A postoperative intelligent monitoring method and system

Abstract

The present invention discloses a postoperative intelligent monitoring method and system, which relates to the technical field of big data processing. By acquiring multi-dimensional heart-related information, a data group HRQ is formed and multi-step preprocessing is performed. By feature extraction, the RR interval standard deviation SDN, the RR interval root mean square deviation RMS and the QT period variability QTV are obtained. The sliding time window is combined to perform time series analysis, and the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR are generated to reflect the short-term and long-term fluctuation trends of the patient's heart. Finally, an arrhythmia risk assessment algorithm is constructed, and the arrhythmia risk index ARI is obtained through comprehensive calculation, and matched with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ, and the patient's physical state assessment strategy is automatically generated to remind medical staff and patients to take intervention measures in time, thereby improving the accuracy of rehabilitation progress assessment after cardiac surgery.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and in particular to a postoperative intelligent monitoring method and system. Background Art

[0002] Postoperative rehabilitation monitoring is an important research and application direction in the medical field, especially after cardiac surgery, the patient's rehabilitation monitoring needs are particularly critical. With the advancement of science and technology, intelligent medical monitoring systems have gradually become one of the key directions of clinical application. Such systems can not only collect patients' physiological data in real time, but also help doctors make more accurate diagnosis and treatment decisions by analyzing these data. After cardiac surgery, the patient's physical condition is complex and changeable. Monitoring key vital signs such as heart rate, blood pressure, blood oxygen and exercise data can effectively evaluate the patient's postoperative recovery.

[0003] In the invention of Chinese patent application number CN202310354952.2, an intelligent monitoring and management method and system for postoperative care training is disclosed, which relates to the technical field of postoperative care training. The method includes: obtaining a preset training data set for a first user; performing stage division and outputting a preset stage training data set; collecting a real-time training data set for the first user, and the real-time training data set includes a real-time muscle strength data set and a real-time sensor data set of equipment for postoperative training; generating a stage monitoring model; inputting the real-time muscle strength data set and the real-time sensor data set of the equipment into the stage monitoring model for analysis, and outputting the real-time stage completion degree; when the real-time stage completion degree is greater than the preset stage completion degree, sending a stage change instruction to the stage monitoring model, which solves the technical problems of insufficient monitoring accuracy and poor monitoring and management effect of postoperative care training for users in the prior art, and achieves the technical effect of improving the accuracy of monitoring and management.

[0004] It can be seen that at present, traditional postoperative rehabilitation monitoring methods mainly rely on a single physiological indicator or match based on historical training data, such as heart rate, blood pressure and muscle strength data. Although this method can reflect the patient's health status to a certain extent, it cannot fully capture the overall progress of postoperative rehabilitation. Especially after cardiac surgery, the patient's recovery process often involves multi-dimensional health factors, such as vital signs, daily activity level, emotional state, etc. If you rely solely on a single indicator, other important rehabilitation signals may be ignored, resulting in the doctor's assessment of the patient's recovery progress is not comprehensive enough, and it is impossible to accurately predict the complexity or potential risks of postoperative recovery. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a postoperative intelligent monitoring method and system, which solves the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a postoperative intelligent monitoring method, comprising:

[0007] Wearable devices, ECG monitors and electrocardiographs are used to obtain heart-related information of patients after surgery to form the collection data group HRQ;

[0008] Performing outlier removal preprocessing, signal filtering preprocessing and normalization preprocessing on the collected data group HRQ to form a processed collected data group HRQ;

[0009] Perform feature extraction on the processed data set HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability during QT, and obtain the RR interval standard deviation SDN, the RR interval root mean square deviation RMS and the QT period variability QTV;

[0010] By applying a sliding time window to perform time series analysis on the extracted RR interval standard deviation SDN, RR interval root mean square deviation RMS and QT period variability QTV, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are obtained to reflect the fluctuation trend of the patient's heart in the short and long term;

[0011] By constructing an arrhythmia risk assessment algorithm formula, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are substituted into the calculation to obtain the arrhythmia risk index ARI, and then matched with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ to obtain the patient's physical condition assessment strategy plan, and prompt and notify the patient and relevant medical staff.

