Patient health early warning processing method and system based on various heart monitoring data

Abstract

The invention discloses a patient health early warning processing method and system based on various heart monitoring data, and relates to the technical field of health diagnos.The early warning processing steps are as follows: S1, acquiring heart monitoring data including electrocardiogram data, blood pressure data and heart ultrasonic data, and constructing a data set after preprocessing the data; s2, feature extraction is conducted on data in the data set, heart disease related feature data are obtained based on big data, and the related feature data comprise electrocardiogram data, blood pressure data and cardiac ultrasound data; and S3, performing vectorization processing on the obtained features, comparing feature data monitored in real time with feature data obtained by big data, and judging whether the feature data are consistent with the feature data. According to the method, electrocardiogram, blood pressure and cardiac ultrasound multi-source data are integrated, the limitation of single data is broken through, the electrical activity, the structure function and the hemodynamics condition of the heart are comprehensively reflected, big data feature analysis and comparison are combined, the evaluation accuracy is improved, and missed diagnosis and misdiagnosis are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of health diagnosis, and in particular to a method and system for processing patient health early warning data based on multiple heart monitoring data. Background Art

[0002] With the increasing aging of society and changes in lifestyle, heart disease has become one of the major diseases threatening human health, characterized by high morbidity, high disability rate and high mortality rate. Timely and accurate monitoring and early warning of heart health status are crucial to preventing the occurrence and development of heart disease, improving patients' quality of life and reducing mortality.

[0003] Currently, there are a variety of monitoring methods in the field of cardiac health monitoring, which can obtain electrocardiogram data, blood pressure data, and cardiac ultrasound data information. However, the existing technologies have many shortcomings. On the one hand, most monitoring systems only analyze and issue warnings for a single type of data, such as using only electrocardiogram data to judge arrhythmias, or assessing cardiovascular risks based solely on blood pressure data. They lack the ability to comprehensively process multiple cardiac monitoring data, making it difficult to fully reflect the patient's cardiac health status, resulting in poor accuracy of warning results and prone to omissions or false alarms. To this end, we proposed a patient health warning processing method and system for multiple cardiac monitoring data. Summary of the Invention

[0004] In order to solve the above technical problems, a patient health warning processing method and system for multiple heart monitoring data are provided. This technical solution solves the above problems.

[0005] To achieve the above objectives, the present invention adopts a technical solution: a patient health early warning processing method based on multiple heart monitoring data, wherein the early warning processing steps are:

[0006] S1. Obtain heart monitoring data, including electrocardiogram data, blood pressure data, and cardiac ultrasound data, and construct a data set after preprocessing the data;

[0007] S2. Extract features from the data in the dataset and obtain feature data related to heart disease based on the big data. The relevant feature data includes electrocardiogram data, blood pressure data, and cardiac ultrasound data.

[0008] S3. Vectorize the acquired features, compare the real-time monitored feature data with the feature data acquired from big data, and determine whether they are consistent;

[0009] S4. Based on the consistent judgment results, weight values are assigned to the monitoring data respectively, and weighted calculation is performed to obtain comprehensive heart health data;

[0010] S5. Based on the obtained comprehensive heart health data, a health threshold is established. When the health threshold is greater than the health threshold, a warning is issued to check for heart safety hazards.

[0011] Preferably, the heart monitoring data in step S1 is obtained through routine hospital examinations, dynamic electrocardiogram monitoring, and wearable devices;

[0012] Data preprocessing includes denoising, outlier removal and image enhancement;

[0013] The preprocessed electrocardiogram data, blood pressure data and cardiac ultrasound data are integrated to construct a data set and establish a retrieval index.

[0014] Preferably, in step S2, the electrocardiogram data feature extraction is performed by identifying the QRS complex using a threshold detection method, searching for P waves and T waves based on the position of the QRS complex, and the extracted features are amplitude and morphological features;

[0015] Blood pressure data feature extraction is done by calculating the mean of systolic pressure, diastolic pressure, and mean arterial pressure to obtain mean features, and calculating the standard deviation and coefficient of variation of blood pressure to obtain fluctuation features;

[0016] Cardiac ultrasound data features include morphological features and functional features.

