heartbeat signal; calling a pre-trained classification model, classifying the target user according to the extracted physiological features, and obtaining a classification result.

[0045] In this embodiment, physiological features mainly include heart rate variability (HRV), heart rate cycle, etc. These features will be used as input for the classification model. Based on the extracted labor-saving features, the user's myocardial infarction risk is assessed through the disease classification subsystem. The classification module uses the trained model to map the input features to the healthy or myocardial infarction state, and finally outputs the test results. If the test results show the risk of myocardial infarction, further medical examinations will be prompted.

[0046] Exemplarily, first, heart rate variability features are calculated, which perform well in distinguishing healthy people from MI patients, including: Time domain features: SDNN: Standard deviation of normal heartbeat intervals, indicating the overall variation of heartbeat intervals.

[0047] RMSSD: The root mean square of the difference between adjacent heartbeats, used to measure the fluctuation of the heartbeat interval.

[0048] NN50 and pNN50: respectively represent the number of times the interval between adjacent heartbeats exceeds 50 milliseconds and its proportion.

[0049] Frequency domain features: These include low frequency (LF) and high frequency (HF) energy, as well as the LF / HF ratio, which are calculated through power spectrum estimation.

[0050] Due to the small amount of sample data, if all features are used directly, it may lead to overfitting. Therefore, l1-SVM is used as the classification model to automatically select the features with the highest correlation with MI, remove redundant features, and optimize model performance. The selected features and their importance levels are as follows: SDNN: 26:4%; RMSSD: 15:3%; NN50: 4:7%; PNN50: 2:9%. Finally, the classification model is used to detect myocardial infarction and obtain the detection results of the target user.

[0051] In summary, this method is different from traditional electrocardiogram (ECG) or blood tests, and this method achieves completely contactless detection. This means that users do not need to wear sensors, connect wires, or perform invasive operations such as blood tests, thus avoiding physiological discomfort for users. Since there is no need for direct contact with the human body, the system can monitor heart rate at any time in daily environments and is particularly suitable for home scenarios. Users can continuously and long-term monitor heart health at home and detect early signs of MI, thereby greatly increasing the possibility of disease prevention.

[0052] Moreover, traditional radio frequency monitoring systems require users to remain still in order to accurately detect heartbeat signals. This method innovatively designs a signal processing algorithm that can tolerate slight physical activities. These activities include reading, eating, surfing the Internet, etc., during which users may slightly lean forward or backward, or move their limbs slightly. This method not only enhances the stability of the signal, but also significantly improves the detection accuracy of the system under slight movement conditions, making it closer to real-life scenarios. For example, when a user is watching TV or working, the system can still monitor their heart rate changes in real time. This adaptability provides great convenience for family and personal health monitoring.

[0053] In addition, this method uses commercially available pulse coherent radar technology. This radar equipment is low-cost and does not require customized hardware support, which makes the system have significant cost advantages. The economy and ease of use of this system make it affordable for ordinary home users, and it does not require complex equipment installation or debugging, making it very suitable for use in home scenarios. Users only need to install the device in the room, and the system can continuously monitor the user's heart health, providing a low-cost solution for the early detection of myocardial infarction.

[0054] The present invention also provides a non-contact myocardial infarction detection device. The non-contact myocardial infarction detection device provided by the present invention is described below. The non-contact myocardial infarction detection device described below and the non-contact myocardial infarction detection method described above can be referenced to each other. Figure 2 is a structural block diagram of the non-contact myocardial infarction detection device provided by the present invention, such as Figure 2 As shown, the device comprises: The acquisition module 201 is used to acquire the radio frequency signal fed back by the target user, and determine the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; An extraction module 202 is used to restore the heartbeat waveform of the target user according to the motion state of the target user and extract the heartbeat signal; The segmentation module 203 is used to segment the heartbeat signal of the target user; The detection module 204 is used to extract physiological features from the target user's segmented heartbeat signal, and classify the target user based on the target user's physiological features to obtain a classification result.

[0055] When the device is in use, first, the acquisition module 201 acquires the radio frequency signal fed back by the target user, and then determines whether the target user is in a static state or a slight body movement state according to the radio frequency signal. Among them, the slight body movement state is a slight physical activity that can be tolerated during the detection process, including reading, eating, surfing the Internet, etc., and the target user will have a slight forward leaning, backward leaning or small movements of the limbs in these activities. Then, the extraction module 202 uses corresponding means to restore the heartbeat waveform of the target user according to the movement state of the target user, and extracts the heartbeat signal. On this basis, the segmentation module 203 then segments the extracted heartbeat signal to extract the specific information of each heartbeat cycle. Finally, the detection module 204 extracts the physiological characteristics of the target user from the segmented heartbeat signal, and classifies the target user based on the physiological characteristics to determine whether the target user is a myocardial infarction population. In the existing detection technology, only when the target user is allowed to remain static can a higher accuracy rate be achieved. In the above process, the RF signal is processed differently according to the target user's motion state, which not only enhances the stability of the RF signal, but also significantly improves the detection accuracy under mild motion conditions, making it closer to real life scenarios. For example, when watching TV or working, users can still monitor their heart rate changes in real time. This adaptability provides great convenience for family and personal health monitoring and solves the problem of poor accuracy in myocardial infarction detection in existing related technologies.

