High-precision electrocardio physiological signal analysis method

Abstract

The invention discloses a high-precision electrocardio physiological signal analysis method, and belongs to the technical field of electrocardio physiological signals, and the method comprises the following steps: S1, collecting and processing an electrocardio physiological real-time signal of a patient, and generating an electrocardio physiological feature signal of the patient; s2, carrying out analysis and pattern recognition on the electrocardiogram physiological characteristic signal of the patient, and determining a high-precision electrocardiogram physiological signal analysis and recognition result; and S3, carrying out timely early warning on heart abnormal events of the patient, generating a high-precision electrocardio physiological signal analysis report, and formulating a treatment scheme for the patient. The problems that the electrocardio physiological signal analysis and recognition accuracy and reliability are low and the abnormal condition of the heart cannot be accurately recognized due to the fact that the electrocardio physiological signal cannot be effectively analyzed in the prior art are solved. The electrocardio physiological signals can be effectively analyzed, the electrocardio physiological signal analysis and recognition accuracy and reliability can be improved, the abnormal condition of the heart can be accurately recognized, and the diagnosis and prognosis accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrocardiophysiological signals, and in particular to a high-precision electrocardiophysiological signal analysis method. Background Art

[0002] The main function of the heart is to provide power for blood flow and to transport blood to all parts of the body. Among them, the electrocardiogram (ECG) signal is an electrical signal generated by cardiac activity, which has a high temporal and spatial resolution and can be used to monitor the diagnosis and treatment of heart disease.

[0003] The Chinese patent with publication number CN2899000Y discloses a multifunctional electrocardiophysiological experimental teaching instrument, which includes a host and a sensor element. The sensor element includes an electrocardiograph electrode, an arterial wave sensor, a venous wave sensor, and a heart sound sensor. Each sensor element is connected to the host through an amplifier; the host is connected with an input device, an output device, and a display device. By synchronously recording a variety of physiological signals of the human body and combining the recorded signals with multimedia teaching, the teaching and research methods are enriched, the students' learning interest is improved, and the students' perceptual understanding is enhanced. The device is suitable for use in experimental teaching of relevant courses in medical schools; however, the patent has the following defects:

[0004] Existing technologies cannot effectively analyze electrocardiographic physiological signals, resulting in low accuracy and reliability in electrocardiographic physiological signal analysis and recognition, and cannot accurately identify cardiac abnormalities, resulting in low accuracy in diagnosis and prognosis. Summary of the invention

[0005] The purpose of the present invention is to provide a high-precision electrophysiological signal analysis method, which can effectively analyze electrophysiological signals, improve the accuracy and reliability of electrophysiological signal analysis and recognition, accurately identify cardiac abnormalities, improve the accuracy of diagnosis and prognosis, and solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A high-precision electrocardiographic signal analysis method comprises the following steps:

[0008] S1. Signal acquisition and processing: Collect the patient's real-time electrophysiological signals based on the smart sensor integrated on the smart wearable device, and process the patient's real-time electrophysiological signals based on the smart sensor to generate the patient's electrophysiological characteristic signals based on the smart sensor;

[0009] S2. Signal analysis and recognition: Train a high-precision electrophysiological signal analysis and recognition model based on deep learning, analyze and pattern recognize the electrophysiological characteristic signals of patients based on smart sensors according to the high-precision electrophysiological signal analysis and recognition model based on deep learning, and determine the high-precision electrophysiological signal analysis and recognition results;

[0010] S3. Visual management of results: Provide timely warnings for abnormal cardiac events in patients and generate high-precision electrophysiological signal analysis reports, which are displayed in a visual form for subsequent follow-up observations and research, and develop treatment plans for patients to adapt to abnormal changes in their heart conditions.

[0011] Preferably, in S1, collecting the patient's real-time electrophysiological signals includes:

[0012] Based on the smart sensors integrated on the smart wearable device, the patient's ECG signal, intracardiac electrogram, heart sound signal and myoelectric signal are monitored and continuously collected in real time in a quiet state to obtain the patient's real-time ECG signal based on the smart sensor;

[0013] Based on the smart sensors integrated on the smart wearable devices, the patient's blood pressure signal, respiratory signal, temperature signal and blood oxygen saturation signal are monitored and collected continuously in real time in a quiet state to obtain the patient's real-time physiological signals based on the smart sensors;

[0014] Wherein, the patient's real-time electrocardiographic and physiological signals based on the smart sensor are determined according to the patient's real-time electrocardiographic signals based on the smart sensor and the patient's real-time physiological signals based on the smart sensor.