[0017] Preferably, the specific calculation and extraction steps of the features in step S2 are:

[0018] The ECG signal is preprocessed, and a numerical standard is set as a threshold. When the value of the preprocessed ECG signal exceeds the threshold, it is determined to be a QRS complex; if it does not exceed the threshold, it is not a QRS complex. After determining the start and end positions of the QRS complex, the positions of the P wave and T wave are found within a specific time range before the start position of the QRS complex and within a specific time range after the end position of the QRS complex.

[0019] Find the maximum value in the electrocardiogram signal parts corresponding to the P wave, QRS complex and T wave respectively. The maximum value is the amplitude of the corresponding wave.

[0020] The morphological characteristics are described by the degree of inclination and curvature of the waveform;

[0021] Mean feature extraction: add up all measured systolic blood pressure values and divide by the total number of measurements to obtain the mean systolic blood pressure value;

[0022] For diastolic blood pressure mean extraction, all measured diastolic blood pressure values were added together and then divided by the total number of measurements to calculate the mean diastolic blood pressure;

[0023] To extract the mean value of mean arterial pressure, add twice the diastolic pressure to the systolic pressure of each measurement and divide by 3 to obtain the mean arterial pressure value for each measurement; add all the mean arterial pressure values and divide by the number of measurements to obtain the mean arterial pressure.

[0024] The fluctuation characteristic calculates the difference between each blood pressure measurement and the corresponding blood pressure mean, squares each difference, adds it up, divides it by the number of measurements, and takes the square root of the result to obtain the standard deviation of the blood pressure, which is used to reflect the degree of dispersion of blood pressure fluctuations.

[0025] The coefficient of variation is calculated by dividing the standard deviation of blood pressure by the corresponding mean blood pressure. The result is the coefficient of variation, which is used to measure the relative size of blood pressure fluctuations;

[0026] The morphological characteristics are measured by ultrasound images to obtain the value of the left ventricular end-diastolic diameter each time. The values of multiple measurements are added together and divided by the number of measurements to obtain the average value of the left ventricular end-diastolic diameter, which is the morphological characteristic value.

[0027] Functional characteristics: The ejection fraction was obtained by subtracting the left ventricular end-systolic volume from the left ventricular end-diastolic volume. The difference was divided by the left ventricular end-diastolic volume and the result was multiplied by 100% to reflect the cardiac systolic function.

[0028] Preferably, the vectorization processing steps in step S3 are:

[0029] After obtaining the electrocardiogram, blood pressure, and cardiac ultrasound features, vectorization processing is performed to integrate and classify different types of features, clarify the feature categories of the electrocardiogram, blood pressure, and cardiac ultrasound, perform feature encoding, convert features of different magnitudes into the same scale, and arrange the processed features in order to form the feature vector of each data.

[0030] Preferably, the step of determining whether the consistency is achieved in step S3 is:

[0031] The two sets of vectors are calculated based on the cosine similarity method, and the feature vectors of the two types of data are compared. If the similarity index is higher than the set threshold, the data is considered consistent; otherwise, it is inconsistent.

[0032] Based on the comparison results, it is determined whether the feature data obtained by real-time monitoring is consistent with that obtained by big data.

[0033] Preferably, the weight value in step S4 is determined based on the analytic hierarchy process.

[0034] Specifically, a hierarchical model was constructed, with the assessment of comprehensive cardiac health status as the target layer, the reliability of data sources, clinical diagnostic importance, and data consistency as the criterion layer, and electrocardiogram, blood pressure, and cardiac ultrasound monitoring data as the scenario layer. Based on the 1-9 scaling method, a judgment matrix was constructed by comparing the scenario layer data pairwise. The consistency index CI was calculated and a consistency test was performed. The result was valid when CI was < 0.1, otherwise the matrix was adjusted. The weights were calculated based on the square root method.

[0035] Preferably, the weighted calculation formula in step S4 is:

[0036] Assume that the eigenvectors of electrocardiogram data, blood pressure data, and cardiac ultrasound data after vectorization are The corresponding weights are w ecg 、w bp 、w ech o , and satisfy w ecg +w bp +w ech o =1, comprehensive heart health data The calculation formula is:

[0037]

[0038] in is the calculated comprehensive heart health data value.

[0039] Preferably, in step S5, an early warning system is constructed based on comprehensive heart health data, and health thresholds are established based on expert analysis;

[0040] During the hidden danger investigation stage, medical personnel will conduct preliminary assessments and give inspection recommendations. In complex cases, expert consultations will be organized to clarify the condition, develop personalized treatment plans, establish a tracking and follow-up mechanism, and monitor data regularly.