[0012] Preferably, the use of a wearable device, an electrocardiogram monitor, and an electrocardiograph to obtain heart-related information of a patient after surgery to form a collection data group HRQ includes:

[0013] During the patient's postoperative recovery period, wearable devices, hospital ECG monitors and electrocardiographs are installed and configured, and multi-dimensional cardiac physiological parameters of the patient are collected by attaching them to the patient's skin, including systolic blood pressure SBP, diastolic blood pressure DBP, heart rate HR, stroke volume SV, RR interval RR and QT interval QT, forming the initial data set CSJ of the patient;

[0014] Among them, systolic blood pressure SBP and diastolic blood pressure DBP are collected and obtained by cuff-type sphygmomanometer and dynamic blood pressure monitor; stroke volume SV is collected and obtained by pulse wave analyzer; RR interval RR and QT interval QT are collected and obtained by electrocardiograph;

[0015] The initial data set CSJ is initially processed to obtain the mean arterial pressure MAP and cardiac output CO, and then reorganized and integrated with the initial data set CSJ to form the acquisition data set HRQ = {HR, RR, QT, MAP, CO};

[0016] The mean arterial pressure MAP is Obtain calculation formula;

[0017] The cardiac output CO is Obtain the calculation formula.

[0018] Preferably, the preprocessing of removing outliers, filtering signals and normalizing the collected data set HRQ to form the processed collected data set HRQ comprises:

[0019] The collected data group HRQ is subjected to outlier removal preprocessing and signal filtering preprocessing. The outlier removal preprocessing includes using the triple standard deviation method to remove abnormal data that exceeds the normal range; the signal filtering preprocessing includes using the Kalman filter to smooth the data and adjust the noise and short-term fluctuations;

[0020] Among them, the triple standard deviation method is used to eliminate abnormal data that exceeds the normal range. The specific elimination process is carried out through the following formula: ;

[0021] In the formula, HRQi represents the i-th parameter in the collected data set HRQ, represents the average value of the i-th parameter in the collected data set HRQ, Indicates 3 times the standard deviation of the i-th parameter in the collected data set HRQ ;

[0022] The standard deviation Obtained through the following calculation formula: ;

[0023] In the formula, n represents the total number of data samples;

[0024] The collected data group HRQ is normalized and preprocessed after outlier removal and signal filtering, including normalization using a Z-score normalization method, to form a processed collected data group HRQ.

[0025] Preferably, the feature extraction of the processed collected data group HRQ, obtaining the long-term and short-term fluctuation information of HRV and the variability during QT, obtaining the RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the variability QTV during QT includes:

[0026] Perform feature extraction on the processed data set HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability of QT during QT, and obtain the standard deviation SDN of the RR interval RR of the long-term fluctuation performance and the root mean square difference RMS of the RR interval RR of the short-term fluctuation performance;

[0027] The RR interval RR standard deviation SDN is obtained by the following calculation formula: ;

[0028] In the formula, n represents the total number of data samples, RRi represents the RR interval of the i-th sampling, represents the average value of RR interval;

[0029] Among them, the average value of RR interval RR Obtained through the following calculation formula: ;

[0030] The root mean square error RMS of the RR interval RR is obtained by the following calculation formula: .

[0031] Perform feature extraction on the processed collected data group HRQ to obtain the variability QTV during the QT interval QT;

[0032] The QT period variability QTV is obtained by the following calculation formula: ;

[0033] Where QTj represents the QT period QT of the i-th sampling, represents the QT interval QT mean;

[0034] Among them, QT interval QT average value Obtained through the following calculation formula: .

[0035] Preferably, the method of performing time series analysis on the extracted RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the QT period variability QTV by applying a sliding time window to obtain the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR to reflect the fluctuation trend of the patient's heart in the short and long term includes:

[0036] Set a sliding time window and divide the time series data into multiple continuous time windows to obtain the outstanding performance of short-term and long-term trends, where the multiple continuous time windows include the RR interval RR standard deviation SDN and the RR interval RR root mean square difference RMS of the short-term volatility performance, divide the time period, and obtain the multi-window data set W={W1, W2, ..., Wz}, Wz={SDNz, RMSz, QTVz};

[0037] Among them, SDNz={SDN1, SDN2, ..., SDNz}; RMSz={RMS1, RMS2, ..., RMSz}; QTVz{QTV1, QTV2, ..., QTVz}.