[0015] Preferably, in S1, processing the patient's real-time electrophysiological signal based on the intelligent sensor includes:

[0016] Acquire the patient's real-time electrophysiological signal based on the intelligent sensor, and perform filtering processing on the patient's real-time electrophysiological signal based on the intelligent sensor based on a median filter;

[0017] Wherein, according to the characteristics of the patient's real-time electrophysiological signal based on the intelligent sensor, the median filter window is determined, and when the size of the median filter window is determined, the size of the median filter window is smaller than the width of the P wave, the QRS complex wave and the T wave in the patient's real-time electrophysiological signal;

[0018] Based on the median filtering algorithm, the median of all data points in the median filter window is calculated to replace the data point in the center of the window, thereby eliminating isolated noise points and removing noise from the real-time electrophysiological signal of the patient based on the smart sensor, while retaining the edges and details of the real-time electrophysiological signal of the patient based on the smart sensor.

[0019] Preferably, determining the median filter window size comprises:

[0020] Retrieve the width of P wave, QRS complex wave and T wave in the patient's real-time electrophysiological signal;

[0021] Retrieving the maximum width from the widths of the P wave, QRS complex wave and T wave in the real-time electrophysiological signal of the patient, and using the time length corresponding to the maximum width as the initial window;

[0022] Extracting the waveform widths of the P wave, QRS complex wave and T wave;

[0023] Obtaining a first window setting coefficient using the waveform widths of the P wave, QRS complex wave, and T wave;

[0024] The first window setting coefficient is obtained by the following formula:

[0025]

[0026] Among them, S 01 Indicates the first window setting coefficient; K p , K q and K t Respectively represent the time length corresponding to the waveform width of P wave, QRS complex wave and T wave; K c Indicates the preset time length reference value;

[0027] comparing the first window setting coefficient with a preset window setting coefficient threshold;

[0028] When the first window setting coefficient is lower than a preset window setting coefficient threshold, the median filter window is set according to a preset initial window size;

[0029] When the first window setting coefficient exceeds a preset window setting coefficient threshold, the median filter window is set using the signal dynamic range of the P wave, QRS complex wave and T wave in combination with the first window setting coefficient.

[0030] Preferably, when the first window setting coefficient exceeds a preset window setting coefficient threshold, the median filter window is set using the signal dynamic range of the P wave, QRS complex wave and T wave in combination with the first window setting coefficient, including:

[0031] When the first window setting coefficient exceeds a preset window setting coefficient threshold, retrieving the signal dynamic range of the P wave, QRS complex wave and T wave;

[0032] Obtaining a second window setting coefficient using the signal dynamic range of the P wave, QRS complex wave and T wave;

[0033] The second window setting coefficient is obtained by the following formula:

[0034]

[0035] Among them, S 02 Indicates the second window setting coefficient; D p , D q and D t Represents the signal dynamic range of P wave, QRS complex wave and T wave respectively; D c Indicates a preset signal dynamic range reference value;

[0036] The median filter window is set using the first window setting coefficient and the second window setting coefficient, wherein the size of the median filter window after the setting is obtained by the following formula:

[0037]

[0038] Among them, K h Indicates the size of the median filter window after setting; K s Indicates the initial window size; S 01 Indicates the first window setting coefficient; S 02 Indicates the second window setting coefficient; S y Indicates the preset window setting coefficient threshold.

[0039] Preferably, in S1, processing the patient's real-time electrophysiological signal based on the intelligent sensor also includes:

[0040] Acquire the real-time electrophysiological signal of the patient based on the intelligent sensor after filtering, and transform the real-time electrophysiological signal of the patient based on the intelligent sensor based on wavelet transform;

[0041] Among them, according to the characteristics of the patient's real-time electrophysiological signal based on the intelligent sensor, a wavelet basis with compact support and symmetry is selected;

[0042] According to the frequency range of the patient's real-time electrophysiological signal based on the smart sensor and the required decomposition details, the number of decomposition layers is determined, and the patient's real-time electrophysiological signal based on the smart sensor is decomposed into approximate coefficients and detail coefficients of different scales using a wavelet basis with compact support and symmetry;

[0043] Useful characteristic parameters are extracted from the wavelet decomposition coefficients and fused to generate the patient's electrophysiological characteristic signals based on intelligent sensors.