[0041] A patient health warning processing system for multiple heart monitoring data, the health warning processing system comprising:

[0042] a data acquisition and processing module configured to acquire heart monitoring data and pre-process it to construct a data set;

[0043] a feature extraction module configured to extract features in the data set and obtain feature data related to heart disease based on the big data;

[0044] The judgment module is configured to perform consistency judgment on two sets of data;

[0045] a comprehensive analysis module configured to assign weight values to the monitored data and perform weighted calculations on comprehensive heart health data;

[0046] The early warning module is configured to perform early warning processing based on health data results.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The present invention integrates multi-source data such as electrocardiogram, blood pressure, and cardiac ultrasound, breaking through the limitations of single data and comprehensively reflecting the cardiac electrical activity, structural function, and hemodynamic status. It combines big data feature analysis and comparison to improve assessment accuracy and reduce missed diagnoses and misdiagnoses. Based on a large amount of historical data and real-time monitoring data, it provides data support for medical decision-making, assists doctors in formulating more scientific diagnosis and treatment plans, and at the same time contributes to medical research and promotes the development of heart disease prevention and treatment technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of the early warning processing steps of the present invention;

[0050] Figure 2 This is a framework diagram of the health warning processing system of the present invention. DETAILED DESCRIPTION

[0051] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0052] Reference Figure 1 As shown, a patient health early warning processing method based on multiple heart monitoring data includes the following steps:

[0053] S1. Obtain heart monitoring data, including electrocardiogram data, blood pressure data, and cardiac ultrasound data, and construct a data set after preprocessing the data;

[0054] S2. Extract features from the data in the dataset and obtain feature data related to heart disease based on the big data. The relevant feature data includes electrocardiogram data, blood pressure data, and cardiac ultrasound data.

[0055] S3. Vectorize the acquired features, compare the real-time monitored feature data with the feature data acquired from big data, and determine whether they are consistent;

[0056] S4. Based on the consistent judgment results, weight values are assigned to the monitoring data respectively, and weighted calculation is performed to obtain comprehensive heart health data;

[0057] S5. Based on the obtained comprehensive heart health data, a health threshold is established. When the health threshold is greater than the health threshold, a warning is issued to check for heart safety hazards.

[0058] This application integrates multi-source data such as electrocardiogram, blood pressure, and cardiac ultrasound, breaking through the limitations of single data, and comprehensively reflecting the electrical activity, structural function, and hemodynamic status of the heart. It combines big data feature analysis and comparison to improve assessment accuracy and reduce missed diagnoses and misdiagnoses; through vector processing and data comparison, monitoring data are weighted according to individual differences, and personalized comprehensive heart health data is calculated by weight, so that the warning fits the patient's actual situation and avoids "one size fits all"; real-time acquisition and processing of data can promptly detect changes in heart health status. Once the data exceeds the health threshold, an immediate warning will be issued to quickly check for safety hazards, buy time for treatment for patients, and reduce the risk of heart disease. A large amount of historical data and real-time monitoring data provide data support for medical decision-making, assist doctors in formulating more scientific diagnosis and treatment plans, and at the same time contribute to medical research and promote the development of heart disease prevention and treatment technologies.

[0059] In step S1, the heart monitoring data is obtained through routine hospital examinations, dynamic electrocardiogram monitoring, and wearable devices;

[0060] Data preprocessing includes denoising, outlier removal and image enhancement;

[0061] The preprocessed electrocardiogram data, blood pressure data and cardiac ultrasound data are integrated to construct a data set and establish a retrieval index.

[0062] This application uses multiple channels to obtain data through routine hospital examinations, Holter monitoring, and wearable devices, covering clinical professional examinations, long-term dynamic monitoring, and daily real-time monitoring scenarios. Routine hospital examinations provide professional and accurate diagnostic data; Holter monitoring can capture long-term heart changes; wearable devices facilitate daily monitoring at any time. The combination of these three methods can not only obtain professional and in-depth data, but also grasp the dynamics of heart health in real time, and comprehensively reflect the patient's heart condition;

[0063] The denoising operation in data preprocessing can effectively remove various interference signals mixed in during the acquisition process, such as power frequency interference and electromyographic interference in the electrocardiogram, making the data clearer and more accurate; removing outliers can eliminate invalid data generated by equipment failure and improper operation, ensuring data authenticity; image enhancement processing for cardiac ultrasound data can improve image contrast and clarity, facilitate subsequent more accurate extraction of image features, significantly improve data quality, and provide reliable guarantees for accurate analysis.