[0038] The standard deviation SDN of the RR interval RR in each time window and the root mean square difference RMS of the RR interval RR of the short-term fluctuation performance are calculated together with the change rate of the adjacent time windows to reflect the trend performance of short-term and long-term fluctuations, and obtain the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR;

[0039] The HRV change rate HVR is obtained by the following calculation formula: ;

[0040] In the formula, T represents the current observation time point, specifically the end time of the current time window, z represents the length of the time window, and SDNt represents the standard deviation SDN of the RR interval at time t;

[0041] The RMS change rate RVR is obtained by the following calculation formula: ;

[0042] Where RMSt represents the RMS rate of change RVR at time t;

[0043] The QTV change rate QTR is obtained by the following calculation formula: ;

[0044] Where QTVt represents the QTV change rate QTR at time t, Indicates the average value of QTV change rate QTR within the elapsed time window;

[0045] Among them, the average pass Formula calculation acquisition;

[0046] The long-term trend of heart rhythm fluctuations is analyzed by acquiring the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR, including HVR data trend set = {HVR1, HVR2, ..., HVRz}, RMS change rate RVR data trend set = {RVR1, RVR2, ..., RVRz}, QTV change rate QTR data trend set = {QTR1, QTR2, ..., QTRz}, and a trend graph of the transformation rate is generated, including a broken line trend graph and a multiple overlapping trend graph.

[0047] Preferably, the arrhythmia risk assessment algorithm formula is constructed, the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR are substituted into the calculation, the arrhythmia risk index ARI is obtained, and it is matched with the preset arrhythmia risk assessment upper limit threshold SZ and lower limit threshold XZ, the patient's physical condition assessment strategy is obtained, and prompts and notifies the patient and relevant medical staff to intervene, including:

[0048] By constructing an arrhythmia risk assessment algorithm formula, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are substituted into the calculation to obtain the arrhythmia risk index ARI of the fluctuation trend of the calculated change rate;

[0049] The arrhythmia risk index ARI is obtained by the following calculation formula: ; In the formula, , and Respectively represent the preset weight values ​​of HRV change rate HVR, RMS change rate RVR and QTV change rate QTR, the specific values ​​are set by the user, and , , , .

[0050] Match the arrhythmia risk index ARI with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ, obtain the patient's physical condition assessment strategy, and prompt and notify the patient and relevant medical staff to intervene;

[0051] The patient's physical condition assessment strategy is obtained by the following matching method:

[0052] When the arrhythmia risk index ARI is less than the lower threshold value XZ, the patient's physical condition assessment result is obtained as qualified, and a qualified strategy plan for the patient's physical condition is generated, including prompting that the heart condition is normal and following the doctor's advice;

[0053] When the lower threshold value XZ≤arrhythmia risk index ARI≤upper threshold value SZ, the patient's physical condition assessment result is obtained as unqualified, and a strategy plan for unqualified patient physical condition is generated. The short-term and long-term fluctuation trends of the current patient are associated with the patient information and sent to the task list of doctors and nurses to be rounded, and the task list is processed through SMS and pop-up windows on the interactive software page;

[0054] When the upper threshold value SZ is less than the arrhythmia risk index ARI, the patient's physical condition assessment result is obtained as a warning, and a patient physical condition warning strategy is generated, including a voice prompt to the patient's associated attending physician to handle it, and the patient's arrhythmia risk index ARI and fluctuation trend are simultaneously associated with the patient's information, and sent to the task list of doctors and nurses to be inspected, and the task list is pinned and marked as a warning, and then a pop-up window flashes through SMS and interactive software pages to prompt the processing of the task list.