[0044] Preferably, in S2, training a high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning includes:

[0045] Collect the patient's historical electrophysiological signals according to the needs of high-precision electrophysiological signal analysis;

[0046] Classifying the collected electrocardiographic physiological history signals of the patients, and dividing the collected electrocardiographic physiological history signals of the patients into a training set and a test set;

[0047] Based on deep learning technology, a training set is used to train the deep learning model, so that the deep learning model can autonomously learn and analyze the high-precision electrocardiographic physiological signal analysis process and patient cardiac pattern recognition, and determine a high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning;

[0048] Use the test set to perform a performance test on the high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning to determine whether the high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning can achieve the expected effect;

[0049] When the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning fails to achieve the expected results, the parameters of the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning are adjusted, and the structure of the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning is continuously iteratively optimized until the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning can achieve the expected results.

[0050] Preferably, in S2, analyzing and pattern recognition of electrophysiological characteristic signals of the patient based on the intelligent sensor includes:

[0051] Obtain a high-precision electrocardiographic signal analysis and recognition model based on deep learning, and deploy the high-precision electrocardiographic signal analysis and recognition model based on deep learning in an actual high-precision electrocardiographic signal analysis environment;

[0052] The patient's electrophysiological characteristic signals based on the smart sensor are input into the high-precision electrophysiological signal analysis and recognition model based on deep learning. The patient's electrophysiological characteristic signals based on the smart sensor are analyzed and pattern recognized according to the high-precision electrophysiological signal analysis and recognition model based on deep learning to judge the patient's heart condition and determine the high-precision electrophysiological signal analysis and recognition results.

[0053] Preferably, in S3, generating a high-precision electrocardiographic physiological signal analysis report includes:

[0054] Based on the high-precision electrocardiographic signal analysis and identification results, timely warnings are given to patients of abnormal cardiac events. Based on the high-precision electrocardiographic signal analysis and identification results, combined with the patient's real-time electrophysiological signals, analysis is performed to generate a high-precision electrocardiographic signal analysis report, which is then stored and transmitted, and displayed to patients and doctors in a visual form for subsequent follow-up observation and research.

[0055] Preferably, in S3, adapting to abnormal changes in the patient's heart includes:

[0056] Doctors develop treatment plans for patients based on high-precision electrophysiological signal analysis reports, and track the treatment plans and patient treatment effects in real time. Doctors make real-time adjustments to treatment plans based on patient treatment effects and patient feedback to adapt to abnormal changes in the patient's heart.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] The present invention collects the patient's real-time electrophysiological signals through the smart sensor integrated on the smart wearable device, and processes the patient's real-time electrophysiological signals based on the smart sensor to generate the patient's electrophysiological characteristic signals based on the smart sensor, analyzes and recognizes the patient's electrophysiological characteristic signals based on the smart sensor according to the high-precision electrophysiological signal analysis and recognition model based on deep learning, determines the high-precision electrophysiological signal analysis and recognition results, and timely warns the patient's abnormal heart events, generates a high-precision electrophysiological signal analysis report, and displays it in a visual form for subsequent tracking observation and research, and formulates a treatment plan for the patient to adapt to the patient's abnormal heart changes. The electrophysiological signals can be effectively analyzed, the accuracy and reliability of electrophysiological signal analysis and recognition can be improved, the abnormal heart conditions can be accurately identified, and the accuracy of diagnosis and prognosis can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 The flowchart of the high-precision electrocardiographic physiological signal analysis method of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0061] In order to solve the problem that the existing technology cannot effectively analyze the electrophysiological signals, resulting in low accuracy and reliability of electrophysiological signal analysis and recognition, and cannot accurately identify cardiac abnormalities, resulting in low accuracy of diagnosis and prognosis, please refer to Figure 1 , this embodiment provides the following technical solutions:

[0062] A high-precision electrocardiographic signal analysis method comprises the following steps:

[0063] S1. Signal acquisition and processing: Collect the patient's real-time electrophysiological signals based on the smart sensor integrated on the smart wearable device, and process the patient's real-time electrophysiological signals based on the smart sensor to generate the patient's electrophysiological characteristic signals based on the smart sensor;

[0064] In this embodiment, collecting the patient's real-time electrophysiological signals includes:

[0065] Based on the smart sensors integrated on the smart wearable device, the patient's ECG signal, intracardiac electrogram, heart sound signal and myoelectric signal are monitored and continuously collected in real time in a quiet state to obtain the patient's real-time ECG signal based on the smart sensor;

[0066] Based on the smart sensors integrated on the smart wearable devices, the patient's blood pressure signal, respiratory signal, temperature signal and blood oxygen saturation signal are monitored and collected continuously in real time in a quiet state to obtain the patient's real-time physiological signals based on the smart sensors;

[0067] Wherein, the patient's real-time electrocardiographic and physiological signals based on the smart sensor are determined according to the patient's real-time electrocardiographic signals based on the smart sensor and the patient's real-time physiological signals based on the smart sensor.