[0064] In step S2, the ECG data feature extraction is performed by using a threshold detection method to identify the QRS complex, searching for P waves and T waves based on the position of the QRS complex, and the extracted features are amplitude and morphological features;

[0065] Blood pressure data feature extraction is done by calculating the mean of systolic pressure, diastolic pressure, and mean arterial pressure to obtain mean features, and calculating the standard deviation and coefficient of variation of blood pressure to obtain fluctuation features;

[0066] Cardiac ultrasound data features include morphological features and functional features.

[0067] The electrocardiogram data of this application identifies the QRS complex through threshold detection, and searches for P waves and T waves based on this, extracting amplitude and morphological features. This method can accurately capture key information about the heart's electrical activity. Accurate identification of the QRS complex helps to determine the ventricular depolarization situation, and the feature extraction of P waves and T waves can reflect the atrial depolarization and ventricular repolarization status. Changes in amplitude and morphology can intuitively reflect abnormalities in the electrical activity of myocardial cells, providing an important basis for the early detection of arrhythmias and myocardial ischemia.

[0068] The specific calculation and extraction steps of the features in step S2 are:

[0069] The ECG signal is preprocessed, and a numerical standard is set as a threshold. When the value of the preprocessed ECG signal exceeds the threshold, it is determined to be a QRS complex; if it does not exceed the threshold, it is not a QRS complex. After determining the start and end positions of the QRS complex, the positions of the P wave and T wave are found within a specific time range before the start position of the QRS complex and within a specific time range after the end position of the QRS complex.

[0070] Find the maximum value in the electrocardiogram signal parts corresponding to the P wave, QRS complex and T wave respectively. The maximum value is the amplitude of the corresponding wave.

[0071] The morphological characteristics are described by the degree of inclination and curvature of the waveform;

[0072] Mean feature extraction: add up all measured systolic blood pressure values and divide by the total number of measurements to obtain the mean systolic blood pressure value;

[0073] For diastolic blood pressure mean extraction, all measured diastolic blood pressure values were added together and then divided by the total number of measurements to calculate the mean diastolic blood pressure;

[0074] To extract the mean value of mean arterial pressure, add twice the diastolic pressure to the systolic pressure of each measurement and divide by 3 to obtain the mean arterial pressure value for each measurement; add all the mean arterial pressure values and divide by the number of measurements to obtain the mean arterial pressure.

[0075] The fluctuation characteristic calculates the difference between each blood pressure measurement and the corresponding blood pressure mean, squares each difference, adds it up, divides it by the number of measurements, and takes the square root of the result to obtain the standard deviation of the blood pressure, which is used to reflect the degree of dispersion of blood pressure fluctuations.

[0076] The coefficient of variation is calculated by dividing the standard deviation of blood pressure by the corresponding mean blood pressure. The result is the coefficient of variation, which is used to measure the relative size of blood pressure fluctuations;

[0077] The morphological characteristics are measured by ultrasound images to obtain the value of the left ventricular end-diastolic diameter each time. The values of multiple measurements are added together and divided by the number of measurements to obtain the average value of the left ventricular end-diastolic diameter, which is the morphological characteristic value.

[0078] Functional characteristics: The ejection fraction was obtained by subtracting the left ventricular end-systolic volume from the left ventricular end-diastolic volume. The difference was divided by the left ventricular end-diastolic volume and the result was multiplied by 100% to reflect the cardiac systolic function.

[0079] The specific calculation formula in this application is:

[0080] ECG related

[0081] QRS wave group determination: suppose the pre-processed electrocardiogram signal is ECG(t), the set threshold is T, QRS(t) represents the determination result of the QRS wave group, then

[0082]

[0083] Amplitude characteristics: Assume ECG P (t), ECG QRS (t), ECG T (t) are the electrocardiogram signals corresponding to the P wave, QRS complex, and T wave, respectively. P 、A QRS 、A T are the amplitudes of the P wave, QRS complex, and T wave, respectively, then:

[0084] A P =max(ECG P (t)) A QRS =max(ECG QRS (t)) A T =max(ECG T (t))

[0085] The morphological characteristics, taking the slope as an example, are set as S:

[0086]

[0087] Where ECG(t) is the electrocardiogram signal of the corresponding wave. The slope is calculated by taking the derivative to describe the morphological characteristics. The degree of curvature and other morphological characteristics can be described by establishing corresponding formulas according to specific methods.

