Intelligent identification system of defibrillator

Abstract

The invention belongs to the technical field of medical equipment, and provides an intelligent recognition system of a defibrillator, which comprises an electrocardiogram acquisition module, a data analysis and processing module, an alarm module, a defibrillation treatment module and a user interaction interface, according to the system, through an automatic process, from collection and analysis of electrocardiogram signals to alarm and defibrillation treatment, response time is greatly shortened, precious treatment time is gained for patients, meanwhile, recognition is conducted by means of a built-in intelligent algorithm, misjudgment caused by human factors is reduced, and diagnosis accuracy and efficiency are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of medical equipment, in particular to an intelligent identification system for a defibrillator. Background Art

[0002] In clinical emergency treatment, the defibrillator is an important medical device used to treat arrhythmias, especially malignant arrhythmias such as ventricular fibrillation.

[0003] However, the operation of existing defibrillators relies on the experience and judgment of medical staff, which not only increases the risk of misdiagnosis but also delays the best time for treatment. At the same time, when there is no one on duty or the medical staff lacks experience, malignant arrhythmias may not be discovered and treated in time, seriously threatening the patient's life safety.

[0004] Therefore, those skilled in the art have proposed an intelligent identification system for a defibrillator to solve the problems raised by the background technology. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an intelligent identification system for a defibrillator to solve the problem that the operation of the defibrillator in the prior art depends on the experience and judgment of medical staff, which not only increases the risk of misjudgment but also delays the best time for treatment; at the same time, in the case of unattended operation or insufficient experience of medical staff, malignant arrhythmias may not be discovered and treated in time, seriously threatening the life safety of patients.

[0006] An intelligent identification system for a defibrillator, comprising:

[0007] An electrocardiogram acquisition module, used for acquiring the electrocardiogram signal of the patient in real time; the number of the electrocardiogram acquisition module is at least one;

[0008] The data analysis and processing module can automatically identify and analyze electrocardiogram signals to determine whether there is malignant arrhythmia;

[0009] Alarm module, when the data analysis and processing module identifies malignant arrhythmia, it can immediately send out sound and light alarm signals;

[0010] The defibrillation treatment module is connected to the data analysis and processing module, and automatically or manually starts the defibrillation treatment program after confirming the malignant arrhythmia;

[0011] The user interaction interface is used to display the ECG waveform, analysis results and alarm information, and allows the user to input operation instructions. It provides intuitive operation guidance and alarm information display through the user interaction interface, which reduces the difficulty of operating the equipment and enables ordinary medical staff and even non-professionals to quickly get started.

[0012] Preferably, the electrocardiogram acquisition module includes an electrode sheet and a signal acquisition circuit, the electrode sheet is used to be attached to the patient's body surface to collect electrocardiogram signals, the signal acquisition circuit is used to process and transmit the collected signals, and the signal processing includes normalizing the signals, and the normalization processing formula is as follows:

[0013]

[0014] in, is the normalized output signal, x is the total output signal, E(x) is the mean output result, Var(x) is the variance, ε is the minimum value, γ and β are constant factors.

[0015] Preferably, the data analysis and processing module includes:

[0016] The preprocessing unit is used to filter and denoise the collected electrocardiogram signal; the steps of the denoising process are as follows:

[0017] S11. First, construct the electrocardiogram signal into an H matrix as follows:

[0018]

[0019] Where N is the number of ECG signals, and L is the window length;

[0020] S12, then perform singular value processing on the H matrix, as follows:

[0021]

[0022] Among them, P is the left singular matrix, Q is the right singular matrix, and V is the diagonal matrix of singular values;

[0023] S13. Finally, the singular values ​​are processed and the signal is reconstructed. After the corresponding singular values ​​are processed, a new diagonal matrix V′ can be obtained. Let the reconstructed H matrix be H′, and the following formula is obtained:

[0024] H′=PV′Q T ;

[0025] Perform anti-diagonal averaging on the reconstructed matrix H′ to achieve denoising of the ECG signal;