[0068] It should be noted that the ECG signal is a signal that records the electrical activity of the heart through electrodes placed on the body surface. The ECG signal includes P wave, QRS complex and T wave, which represent the processes of atrial depolarization, ventricular depolarization and ventricular repolarization respectively.

[0069] Intracardiac electrograms are electrocardiograms recorded through intracardiac electrodes, providing more detailed information about the electrical activity of the heart, including the electrical activity of the atria, ventricles, and His bundle.

[0070] The heart sound signal is a heart sound signal recorded by a microphone or sensor placed on the body surface. The heart sound signal includes the first heart sound and the second heart sound, which represent the beginning of ventricular contraction and relaxation respectively.

[0071] EMG signals are muscle electrical activity signals recorded by electrodes placed on the body surface. In electrophysiology, EMG signals are mainly used to detect muscle activity around the heart.

[0072] The blood pressure signal is a blood pressure change signal recorded by a blood pressure sensor. The blood pressure signal includes systolic pressure and diastolic pressure, which represent the blood pressure levels during heart contraction and relaxation, respectively.

[0073] The respiratory signal is a respiratory movement signal recorded by a respiratory sensor. The respiratory signal includes respiratory frequency and respiratory depth, which reflects the functional state of the respiratory system.

[0074] The temperature signal is a body temperature change signal recorded by a temperature sensor. The body temperature change can reflect the metabolic state of the heart and body.

[0075] The blood oxygen saturation signal is a blood oxygen saturation change signal recorded by a blood oxygen saturation meter. The blood oxygen saturation reflects the oxygen content in the blood, which can reflect the functional status of the heart and respiratory system.

[0076] In this embodiment, the real-time electrophysiological signals of the patient based on the intelligent sensor are processed, including:

[0077] Acquire the patient's real-time electrophysiological signal based on the intelligent sensor, and perform filtering processing on the patient's real-time electrophysiological signal based on the intelligent sensor based on a median filter;

[0078] Wherein, according to the characteristics of the patient's real-time electrophysiological signal based on the intelligent sensor, the median filter window is determined, and when the size of the median filter window is determined, the size of the median filter window is smaller than the width of the P wave, the QRS complex wave and the T wave in the patient's real-time electrophysiological signal;

[0079] Based on the median filtering algorithm, the median of all data points in the median filter window is calculated to replace the data point in the center of the window, thereby eliminating isolated noise points and removing noise from the real-time electrophysiological signal of the patient based on the smart sensor, while retaining the edges and details of the real-time electrophysiological signal of the patient based on the smart sensor.

[0080] Specifically, determining the median filter window size includes:

[0081] Retrieve the width of P wave, QRS complex wave and T wave in the patient's real-time electrophysiological signal;

[0082] Retrieving the maximum width from the widths of the P wave, QRS complex wave and T wave in the real-time electrophysiological signal of the patient, and using the time length corresponding to the maximum width as the initial window;

[0083] Extracting the waveform widths of the P wave, QRS complex wave and T wave;

[0084] Obtaining a first window setting coefficient using the waveform widths of the P wave, QRS complex wave, and T wave;

[0085] The first window setting coefficient is obtained by the following formula:

[0086]

[0087] Among them, S 01 Indicates the first window setting coefficient; K p , K q and K t Respectively represent the time length corresponding to the waveform width of P wave, QRS complex wave and T wave; K c Indicates the preset time length reference value;

[0088] comparing the first window setting coefficient with a preset window setting coefficient threshold;

[0089] When the first window setting coefficient is lower than a preset window setting coefficient threshold, the median filter window is set according to a preset initial window size;

[0090] When the first window setting coefficient exceeds a preset window setting coefficient threshold, the median filter window is set using the signal dynamic range of the P wave, QRS complex wave and T wave in combination with the first window setting coefficient.