[0088] Mean characteristics: Let the number of measurements be n, and the systolic blood pressure measured at the i-th time be SBP i , diastolic blood pressure is DBPi MAP i The mean systolic blood pressure is The mean diastolic blood pressure is The mean arterial pressure is but:

[0089]

[0090]

[0091] Fluctuation characteristic standard deviation σ:

[0092]

[0093] BP i is the corresponding blood pressure value measured for the i-th time, is the mean of the corresponding blood pressure,

[0094] Coefficient of variation CV:

[0095]

[0096] Where BP is SBP, DBP, MAP;

[0097] Cardiac ultrasound related

[0098] Morphological characteristics, mean left ventricular end-diastolic diameter Assume that the number of measurements is m, and the left ventricular end-diastolic diameter measured at the jth time is LVEDD j ,but

[0099] Functional characteristics, ejection fraction EF: Let the left ventricular end-diastolic volume be LVEDV and the left ventricular end-systolic volume be LVESV, then:

[0100]

[0101] EF is the calculated ejection fraction.

[0102] The vectorization processing steps in step S3 are:

[0103] After obtaining the electrocardiogram, blood pressure, and cardiac ultrasound features, vectorization processing is performed to integrate and classify different types of features, clarify the feature categories of the electrocardiogram, blood pressure, and cardiac ultrasound, perform feature encoding, convert features of different magnitudes into the same scale, and arrange the processed features in order to form the feature vector of each data.

[0104] The steps for determining whether consistency occurs in step S3 are:

[0105] The two sets of vectors are calculated based on the cosine similarity method, and the feature vectors of the two types of data are compared. If the similarity index is higher than the set threshold, the data is considered consistent; otherwise, it is inconsistent.

[0106] Based on the comparison results, it is determined whether the feature data obtained by real-time monitoring is consistent with that obtained by big data.

[0107] The cosine similarity of the present application measures the similarity of two vectors by calculating the cosine value of the angle between them. In the comparison of feature vectors, it can accurately capture the directional differences between vectors without being affected by the length of the vectors. This is very effective for processing data of different magnitudes but with similar feature patterns. It can more accurately reflect the inherent similarity between data and avoid misjudgments caused by differences in data magnitude. The calculation of cosine similarity is relatively simple, involving only the calculation of the dot product and modulus of the vectors, and has low computational complexity. When processing large amounts of heart monitoring data, it can quickly obtain similarity results, improving the real-time nature of early warning processing and meeting the needs of timely monitoring and rapid response to patient health status.

[0108] The weight value in step S4 is determined based on the analytic hierarchy process.

[0109] Specifically, a hierarchical model was constructed, with the assessment of comprehensive cardiac health status as the target layer, the reliability of data sources, clinical diagnostic importance, and data consistency as the criterion layer, and electrocardiogram, blood pressure, and cardiac ultrasound monitoring data as the scenario layer. Based on the 1-9 scaling method, a judgment matrix was constructed by comparing the scenario layer data pairwise. The consistency index CI was calculated and a consistency test was performed. The result was valid when CI was < 0.1, otherwise the matrix was adjusted. The weights were calculated based on the square root method.

[0110] This application calculates weights based on the square root method and provides specific quantitative indicators for the importance of each monitoring data in assessing the comprehensive heart health status. These weights can help doctors or relevant personnel pay more targeted attention to different data in actual work and reasonably allocate resources and attention. If the weight of the electrocardiogram data is higher, more attention will be paid to the information provided by the electrocardiogram when analyzing the heart health status, making the evaluation process more operational and practical.

[0111] The weighted calculation formula in step S4 is:

[0112] Assume that the eigenvectors of electrocardiogram data, blood pressure data, and cardiac ultrasound data after vectorization are The corresponding weights are w ecg 、w bp 、w ech o , and satisfy w ecg +w bp +w ech o =1, comprehensive heart health data The calculation formula is:

[0113]

[0114] in is the calculated comprehensive heart health data value.