[0055] A postoperative intelligent monitoring system, comprising a data acquisition module, a data preprocessing module, a feature extraction module, a fluctuation trend analysis module and a patient status assessment and decision-making module;

[0056] The data acquisition module obtains heart-related information of the patient after surgery by using a wearable device, an electrocardiogram monitor and an electrocardiograph to form a collection data group HRQ;

[0057] The data preprocessing module performs outlier removal preprocessing, signal filtering preprocessing and normalization preprocessing on the collected data group HRQ to form a processed collected data group HRQ;

[0058] The feature extraction module extracts features from the processed collected data group HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability during QT, and obtains the RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the variability QTV during QT;

[0059] The fluctuation trend analysis module performs time series analysis on the extracted RR interval standard deviation SDN, RR interval root mean square deviation RMS and QT period variability QTV by applying a sliding time window to obtain the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR to reflect the fluctuation trend of the patient's heart in the short and long term;

[0060] The patient status assessment and decision-making module constructs an arrhythmia risk assessment algorithm formula, substitutes the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR into the calculation, obtains the arrhythmia risk index ARI, and matches it with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ, obtains the patient's physical status assessment strategy plan, and prompts and notifies the patient and relevant medical staff to intervene.

[0061] The present invention provides a postoperative intelligent monitoring method and system, which has the following beneficial effects:

[0062] The present invention effectively solves the shortcomings of the existing single indicator evaluation in postoperative rehabilitation monitoring, and provides a more comprehensive monitoring of heart health status. Multi-dimensional heart-related information is obtained through wearable devices and medical devices, and the data group HRQ is composed and multi-step preprocessing is performed to remove outliers, filter signals and normalize them to ensure the accuracy and consistency of the data. Then, through feature extraction, the RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the QT period variability QTV are obtained, and time series analysis is performed in combination with the sliding time window to generate the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR, which reflect the short-term and long-term fluctuation trends of the patient's heart. Finally, an arrhythmia risk assessment algorithm was constructed, and the arrhythmia risk index ARI was obtained through comprehensive calculation. It was matched with the preset upper and lower thresholds SZ and XZ of the arrhythmia risk assessment, and the patient's physical condition assessment strategy was automatically generated to remind medical staff and patients to take intervention measures in time, thereby improving the accuracy of the assessment of rehabilitation progress after cardiac surgery. It can also provide early warning before abnormal heart rhythm fluctuations occur, reducing the risk of complications such as arrhythmias, and ensuring that the patient's postoperative rehabilitation process is safer, timely and effective.

[0063] The ability to capture the heart rhythm fluctuation trend is effectively improved by setting a sliding time window and constructing a multi-window data set. In this method, the RR interval standard deviation SDN, the RR interval root mean square deviation RMS and the QT period variability QTV are divided into multiple continuous time windows, and the short-term and long-term fluctuations in each window are analyzed to generate the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR. This processing method ensures the dual measurement of the short-term sensitivity and long-term stability of the heart rhythm fluctuation trend, and can dynamically reflect the subtle changes of the patient's heart during postoperative rehabilitation. At the same time, by generating a broken line trend chart and multiple overlapping trend charts, doctors can more intuitively observe the long-term change trend of various heart indicators, providing the ability to predict the risk of arrhythmia in advance. Compared with traditional methods, this technology can more accurately predict the potential risks and recovery of the heart rhythm fluctuation, especially suitable for capturing subtle fluctuations in long-term observation, reducing risk signals that may be ignored, and improving the safety and predictability of postoperative rehabilitation.

[0064] By constructing an arrhythmia risk assessment algorithm, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are substituted into the calculation to generate the arrhythmia risk index ARI, and matched with the preset arrhythmia risk assessment upper and lower thresholds SZ and XZ to provide a patient's physical condition assessment strategy. According to different risk levels, the system can not only determine whether the patient's heart condition is qualified or not, but also automatically trigger an early warning in high-risk situations to notify medical staff to intervene in time. This process is real-time prompts through text messages, voice, and interactive software page pop-ups, ensuring that doctors and nurses can quickly obtain patients' risk information and take necessary treatment measures. Compared with traditional passive monitoring, this method is more proactive and intelligent, and can dynamically track the fluctuation trend of patients' heart conditions, and provide personalized risk intervention strategies to minimize the incidence of sudden arrhythmias. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of the steps of a postoperative intelligent monitoring method of the present invention;

[0066] Figure 2 The present invention is a block diagram of a postoperative intelligent monitoring system. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0068] Example 1: The present invention provides a postoperative intelligent monitoring method, please refer to Figure 1 , including the following steps:

[0069] S1: Use wearable devices, ECG monitors and electrocardiographs to obtain heart-related information of patients after surgery, forming the collection data group HRQ;