[0091] The technical effect of the above technical scheme is: the scheme retrieves the width of the P wave, QRS complex wave and T wave in the patient's real-time electrophysiological signal, and determines the size of the median filter window according to the characteristics of these waveforms. This method enables the filter window size to adapt to the patient's electrocardiographic signal characteristics, thereby improving the pertinence and effectiveness of the filtering. The scheme mentions that the maximum width is first determined as the initial window, and then the window size is further adjusted by calculating the first window setting coefficient. This method combines the main waveform characteristics of the electrocardiographic signal, so that the window size is more in line with the actual situation of the electrocardiographic signal, thereby improving the accuracy of the filtering. The scheme introduces a comparison mechanism between the first window setting coefficient and the preset window setting coefficient threshold. When the first window setting coefficient is lower than the threshold, the preset initial window size is adopted; when it exceeds the threshold, it is adjusted in combination with the signal dynamic range. This dynamic adjustment capability enables the filter to better adapt to the changes of different electrocardiographic signals and improves the stability and reliability of the filtering. By comprehensively considering the waveform width of the P wave, QRS complex wave and T wave and the signal dynamic range, the scheme can more accurately set the median filter window size, thereby optimizing the filtering effect. This helps reduce noise interference, improves the clarity and readability of ECG signals, and provides doctors with more accurate diagnostic evidence. Since the scheme can adaptively adjust the filter window size and optimize the filtering effect, it can more effectively extract and analyze ECG signal features. This helps doctors make diagnoses faster and improves diagnostic efficiency.

[0092] In summary, this technical solution improves the accuracy and stability of ECG signal processing by adaptively determining the median filter window size, optimizes the filtering effect, provides doctors with more accurate diagnostic basis, and also improves diagnostic efficiency.

[0093] Specifically, when the first window setting coefficient exceeds a preset window setting coefficient threshold, the median filter window is set by using the signal dynamic range of the P wave, QRS complex wave and T wave in combination with the first window setting coefficient, including:

[0094] When the first window setting coefficient exceeds a preset window setting coefficient threshold, retrieving the signal dynamic range of the P wave, QRS complex wave and T wave;

[0095] Obtaining a second window setting coefficient using the signal dynamic range of the P wave, QRS complex wave and T wave;

[0096] The second window setting coefficient is obtained by the following formula:

[0097]

[0098] Among them, S 02 Indicates the second window setting coefficient; D p , D q and D t Represents the signal dynamic range of P wave, QRS complex wave and T wave respectively; D c Indicates a preset signal dynamic range reference value;

[0099] The median filter window is set using the first window setting coefficient and the second window setting coefficient, wherein the size of the median filter window after the setting is obtained by the following formula:

[0100]

[0101] Among them, K h Indicates the size of the median filter window after setting; K s Indicates the initial window size; S 01 Indicates the first window setting coefficient; S 02 Indicates the second window setting coefficient; S y Indicates the preset window setting coefficient threshold.

[0102] The technical effect of the above technical solution is: when the first window setting coefficient exceeds the preset window setting coefficient threshold, the solution no longer relies solely on the initial window size or a single waveform width feature, but further considers the signal dynamic range of the P wave, QRS complex wave and T wave. This method of comprehensively considering multiple ECG signal characteristics makes the setting of the median filter window size more flexible and adaptable. By introducing the second window setting coefficient, the solution can more accurately reflect the changing characteristics of the ECG signal. By setting the median filter window in combination with the first window setting coefficient and the second window setting coefficient, the noise in the ECG signal can be more effectively removed while retaining the important features of the signal, thereby improving the accuracy of filtering. The solution uses two window setting coefficients to jointly determine the size of the median filter window, so that the performance of the filter is optimized. Through the set median filter window size formula, it can be ensured that the filter can achieve a good filtering effect under different signal characteristics, which can remove noise and retain the detailed information of the signal. Since the solution can adaptively adjust the median filter window size and optimize the filtering effect, the ECG signal features can be more effectively extracted and analyzed. This helps doctors to more accurately judge the patient's heart condition and improve the accuracy of diagnosis. This solution comprehensively considers multiple characteristics of ECG signals, including waveform width and signal dynamic range, so that the system can have better adaptability to different types of ECG signals. This design enhances the robustness of the system, allowing the system to maintain stable performance when facing different patients or different ECG signals.

[0103] In summary, this technical solution adaptively adjusts the median filter window size by comprehensively considering multiple characteristics of the ECG signal, thereby improving the adaptability and filtering accuracy of the filter, optimizing the filtering effect, improving the diagnostic accuracy, and enhancing the robustness of the system.