[0026] The feature extraction unit is used to extract key feature parameters from the preprocessed electrocardiogram signal; the extracted key feature parameters are processed using feature extraction technology, and the formula of the feature extraction technology is as follows:

[0027]

[0028] Among them, h represents the height of the feature map, l represents the width of the feature map, r represents the convolution kernel size of the feature map, and x p Indicates the number of regular convolution channels;

[0029] An intelligent recognition unit, based on a deep learning algorithm, automatically identifies the type of malignant arrhythmia according to characteristic parameters; the deep learning algorithm includes automatic learning and recognition of extracted characteristic parameters; the deep learning algorithm can automatically learn feature representations from data and perform classification or recognition tasks based on these features;

[0030] A decision unit, which determines whether to trigger the alarm module according to the output result of the intelligent recognition unit, and the decision unit uses a K filter algorithm to predict the output result, and the K filter algorithm includes a prediction period and an update period;

[0031] The prediction period includes observation value prediction and covariance prediction, and can predict the state at the current moment through the optimal estimate at the previous moment, as shown in formula (1) and formula (2):

[0032]

[0033] Q t * =HQ t-1 A T +S (2)

[0034] in, represents the optimal estimate of the state at time t-1, represents the prior estimate of the state at time t, that is, the state prediction value at t; H is the state transfer matrix; M is the control input matrix; u t-1 is the input at time t-1; Q t-1 is the covariance matrix at time t-1, Q t * represents the covariance matrix of the prior estimate at time t; S is the covariance matrix of the process noise;

[0035] The update period includes K gain calculation, current optimal estimate calculation and error covariance update, as shown in equations (3) to (5):

[0036] k t =Q t * L T (LQ t * L T +R) -1 (3)

[0037]

[0038] Q t=(IK t L)Q t * (5)

[0039] Among them, K t represents K gain; L represents the transfer matrix from state variables to observations; R represents the covariance of measurement noise; z t represents the observation equation; I represents the identity matrix; Q t is the covariance matrix at the current moment;

[0040] By using K filtering algorithm to track and estimate it in real time, the real-time and accuracy of structural damage warning can be improved;

[0041] At the same time, a hybrid kernel function K(m,n) with superior performance is introduced. The hybrid kernel function K(m,n) has good learning and interpolation capabilities, and its generalization and prediction capabilities are relatively high. Its performance has been greatly improved, and its accuracy and reliability for alarm prediction have been improved. The formula of the kernel function K is as follows:

[0042]

[0043] Among them, ρ represents the influence coefficient of the mixed kernel function, and d represents the order of the kernel function.

[0044] Preferably, the alarm module comprises:

[0045] An audible and visual alarm device that emits a high-decibel sound and flashes a strong light when a malignant arrhythmia is identified;

[0046] The wireless communication module can send alarm information to the remote monitoring center or the mobile device of designated medical staff.

[0047] A processor is configured to execute the above-mentioned intelligent identification system of the defibrillator.

[0048] A computer-readable storage medium stores a computer program, wherein the computer program is executed by a processor to generate the above-mentioned intelligent identification system for a defibrillator.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. Through the built-in intelligent algorithm, the present invention can automatically and accurately identify malignant arrhythmias and reduce the possibility of human misjudgment.

[0051] 2. After the malignant arrhythmia is identified, the present invention can immediately send out an audible and visual alarm signal, and send the alarm information to a remote monitoring center or a mobile device of designated medical staff through a wireless communication module, thereby shortening the response time and buying precious treatment time for the patient.