[0070] S2: performing outlier removal preprocessing, signal filtering preprocessing and normalization preprocessing on the collected data group HRQ to form a processed collected data group HRQ;

[0071] S3: extract features from the processed data set HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability during QT, and obtain the RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the variability QTV during QT;

[0072] S4: By applying a sliding time window, the extracted RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the QT period variability QTV are analyzed in time series to obtain the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR to reflect the fluctuation trend of the patient's heart in the short and long term;

[0073] S5: By constructing an arrhythmia risk assessment algorithm formula, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are substituted into the calculation to obtain the arrhythmia risk index ARI, and then matched with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ to obtain the patient's physical condition assessment strategy plan, and prompt and notify the patient and relevant medical staff to intervene.

[0074] In this embodiment, through S1 to S5, the shortcomings of single indicator evaluation in existing postoperative rehabilitation monitoring are effectively solved, and a more comprehensive monitoring of heart health status is provided. Multi-dimensional heart-related information is obtained through wearable devices and medical devices, and the data group HRQ is composed and multi-step preprocessing is performed to remove outliers, filter signals and normalize them to ensure the accuracy and consistency of the data. Then, through feature extraction, the RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the QT period variability QTV are obtained, and time series analysis is performed in combination with the sliding time window to generate the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR, which reflect the short-term and long-term fluctuation trends of the patient's heart. Finally, an arrhythmia risk assessment algorithm was constructed, and the arrhythmia risk index ARI was obtained through comprehensive calculation. It was matched with the preset upper and lower thresholds SZ and XZ of the arrhythmia risk assessment, and the patient's physical condition assessment strategy was automatically generated to remind medical staff and patients to take intervention measures in time, thereby improving the accuracy of the assessment of rehabilitation progress after cardiac surgery. It can also provide early warning before abnormal heart rhythm fluctuations occur, reducing the risk of complications such as arrhythmias, and ensuring that the patient's postoperative rehabilitation process is safer, timely and effective.

[0075] Example 2: This example is an explanation of Example 1. Figure 1 ,Specifically, S1 includes S11 and S12;

[0076] S11: During the patient's postoperative recovery period, install and configure wearable devices, hospital ECG monitors and electrocardiographs, and collect multi-dimensional cardiac physiological parameters of the patient by fitting them to the patient's skin, including systolic blood pressure SBP, diastolic blood pressure DBP, heart rate HR, stroke volume SV, RR interval RR and QT interval QT, to form the initial data set CSJ of the patient;

[0077] Among them, systolic blood pressure SBP and diastolic blood pressure DBP are collected and obtained by cuff-type sphygmomanometer and dynamic blood pressure monitor; stroke volume SV is collected and obtained by pulse wave analyzer; RR interval RR and QT interval QT are collected and obtained by electrocardiograph;

[0078] S12: Perform initial processing on the initial data set CSJ to obtain the mean arterial pressure MAP and cardiac output CO, and then reorganize and integrate them with the initial data set CSJ to form a collection data set HRQ={HR, RR, QT, MAP, CO};

[0079] Mean arterial pressure MAP Obtain calculation formula;

[0080] Cardiac output CO Obtain the calculation formula.

[0081] S2 includes S21 and S22;

[0082] S21: performing outlier removal preprocessing and signal filtering preprocessing on the collected data group HRQ. The outlier removal preprocessing includes using the triple standard deviation method to remove abnormal data that exceeds the normal range; the signal filtering preprocessing includes using the Kalman filter to smooth the data and adjust the noise and short-term fluctuations;

[0083] Among them, the triple standard deviation method is used to eliminate abnormal data that exceeds the normal range. The specific elimination process is carried out through the following formula: ;

[0084] In the formula, HRQi represents the i-th parameter in the collected data set HRQ, represents the average value of the i-th parameter in the collected data set HRQ, Indicates 3 times the standard deviation of the i-th parameter in the collected data set HRQ ;

[0085] The standard deviation Obtained through the following calculation formula: ;

[0086] In the formula, n represents the total number of data samples;

[0087] S22: performing normalization preprocessing on the collected data group HRQ after the outlier removal preprocessing and the signal filtering preprocessing, including performing normalization processing using a Z-score normalization method, to form a processed collected data group HRQ.