[0104] In this embodiment, the real-time electrophysiological signal of the patient based on the intelligent sensor is processed, and further includes:

[0105] Acquire the real-time electrophysiological signal of the patient based on the intelligent sensor after filtering, and transform the real-time electrophysiological signal of the patient based on the intelligent sensor based on wavelet transform;

[0106] Among them, according to the characteristics of the patient's real-time electrophysiological signal based on the intelligent sensor, a wavelet basis with compact support and symmetry is selected;

[0107] According to the frequency range of the patient's real-time electrophysiological signal based on the smart sensor and the required decomposition details, the number of decomposition layers is determined, and the patient's real-time electrophysiological signal based on the smart sensor is decomposed into approximate coefficients and detail coefficients of different scales using a wavelet basis with compact support and symmetry;

[0108] Useful characteristic parameters are extracted from the wavelet decomposition coefficients and fused to generate the patient's electrophysiological characteristic signals based on intelligent sensors.

[0109] S2. Signal analysis and recognition: Train a high-precision electrophysiological signal analysis and recognition model based on deep learning, analyze and pattern recognize the electrophysiological characteristic signals of patients based on smart sensors according to the high-precision electrophysiological signal analysis and recognition model based on deep learning, and determine the high-precision electrophysiological signal analysis and recognition results;

[0110] In this embodiment, training a high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning includes:

[0111] Collect the patient's historical electrophysiological signals according to the needs of high-precision electrophysiological signal analysis;

[0112] Classifying the collected electrocardiographic physiological history signals of the patients, and dividing the collected electrocardiographic physiological history signals of the patients into a training set and a test set;

[0113] Based on deep learning technology, a training set is used to train the deep learning model, so that the deep learning model can autonomously learn and analyze the high-precision electrocardiographic physiological signal analysis process and patient cardiac pattern recognition, and determine a high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning;

[0114] Use the test set to perform a performance test on the high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning to determine whether the high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning can achieve the expected effect;

[0115] When the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning fails to achieve the expected results, the parameters of the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning are adjusted, and the structure of the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning is continuously iteratively optimized until the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning can achieve the expected results.

[0116] In this embodiment, the analysis and pattern recognition of the electrophysiological characteristic signals of the patient based on the smart sensor include:

[0117] Obtain a high-precision electrocardiographic signal analysis and recognition model based on deep learning, and deploy the high-precision electrocardiographic signal analysis and recognition model based on deep learning in an actual high-precision electrocardiographic signal analysis environment;

[0118] The patient's electrophysiological characteristic signals based on the smart sensor are input into the high-precision electrophysiological signal analysis and recognition model based on deep learning. The patient's electrophysiological characteristic signals based on the smart sensor are analyzed and pattern recognized according to the high-precision electrophysiological signal analysis and recognition model based on deep learning to judge the patient's heart condition and determine the high-precision electrophysiological signal analysis and recognition results.

[0119] S3. Visual management of results: Provide timely warnings for abnormal cardiac events in patients and generate high-precision electrophysiological signal analysis reports, which are displayed in a visual form for subsequent follow-up observations and research, and develop treatment plans for patients to adapt to abnormal changes in their heart conditions.

[0120] In this embodiment, a high-precision electrocardiographic signal analysis report is generated, including:

[0121] Based on the high-precision electrocardiographic signal analysis and identification results, timely warnings are given to patients of abnormal cardiac events. Based on the high-precision electrocardiographic signal analysis and identification results, combined with the patient's real-time electrophysiological signals, analysis is performed to generate a high-precision electrocardiographic signal analysis report, which is then stored and transmitted, and displayed to patients and doctors in a visual form for subsequent follow-up observation and research.

[0122] In this embodiment, adapting to abnormal changes in the patient's heart includes:

[0123] Doctors develop treatment plans for patients based on high-precision electrophysiological signal analysis reports, and track the treatment plans and patient treatment effects in real time. Doctors make real-time adjustments to treatment plans based on patient treatment effects and patient feedback to adapt to abnormal changes in the patient's heart.

[0124] In summary, the patient's real-time electrophysiological signals are collected through the smart sensors integrated on the smart wearable devices, and the patient's real-time electrophysiological signals based on the smart sensors are processed to generate the patient's electrophysiological characteristic signals based on the smart sensors. The patient's electrophysiological characteristic signals based on the smart sensors are analyzed and pattern recognized according to the high-precision electrophysiological signal analysis and recognition model based on deep learning, the high-precision electrophysiological signal analysis and recognition results are determined, and timely warnings are given to the patient's abnormal cardiac events. A high-precision electrophysiological signal analysis report is generated and displayed in a visual form for subsequent tracking observation and research, and treatment plans are formulated for patients to adapt to the patient's abnormal heart changes. The electrophysiological signals can be effectively analyzed, the accuracy and reliability of electrophysiological signal analysis and recognition can be improved, and abnormal cardiac conditions can be accurately identified, which can improve the accuracy of diagnosis and prognosis.