[0052] 3. The intelligent recognition function of the present invention enables the defibrillator to intervene in the early stages of malignant arrhythmias, and cooperate with timely defibrillation treatment to significantly improve the patient's success rate of treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The figure is a framework diagram of the intelligent identification system of the defibrillator of the present invention. DETAILED DESCRIPTION

[0054] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0055] like Figure 1 As shown:

[0056] Embodiment 1: The present invention provides an intelligent identification system for a defibrillator, comprising an electrocardiogram acquisition module, a data analysis and processing module, an alarm module, a defibrillation treatment module and a user interaction interface, wherein the electrocardiogram acquisition module, the data analysis and processing module, the alarm module, the defibrillation treatment module and the user interaction interface are electrically connected in sequence:

[0057] The electrocardiogram acquisition module is used to collect the electrocardiogram signal of the patient in real time; the electrocardiogram acquisition module includes an electrode sheet and a signal acquisition circuit, the electrode sheet is used to be attached to the patient's body surface to collect the electrocardiogram signal, and the signal acquisition circuit is used to process and transmit the collected signal; the signal processing includes normalizing the signal, and the normalization processing formula is as follows:

[0058]

[0059] in, is the normalized output signal, x is the total output signal, E(x) is the mean output result, Var(x) is the variance, ε is the minimum value, γ and β are constant factors;

[0060] The data analysis and processing module can automatically identify and analyze the electrocardiogram signal to determine whether there is a malignant arrhythmia; the data analysis and processing module includes:

[0061] The preprocessing unit is used to filter and denoise the collected electrocardiogram signal; the steps of the denoising process are as follows:

[0062] S11. First, construct the electrocardiogram signal into an H matrix as follows:

[0063]

[0064] Where N is the number of ECG signals, and L is the window length;

[0065] S12, then perform singular value processing on the H matrix, as follows:

[0066]

[0067] Among them, P is the left singular matrix, Q is the right singular matrix, and V is the diagonal matrix of singular values;

[0068] S13. Finally, the singular values ​​are processed and the signal is reconstructed. After the corresponding singular values ​​are processed, a new diagonal matrix V′ can be obtained. Let the reconstructed H matrix be H′, and the following formula is obtained:

[0069] H′=PV′Q T ;

[0070] The reconstruction matrix H′ is processed by anti-diagonal averaging to achieve denoising of the ECG signal; the denoising algorithm in the preprocessing unit effectively removes the noise in the ECG signal through singular value decomposition (SVD) and reconstruction, improves the signal quality, and makes subsequent analysis more accurate;

[0071] The feature extraction unit is used to extract key feature parameters from the preprocessed electrocardiogram signal; the extracted key feature parameters are processed using feature extraction technology, and the formula of the feature extraction technology is as follows:

[0072]

[0073] Among them, h represents the height of the feature map, l represents the width of the feature map, r represents the convolution kernel size of the feature map, and x p Indicates the number of conventional convolution channels; the feature extraction technology used by the feature extraction unit can accurately extract key feature parameters from the preprocessed electrocardiogram signal;

[0074] An intelligent recognition unit, based on a deep learning algorithm, automatically identifies the type of malignant arrhythmia according to characteristic parameters; the deep learning algorithm includes automatic learning and recognition of extracted characteristic parameters; the deep learning algorithm can automatically learn feature representations from data and perform classification or recognition tasks based on these features, so that the recognition results are more accurate and reliable;

[0075] A decision unit, which determines whether to trigger the alarm module according to the output result of the intelligent recognition unit, and the decision unit uses a K filter algorithm to predict the output result, and the K filter algorithm includes a prediction period and an update period;

[0076] The prediction period includes observation value prediction and covariance prediction, and can predict the state at the current moment through the optimal estimate at the previous moment, as shown in formula (1) and formula (2):

[0077]

[0078] Q t * =HQ t-1 A T +S (2)

[0079] in, represents the optimal estimate of the state at time t-1, represents the prior estimate of the state at time t, that is, the state prediction value at t; H is the state transfer matrix; M is the control input matrix; u t-1 is the input at time t-1; Q t-1 is the covariance matrix at time t-1, Q t * represents the covariance matrix of the prior estimate at time t; S is the covariance matrix of the process noise;

[0080] The update period includes K gain calculation, current optimal estimate calculation and error covariance update, as shown in equations (3) to (5):

[0081] k t =Q t * L T (LQ t * L T +R) -1 (3)

[0082]

[0083] Q t =(IK t L)Q t * (5)