[0115] In step S5, an early warning system is built based on comprehensive heart health data, and health thresholds are established based on expert analysis methods;

[0116] During the hidden danger investigation stage, medical personnel will conduct preliminary assessments and give inspection recommendations. In complex cases, expert consultations will be organized to clarify the condition, develop personalized treatment plans, establish a tracking and follow-up mechanism, and monitor data regularly.

[0117] A patient health warning processing system for multiple heart monitoring data, the health warning processing system comprising:

[0118] a data acquisition and processing module configured to acquire heart monitoring data and pre-process it to construct a data set;

[0119] a feature extraction module configured to extract features in the data set and obtain feature data related to heart disease based on the big data;

[0120] The judgment module is configured to perform consistency judgment on two sets of data;

[0121] a comprehensive analysis module configured to assign weight values to the monitored data and perform weighted calculations on comprehensive heart health data;

[0122] The early warning module is configured to perform early warning processing based on health data results.

[0123] This application preprocesses cardiac monitoring data to remove noise and outlier interference factors in the data, improve the accuracy and reliability of the data, and provide a high-quality data foundation for subsequent analysis and processing. Constructing a data set helps to effectively organize and manage the data, making it easier for subsequent modules to call and process it.

[0124] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. Various changes and improvements are possible without departing from the spirit and scope of the present invention, and such changes and improvements fall within the scope of the invention as claimed.

Claims

1. A patient health warning processing method for multiple heart monitoring data, characterized in that: The steps for early warning processing are: S1. Obtain heart monitoring data, including electrocardiogram data, blood pressure data, and cardiac ultrasound data, and construct a data set after preprocessing the data; S2. Extract features from the data in the dataset and obtain feature data related to heart disease based on the big data. The relevant feature data includes electrocardiogram data, blood pressure data, and cardiac ultrasound data. S3. Vectorize the acquired features, compare the real-time monitored feature data with the feature data acquired from big data, and determine whether they are consistent; S4. Based on the consistent judgment results, weight values are assigned to the monitoring data respectively, and weighted calculation is performed to obtain comprehensive heart health data; S5. Based on the obtained comprehensive heart health data, a health threshold is established. When the health threshold is greater than the health threshold, a warning is issued to check for heart safety hazards.

2. The patient health warning processing method of multiple heart monitoring data according to claim 1 is characterized in that: In step S1, the heart monitoring data is obtained through routine hospital examinations, dynamic electrocardiogram monitoring, and wearable devices; Data preprocessing includes denoising, outlier removal and image enhancement; The preprocessed electrocardiogram data, blood pressure data and cardiac ultrasound data are integrated to construct a data set and establish a retrieval index.

3. The patient health warning processing method of multiple heart monitoring data according to claim 1 is characterized in that: In step S2, the ECG data feature extraction is performed by using a threshold detection method to identify the QRS complex, searching for P waves and T waves based on the position of the QRS complex, and the extracted features are amplitude and morphological features; Blood pressure data feature extraction is done by calculating the mean of systolic pressure, diastolic pressure, and mean arterial pressure to obtain mean features, and calculating the standard deviation and coefficient of variation of blood pressure to obtain fluctuation features; Cardiac ultrasound data features include morphological features and functional features.