[0125] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, method, article or device.

[0126] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A high-precision electrocardiographic signal analysis method, characterized in that: The steps include: S1. Signal acquisition and processing: Collect the patient's real-time electrophysiological signals based on the smart sensor integrated on the smart wearable device, and process the patient's real-time electrophysiological signals based on the smart sensor to generate the patient's electrophysiological characteristic signals based on the smart sensor; S2. Signal analysis and recognition: Train a high-precision electrophysiological signal analysis and recognition model based on deep learning, analyze and pattern recognize the electrophysiological characteristic signals of patients based on smart sensors according to the high-precision electrophysiological signal analysis and recognition model based on deep learning, and determine the high-precision electrophysiological signal analysis and recognition results; S3. Visual management of results: Provide timely warnings for abnormal cardiac events in patients and generate high-precision electrophysiological signal analysis reports, which are displayed in a visual form for subsequent follow-up observations and research, and develop treatment plans for patients to adapt to abnormal changes in their heart conditions.

2. A high-precision electrocardiographic signal analysis method as claimed in claim 1, characterized in that: In S1, collecting the patient's real-time electrophysiological signals includes: Based on the smart sensors integrated on the smart wearable device, the patient's ECG signal, intracardiac electrogram, heart sound signal and myoelectric signal are monitored and continuously collected in real time in a quiet state to obtain the patient's real-time ECG signal based on the smart sensor; Based on the smart sensors integrated on the smart wearable devices, the patient's blood pressure signal, respiratory signal, temperature signal and blood oxygen saturation signal are monitored and collected continuously in real time in a quiet state to obtain the patient's real-time physiological signals based on the smart sensors; Wherein, the patient's real-time electrocardiographic and physiological signals based on the smart sensor are determined according to the patient's real-time electrocardiographic signals based on the smart sensor and the patient's real-time physiological signals based on the smart sensor.

3. A high-precision electrocardiographic signal analysis method as claimed in claim 2, characterized in that: In S1, the real-time electrophysiological signal of the patient based on the intelligent sensor is processed, including: Acquire the patient's real-time electrophysiological signal based on the intelligent sensor, and perform filtering processing on the patient's real-time electrophysiological signal based on the intelligent sensor based on a median filter; Wherein, according to the characteristics of the patient's real-time electrophysiological signal based on the intelligent sensor, the median filter window is determined, and when the size of the median filter window is determined, the size of the median filter window is smaller than the width of the P wave, the QRS complex wave and the T wave in the patient's real-time electrophysiological signal; Based on the median filtering algorithm, the median of all data points in the median filter window is calculated to replace the data point in the center of the window, thereby eliminating isolated noise points and removing noise from the real-time electrophysiological signal of the patient based on the smart sensor, while retaining the edges and details of the real-time electrophysiological signal of the patient based on the smart sensor.

4. A high-precision electrocardiographic signal analysis method as claimed in claim 3, characterized in that: Determining the median filter window size comprises: Retrieve the width of P wave, QRS complex wave and T wave in the patient's real-time electrophysiological signal; Retrieving the maximum width from the widths of the P wave, QRS complex wave and T wave in the real-time electrophysiological signal of the patient, and using the time length corresponding to the maximum width as the initial window; Extracting the waveform widths of the P wave, QRS complex wave and T wave; Obtaining a first window setting coefficient using the waveform widths of the P wave, QRS complex wave, and T wave; The first window setting coefficient is obtained by the following formula: Among them, S 01 Indicates the first window setting coefficient; K p , K q and K t Respectively represent the time length corresponding to the waveform width of P wave, QRS complex wave and T wave; K c Indicates the preset time length reference value; comparing the first window setting coefficient with a preset window setting coefficient threshold; When the first window setting coefficient is lower than a preset window setting coefficient threshold, the median filter window is set according to a preset initial window size; When the first window setting coefficient exceeds a preset window setting coefficient threshold, the median filter window is set using the signal dynamic range of the P wave, QRS complex wave and T wave in combination with the first window setting coefficient.