[0088] S3 includes S31 and S32;

[0089] S31: extracting features from the processed data set HRQ, obtaining the long-term and short-term fluctuation information of HRV and the variability of QT during QT, obtaining the standard deviation SDN of the RR interval RR of the long-term fluctuation performance and the root mean square difference RMS of the RR interval RR of the short-term fluctuation performance;

[0090] The RR interval RR standard deviation SDN is obtained by the following calculation formula: ;

[0091] In the formula, n represents the total number of data samples, RRi represents the RR interval of the i-th sampling, represents the average value of RR interval;

[0092] Among them, the average value of RR interval RR Obtained through the following calculation formula: ;

[0093] The root mean square error RMS of the RR interval RR is obtained by the following calculation formula: .

[0094] S32: extracting features from the processed collected data group HRQ to obtain the variability QTV during the QT interval QT;

[0095] The QT period variability QTV is obtained by the following calculation formula: ;

[0096] Where QTj represents the QT period QT of the i-th sampling, represents the QT interval QT mean;

[0097] Among them, QT interval QT average value Obtained through the following calculation formula: .

[0098] In this embodiment, wearable devices and a variety of medical monitoring instruments are used to collect multi-dimensional physiological parameters of the heart, ensuring the comprehensiveness and reliability of the data source. Then, through a variety of pre-processing techniques such as triple standard deviation method, Kalman filter and smoothing, abnormal data and noise are removed to ensure the accuracy of the data. Furthermore, combined with Z-score normalization and feature extraction technology, the key cardiac features of long-term and short-term fluctuations of RR interval RR and QT interval QT are obtained, further improving the depth of data analysis. Through these steps, the short-term and long-term fluctuations of the patient's heart can be captured more carefully, especially in the complex rehabilitation process, with higher sensitivity to potential abnormal fluctuations. Compared with traditional single monitoring methods, this method can provide doctors with a more multi-dimensional evaluation basis while ensuring data quality, thereby improving the accuracy and effectiveness of postoperative cardiac health management. This multi-dimensional, high-precision monitoring and analysis greatly reduces the omission of postoperative cardiac risks and provides scientific support for personalized rehabilitation interventions.

[0099] Example 3: This example is an explanation of Example 2. Figure 1 ,Specifically, S4 includes S41, S42 and S43;

[0100] S41: Set a sliding time window and divide the time series data into multiple continuous time windows to obtain the outstanding performance of short-term and long-term trends, wherein the multiple continuous time windows include the RR interval RR standard deviation SDN and the RR interval RR root mean square difference RMS of the short-term volatility performance, divide the time period, and obtain a multi-window data set W={W1, W2, ..., Wz}, Wz={SDNz, RMSz, QTVz};

[0101] Among them, SDNz={SDN1, SDN2, ..., SDNz}; RMSz={RMS1, RMS2, ..., RMSz}; QTVz{QTV1, QTV2, ..., QTVz}.

[0102] S42: Calculate the RR standard deviation SDN of the RR interval RR in each time window and the root mean square difference RMS of the RR interval RR of the short-term fluctuation performance and the change rate of the adjacent time windows to reflect the trend performance of short-term and long-term fluctuations, and obtain the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR;

[0103] The HRV change rate HVR is obtained by the following calculation formula: ;

[0104] In the formula, T represents the current observation time point, specifically the end time of the current time window, z represents the length of the time window, and SDNt represents the standard deviation SDN of the RR interval at time t;

[0105] The RMS change rate RVR is obtained by the following calculation formula: ;

[0106] Where RMSt represents the RMS rate of change RVR at time t;

[0107] The QTV change rate QTR is obtained by the following calculation formula: ;

[0108] Where QTVt represents the QTV change rate QTR at time t, Indicates the average value of QTV change rate QTR within the elapsed time window;

[0109] Among them, the average pass Formula calculation acquisition;

[0110] S43: Analyze the long-term trend of heart rhythm fluctuations by acquiring the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR, including HVR data trend set = {HVR1, HVR2, ..., HVRz}, RMS change rate RVR data trend set = {RVR1, RVR2, ..., RVRz}, QTV change rate QTR data trend set = {QTR1, QTR2, ..., QTRz}, and generate trend graphs of the transformation rate, including generating line trend graphs and multiple overlapping trend graphs.