[0084] Among them, K t represents K gain; L represents the transfer matrix from state variables to observations; R represents the covariance of measurement noise; z t represents the observation equation; I represents the identity matrix; Q t is the covariance matrix at the current moment;

[0085] The decision-making unit uses the K-filter algorithm to predict the output results, including the prediction period and the update period, to achieve real-time tracking and estimation of malignant arrhythmias, improve the real-time and accuracy of structural damage warning, and ensure timely response;

[0086] At the same time, a hybrid kernel function K(m,n) with superior performance is introduced. The hybrid kernel function K(m,n) has good learning and interpolation capabilities, and its generalization and prediction capabilities are relatively high. Its performance has been greatly improved, and its accuracy and reliability for alarm prediction have been improved. The formula of the kernel function K is as follows:

[0087]

[0088] Among them, ρ represents the influence coefficient of the hybrid kernel function, and d represents the order of the kernel function. By introducing the hybrid kernel function K(m,n), the learning, interpolation, generalization and prediction capabilities of the algorithm are improved, so that the accuracy and reliability of alarm prediction are significantly improved.

[0089] The alarm module can immediately send out an audible and visual alarm signal when the data analysis and processing module identifies a malignant arrhythmia; the alarm module includes:

[0090] An audible and visual alarm device that emits a high-decibel sound and flashes a strong light when a malignant arrhythmia is identified;

[0091] Wireless communication module, capable of sending alarm information to a remote monitoring center or a designated medical staff’s mobile device;

[0092] The defibrillation treatment module is connected to the data analysis and processing module, and automatically or manually starts the defibrillation treatment program after confirming the malignant arrhythmia;

[0093] The user interaction interface is used to display the ECG waveform, analysis results and alarm information, and allows the user to input operation instructions.

[0094] From the above, we can see that the system greatly shortens the response time through automated processes, from the collection and analysis of ECG signals to alarm and defibrillation treatment, thus buying precious treatment time for patients. At the same time, it relies on built-in intelligent algorithms for identification, reduces misjudgments caused by human factors, and improves the accuracy and efficiency of diagnosis.

[0095] The performance indicators of the existing method (i.e., the identification of malignant arrhythmia by the defibrillator in the prior art) and the intelligent identification system of the defibrillator provided in this embodiment are compared, and the comparison results are shown in Table 1:

[0096] Table 1- Performance index comparison results

[0097] Existing methods This method Signal recognition efficiency 68.7％ 93.8％ Identify real-time Unable to identify in real time Automatic real-time recognition Human intervention required Reliance on manual inspection No manual work required False positive rate 32.8％ 4.3％

[0098] It can be seen from Table 1 that, compared with the existing methods, the intelligent recognition system for a defibrillator provided in this embodiment has better effects in signal recognition efficiency, recognition real-time performance, manual intervention requirements, and false positive and missed detection rates.

[0099] The embodiment of the present application provides an electronic device, which is applicable to the above-mentioned intelligent identification system of the defibrillator, including:

[0100] Memory, used to protect computer programs and data;

[0101] Processor, used to run system programs.

[0102] The embodiment of the present application provides a computer storage medium, which is applicable to the above-mentioned intelligent identification system of the defibrillator, and performs hierarchical confidentiality management on the above-mentioned system and data in accordance with confidentiality management requirements.

[0103] Those skilled in the art will appreciate that the embodiments of the present application may be provided as a system or a computer program product. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0104] The present application is described with reference to the flowcharts and / or block diagrams of the devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0105] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1The steps for the functions specified in one or more boxes.

[0107] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0108] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0109] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include transitory media such as modulated data signals and carrier waves.

[0110] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, commodity or device including the elements.

[0111] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. An intelligent identification system for a defibrillator, characterized in that: include: Electrocardiogram acquisition module, used to collect the patient's electrocardiogram signal in real time; The data analysis and processing module can automatically identify and analyze electrocardiogram signals to determine whether there is malignant arrhythmia; Alarm module, when the data analysis and processing module identifies malignant arrhythmia, it can immediately send out sound and light alarm signals; The defibrillation treatment module is connected to the data analysis and processing module, and automatically or manually starts the defibrillation treatment program after confirming the malignant arrhythmia; The user interaction interface is used to display the ECG waveform, analysis results and alarm information, and allows the user to input operation instructions.