[0056] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3 As shown, the electronic device may include: a processor 301, a communications interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communications interface 302 and the memory 303 communicate with each other via the communication bus 304. The processor 301 may call the logic instructions in the memory 303 to execute the non-contact myocardial infarction detection method, which includes: Collect the radio frequency signal fed back by the target user, and determine the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; According to the target user's motion state, restore the target user's heartbeat waveform and extract the heartbeat signal; Segment the heartbeat signal of the target user; The physiological features are extracted from the target user's segmented heartbeat signal, and the target user is classified based on the target user's physiological features to obtain a classification result.

[0057] In addition, the logic instructions in the above-mentioned memory 303 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0058] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the non-contact myocardial infarction detection method provided by the above methods, the method includes: Collect the radio frequency signal fed back by the target user, and determine the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; According to the target user's motion state, restore the target user's heartbeat waveform and extract the heartbeat signal; Segment the heartbeat signal of the target user; The physiological features are extracted from the target user's segmented heartbeat signal, and the target user is classified based on the target user's physiological features to obtain a classification result.

[0059] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform the non-contact myocardial infarction detection method provided by the above methods, the method comprising: Collect the radio frequency signal fed back by the target user, and determine the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; According to the target user's motion state, restore the target user's heartbeat waveform and extract the heartbeat signal; Segment the heartbeat signal of the target user; The physiological features are extracted from the target user's segmented heartbeat signal, and the target user is classified based on the target user's physiological features to obtain a classification result.

[0060] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0061] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A non-contact myocardial infarction detection method, characterized in that: include: Collecting the radio frequency signal fed back by the target user, and determining the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; According to the motion state of the target user, restore the heartbeat waveform of the target user and extract the heartbeat signal; Segmenting the heartbeat signal of the target user; Extracting physiological features from the target user's segmented heartbeat signal, and classifying the target user based on the target user's physiological features to obtain a classification result.

2. The non-contact myocardial infarction detection method according to claim 1, characterized in that: Collecting the radio frequency signal fed back by the target user, and determining the motion state of the target user based on the radio frequency signal, including: Collecting and transmitting a radio frequency signal modulated by the breathing and heartbeat of the target user through a pulse coherent radar; the radio frequency signal includes a plurality of continuous IQ data; determining the amplitude of motion between two pulses based on a phase difference between two consecutive IQ data; The amplitude variance of the motion amplitude is compared with a preset detection threshold to determine the motion state of the target user.

3. The non-contact myocardial infarction detection method according to claim 2, characterized in that: When the target user is in a stationary state, the heartbeat waveform of the target user is restored according to the motion state of the target user, and the heartbeat signal is extracted, including: Combining the phase differences of the IQ data of the target user at each time to obtain a phase difference sequence signal; Performing a fast Fourier transform on the phase difference sequence signal, and constructing a bandpass filter to filter the radio frequency signal; The filtered radio frequency signal is subjected to inverse Fourier transform to obtain a heartbeat waveform of the target user, and a heartbeat signal is extracted based on the heartbeat waveform.

4. The non-contact myocardial infarction detection method according to claim 2, characterized in that: When the target user is in a slight body motion state, restoring the heartbeat waveform of the target user according to the motion state of the target user and extracting the heartbeat signal includes: Combining the phase differences of the IQ data of the target user at each time to obtain a phase difference sequence signal; Adding white noise to the phase difference sequence signal, and performing empirical mode decomposition to obtain an intrinsic mode function sequence; Convert each of the intrinsic mode function sequences to the frequency domain to determine the main frequency of each of the intrinsic mode function sequences; The heartbeat frequency range of the target user is determined based on the main frequency of the intrinsic mode function sequence, and the heartbeat signal of the target user is extracted.

5. The non-contact myocardial infarction detection method according to claim 3, characterized in that: When the target user is in a stationary state, segmenting the heartbeat signal of the target user includes: Reversely process the phase difference sequence signal, traverse the phase difference sequence signal, determine the peak function of each point, and generate a peak set; The peak value set is traversed in order of index, points in the peak value set are sorted and filtered, and divided into multiple segments to form a heartbeat signal segment set.

6. The non-contact myocardial infarction detection method according to claim 4, characterized in that: When the target user is in a stationary state, segmenting the heartbeat signal of the target user includes: Set the standard template for heartbeat signal; By means of a dynamic programming method, the standard template is iteratively updated with the goal of minimizing the error between the signal segment of the heartbeat signal and the standard template until the iteration converges, thereby obtaining the segmentation result of the heartbeat signal and the target template.

7. The non-contact myocardial infarction detection method according to claim 1, characterized in that: Extracting physiological features from the heartbeat signal of the target user after segmentation processing, and classifying the target user based on the physiological features of the target user to obtain a classification result, including: Extracting physiological features from the segmented heartbeat signal; The pre-trained classification model is called to classify the target user according to the extracted physiological features to obtain a classification result.

8. A non-contact myocardial infarction detection device, characterized in that: include: A collection module, used for collecting the radio frequency signal fed back by the target user, and determining the motion state of the target user based on the radio frequency signal; the motion state includes a static state and a slight body movement state; An extraction module, used to restore the heartbeat waveform of the target user and extract the heartbeat signal according to the motion state of the target user; A segmentation module, used for segmenting the heartbeat signal of the target user; The detection module is used to extract physiological features from the heartbeat signal of the target user after segmentation processing, and classify the target user based on the physiological features of the target user to obtain a classification result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the non-contact myocardial infarction detection method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the non-contact myocardial infarction detection method according to any one of claims 1 to 7 is implemented.

Images