5. A high-precision electrocardiographic signal analysis method as claimed in claim 4, characterized in that: When the first window setting coefficient exceeds a preset window setting coefficient threshold, the median filter window is set by using the signal dynamic range of the P wave, the QRS complex wave and the T wave in combination with the first window setting coefficient, including: When the first window setting coefficient exceeds a preset window setting coefficient threshold, retrieving the signal dynamic range of the P wave, QRS complex wave and T wave; Obtaining a second window setting coefficient using the signal dynamic range of the P wave, QRS complex wave and T wave; The second window setting coefficient is obtained by the following formula: Among them, S 02 Indicates the second window setting coefficient; D p , D q and D t Represents the signal dynamic range of P wave, QRS complex wave and T wave respectively; D c Indicates a preset signal dynamic range reference value; The median filter window is set using the first window setting coefficient and the second window setting coefficient, wherein the size of the median filter window after the setting is obtained by the following formula: Among them, K h Indicates the size of the median filter window after setting; K s Indicates the initial window size; S 01 Indicates the first window setting coefficient; S 02 Indicates the second window setting coefficient; S y Indicates the preset window setting coefficient threshold.

6. A high-precision electrocardiographic signal analysis method as claimed in claim 3, characterized in that: In the above S1, the real-time electrophysiological signal of the patient based on the intelligent sensor is processed, and further includes: Acquire the real-time electrophysiological signal of the patient based on the intelligent sensor after filtering, and transform the real-time electrophysiological signal of the patient based on the intelligent sensor based on wavelet transform; Among them, according to the characteristics of the patient's real-time electrophysiological signal based on the intelligent sensor, a wavelet basis with compact support and symmetry is selected; According to the frequency range of the patient's real-time electrophysiological signal based on the smart sensor and the required decomposition details, the number of decomposition layers is determined, and the patient's real-time electrophysiological signal based on the smart sensor is decomposed into approximate coefficients and detail coefficients of different scales using a wavelet basis with compact support and symmetry; Useful characteristic parameters are extracted from the wavelet decomposition coefficients and fused to generate the patient's electrophysiological characteristic signals based on intelligent sensors.

7. A high-precision electrocardiographic signal analysis method as claimed in claim 6, characterized in that: In S2, a high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning is trained, including: Collect the patient's historical electrophysiological signals according to the needs of high-precision electrophysiological signal analysis; Classifying the collected electrocardiographic physiological history signals of the patients, and dividing the collected electrocardiographic physiological history signals of the patients into a training set and a test set; Based on deep learning technology, a training set is used to train the deep learning model, so that the deep learning model can autonomously learn and analyze the high-precision electrocardiographic physiological signal analysis process and patient cardiac pattern recognition, and determine a high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning; Use the test set to perform a performance test on the high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning to determine whether the high-precision electrocardiographic physiological signal analysis and recognition model based on deep learning can achieve the expected effect; When the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning fails to achieve the expected results, the parameters of the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning are adjusted, and the structure of the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning is continuously iteratively optimized until the high-precision electrocardiophysiological signal analysis and recognition model based on deep learning can achieve the expected results.

8. A high-precision electrocardiographic signal analysis method as claimed in claim 7, characterized in that: In S2, the electrophysiological characteristic signals of the patient based on the intelligent sensor are analyzed and pattern recognized, including: Obtain a high-precision electrocardiographic signal analysis and recognition model based on deep learning, and deploy the high-precision electrocardiographic signal analysis and recognition model based on deep learning in an actual high-precision electrocardiographic signal analysis environment; The patient's electrophysiological characteristic signals based on the smart sensor are input into the high-precision electrophysiological signal analysis and recognition model based on deep learning. The patient's electrophysiological characteristic signals based on the smart sensor are analyzed and pattern recognized according to the high-precision electrophysiological signal analysis and recognition model based on deep learning to judge the patient's heart condition and determine the high-precision electrophysiological signal analysis and recognition results.

9. A high-precision electrocardiographic signal analysis method as claimed in claim 8, characterized in that: In S3, a high-precision electrocardiographic physiological signal analysis report is generated, including: Based on the high-precision electrocardiographic signal analysis and identification results, timely warnings are given to patients of abnormal cardiac events. Based on the high-precision electrocardiographic signal analysis and identification results, combined with the patient's real-time electrophysiological signals, analysis is performed to generate a high-precision electrocardiographic signal analysis report, which is then stored and transmitted, and displayed to patients and doctors in a visual form for subsequent follow-up observation and research.

10. A high-precision electrocardiographic signal analysis method as claimed in claim 9, characterized in that: In S3, adapting to abnormal changes in the patient's heart includes: Doctors develop treatment plans for patients based on high-precision electrophysiological signal analysis reports, and track the treatment plans and patient treatment effects in real time. Doctors make real-time adjustments to treatment plans based on patient treatment effects and patient feedback to adapt to abnormal changes in the patient's heart.

Images