2. The intelligent identification system for a defibrillator as claimed in claim 1, characterized in that: The electrocardiogram acquisition module includes an electrode sheet and a signal acquisition circuit. The electrode sheet is used to be attached to the patient's body surface to collect electrocardiogram signals, and the signal acquisition circuit is used to process and transmit the collected signals.

3. The intelligent identification system for a defibrillator as claimed in claim 1, characterized in that: The data analysis and processing module includes: The preprocessing unit is used to filter and denoise the collected electrocardiogram signal; the steps of the denoising process are as follows: S11. First, construct the electrocardiogram signal into an H matrix as follows: Where N is the number of ECG signals, and L is the window length; S12, then perform singular value processing on the H matrix, as follows: Among them, P is the left singular matrix, Q is the right singular matrix, and V is the diagonal matrix of singular values; S13, finally, the singular values ​​are processed and the signal is reconstructed; that is, after the corresponding singular values ​​are processed, a new diagonal matrix V′ can be obtained, and the reconstructed H matrix is ​​set to H′, and the following formula is obtained: H′=PV′Q T ; Perform anti-diagonal averaging on the reconstructed matrix H′ to achieve denoising of the ECG signal; The feature extraction unit is used to extract key feature parameters from the preprocessed electrocardiogram signal; the extracted key feature parameters are processed using feature extraction technology, and the formula of the feature extraction technology is as follows: Among them, h represents the height of the feature map, l represents the width of the feature map, r represents the convolution kernel size of the feature map, and x p Indicates the number of regular convolution channels; Intelligent recognition unit, based on deep learning algorithm, automatically identifies the type of malignant arrhythmia according to characteristic parameters; A decision unit, which determines whether to trigger the alarm module according to the output result of the intelligent recognition unit, and the decision unit uses a K filter algorithm to predict the output result, and the K filter algorithm includes a prediction period and an update period; The prediction period includes observation value prediction and covariance prediction, and can predict the state at the current moment through the optimal estimate at the previous moment, as shown in formula (1) and formula (2): Q t * =HQ t-1 TO T +S (2) in, represents the optimal estimate of the state at time t-1, represents the prior estimate of the state at time t, that is, the state prediction value at t; H is the state transfer matrix; M is the control input matrix; u t-1 is the input at time t-1; Q t-1 is the covariance matrix at time t-1, Q t * represents the covariance matrix of the prior estimate at time t; S is the covariance matrix of the process noise; The update period includes K gain calculation, current optimal estimate calculation and error covariance update, as shown in equations (3) to (5): k t =Q t * L T (LQ t * L T +R) -1 (3) Q t =(I-K t L)Q t * (5) Among them, K t represents K gain; L represents the transfer matrix from state variables to observations; R represents the covariance of measurement noise; z t represents the observation equation; I represents the identity matrix; Q t is the covariance matrix at the current moment; At the same time, a hybrid kernel function K(m,n) with superior performance is introduced. The hybrid kernel function K(m,n) has good learning and interpolation capabilities, and its generalization and prediction capabilities are relatively high. Its performance has been greatly improved, and its accuracy and reliability for alarm prediction have been improved. The formula of the kernel function K is as follows: Among them, ρ represents the influence coefficient of the mixed kernel function, and d represents the order of the kernel function.

4. The intelligent identification system for a defibrillator as claimed in claim 1, characterized in that: The alarm module comprises: An audible and visual alarm device that emits a high-decibel sound and flashes a strong light when a malignant arrhythmia is identified; The wireless communication module can send alarm information to the remote monitoring center or the mobile device of designated medical staff.

5. A processor, characterized in that: An intelligent identification system configured to implement a defibrillator according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the intelligent identification system of the defibrillator according to any one of claims 1 to 4 is implemented.

Images