4. The patient health warning processing method of multiple heart monitoring data according to claim 3 is characterized in that: The specific calculation and extraction steps of the features in step S2 are: The ECG signal is preprocessed, and a numerical standard is set as a threshold. When the value of the preprocessed ECG signal exceeds the threshold, it is determined to be a QRS complex; if it does not exceed the threshold, it is not a QRS complex. After determining the start and end positions of the QRS complex, the positions of the P wave and T wave are found within a specific time range before the start position of the QRS complex and within a specific time range after the end position of the QRS complex. Find the maximum value in the electrocardiogram signal parts corresponding to the P wave, QRS complex and T wave respectively. The maximum value is the amplitude of the corresponding wave. The morphological characteristics are described by the degree of inclination and curvature of the waveform; Mean feature extraction: add up all measured systolic blood pressure values and divide by the total number of measurements to obtain the mean systolic blood pressure value; For diastolic blood pressure mean extraction, all measured diastolic blood pressure values were added together and then divided by the total number of measurements to calculate the mean diastolic blood pressure; To extract the mean value of mean arterial pressure, add twice the diastolic pressure to the systolic pressure of each measurement and divide by 3 to obtain the mean arterial pressure value for each measurement; add all the mean arterial pressure values and divide by the number of measurements to obtain the mean arterial pressure. fluctuation The characteristic is to calculate the difference between each blood pressure measurement value and the corresponding blood pressure mean, square the differences, add them together, divide by the number of measurements, and take the square root of the result to obtain the standard deviation of the blood pressure, which is used to reflect the degree of dispersion of blood pressure fluctuations; The coefficient of variation is calculated by dividing the standard deviation of blood pressure by the corresponding mean blood pressure. The result is the coefficient of variation, which is used to measure the relative size of blood pressure fluctuations; The morphological characteristics are measured by ultrasound images to obtain the value of the left ventricular end-diastolic diameter each time. The values of multiple measurements are added together and divided by the number of measurements to obtain the average value of the left ventricular end-diastolic diameter, which is the morphological characteristic value. Functional characteristics: The ejection fraction was obtained by subtracting the left ventricular end-systolic volume from the left ventricular end-diastolic volume. The difference was divided by the left ventricular end-diastolic volume and the result was multiplied by 100% to reflect the cardiac systolic function.

5. The patient health warning processing method of multiple heart monitoring data according to claim 1 is characterized in that: The vectorization processing steps in step S3 are: After obtaining the electrocardiogram, blood pressure, and cardiac ultrasound features, vectorization processing is performed to integrate and classify different types of features, clarify the feature categories of the electrocardiogram, blood pressure, and cardiac ultrasound, perform feature encoding, convert features of different magnitudes into the same scale, and arrange the processed features in order to form the feature vector of each data.

6. The patient health warning processing method of multiple heart monitoring data according to claim 1 is characterized in that: The steps for determining whether consistency occurs in step S3 are: The two sets of vectors are calculated based on the cosine similarity method, and the feature vectors of the two types of data are compared. If the similarity index is higher than the set threshold, the data is considered consistent; otherwise, it is inconsistent. Based on the comparison results, it is determined whether the feature data obtained by real-time monitoring is consistent with that obtained by big data.

7. The patient health warning processing method of multiple heart monitoring data according to claim 1 is characterized in that: The weight value in step S4 is determined based on the analytic hierarchy process. Specifically, a hierarchical model was constructed, with the assessment of comprehensive heart health status as the target layer, data source reliability, clinical diagnostic importance, and data consistency as the criterion layer, and electrocardiogram, blood pressure, and cardiac ultrasound monitoring data as the solution layer. Based on the 1-9 scaling method, a judgment matrix was constructed by comparing the solution layer data pairwise. Calculate the consistency index CI and perform consistency test. The result is valid when CI is less than 0.1, otherwise adjust the matrix; calculate the weight based on the square root method.

8. The patient health warning processing method of multiple heart monitoring data according to claim 1 is characterized in that: The weighted calculation formula in step S4 is: Assume that the eigenvectors of electrocardiogram data, blood pressure data, and cardiac ultrasound data after vectorization are The corresponding weights are w ecg 、w bp 、w ech o , and satisfy w ecg +w bp +w ech o =1, comprehensive heart health data The calculation formula is: in is the calculated comprehensive heart health data value.

9. The patient health warning processing method of multiple heart monitoring data according to claim 1, characterized in that: In step S5, an early warning system is built based on comprehensive heart health data, and health thresholds are established based on expert analysis methods; During the hidden danger investigation stage, medical personnel will conduct preliminary assessments and give inspection recommendations. In complex cases, expert consultations will be organized to clarify the condition, develop personalized treatment plans, establish a tracking and follow-up mechanism, and monitor data regularly.

10. A patient health early warning processing system for multiple heart monitoring data, characterized in that: The health warning processing system includes: a data acquisition and processing module configured to acquire heart monitoring data and pre-process it to construct a data set; a feature extraction module configured to extract features within the data set and obtain feature data related to heart disease based on the big data; The judgment module is configured to perform consistency judgment on two sets of data; A comprehensive analysis module is configured to assign weight values to the monitored data and perform weighted calculations on comprehensive heart health data; The early warning module is configured to perform early warning processing based on health data results.

Images