[0111] In this embodiment, the ability to capture the heart rhythm fluctuation trend is effectively improved by setting a sliding time window and constructing a multi-window data set. In the method, the RR interval standard deviation SDN, the root mean square error RMS of the RR interval and the QT period variability QTV are divided into multiple continuous time windows, and the short-term and long-term fluctuations in each window are analyzed to generate the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR. This processing method ensures the dual measurement of the short-term sensitivity and long-term stability of the heart rhythm fluctuation trend, and can dynamically reflect the subtle changes of the patient's heart during postoperative rehabilitation. At the same time, by generating a broken line trend chart and multiple overlapping trend charts, doctors can more intuitively observe the long-term change trends of various heart indicators, providing the ability to predict the risk of arrhythmia in advance. Compared with traditional methods, this technology can more accurately predict the potential risks and recovery of the heart rhythm fluctuation, especially suitable for capturing subtle fluctuations in long-term observation, reducing risk signals that may be ignored, and improving the safety and predictability of postoperative rehabilitation.

[0112] Example 4: This example is an explanation of Example 3. Figure 1,Specifically, S5 includes S51 and S52;

[0113] S51: constructing an arrhythmia risk assessment algorithm formula, substituting the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR into the calculation, and obtaining the arrhythmia risk index ARI of the fluctuation trend of the calculated change rate;

[0114] The arrhythmia risk index (ARI) is calculated using the following formula: ; In the formula, , and Respectively represent the preset weight values ​​of HRV change rate HVR, RMS change rate RVR and QTV change rate QTR, the specific values ​​are set by the user, and , , , .

[0115] S52: Matching the arrhythmia risk index ARI with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ, obtaining the patient's physical condition assessment strategy, and prompting and notifying the patient and relevant medical staff to intervene;

[0116] The patient's physical status assessment strategy is obtained through the following matching methods:

[0117] When the arrhythmia risk index ARI is less than the lower threshold value XZ, the patient's physical condition assessment result is obtained as qualified, and a qualified strategy plan for the patient's physical condition is generated, including prompting that the heart condition is normal and following the doctor's advice;

[0118] When the lower threshold value XZ≤arrhythmia risk index ARI≤upper threshold value SZ, the patient's physical condition assessment result is obtained as unqualified, and a strategy plan for unqualified patient physical condition is generated. The short-term and long-term fluctuation trends of the current patient are associated with the patient information and sent to the task list of doctors and nurses to be rounded, and the task list is processed through SMS and pop-up windows on the interactive software page;

[0119] When the upper threshold value SZ is less than the arrhythmia risk index ARI, the patient's physical condition assessment result is obtained as a warning, and a patient physical condition warning strategy is generated, including a voice prompt to the patient's associated attending physician to handle it, and the patient's arrhythmia risk index ARI and fluctuation trend are simultaneously associated with the patient's information, and sent to the task list of doctors and nurses to be inspected, and the task list is pinned and marked as a warning, and then a pop-up window flashes through SMS and interactive software pages to prompt the processing of the task list.

[0120] In this embodiment, by constructing an arrhythmia risk assessment algorithm, the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR are substituted into the calculation to generate the arrhythmia risk index ARI, and matched with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ to provide a patient's physical condition assessment strategy. According to different risk levels, the system can not only judge whether the patient's heart state is qualified or not, but also automatically trigger an early warning in high-risk situations to notify medical staff to intervene in time. This process is carried out in real time through text messages, voice and interactive software page pop-up windows and other methods to ensure that doctors and nurses can quickly obtain the patient's risk information and take necessary treatment measures. Compared with traditional passive monitoring, this method is more proactive and intelligent, can dynamically track the fluctuation trend of the patient's heart state, and provide personalized risk intervention strategies to minimize the incidence of sudden arrhythmias.

[0121] Example 5: A postoperative intelligent monitoring system, please refer to Figure 2 ,Specifically, it includes data acquisition module, data preprocessing module, ,feature extraction module, fluctuation trend analysis module and patient ,status assessment and decision module;

[0122] The data collection module uses wearable devices, ECG monitors and electrocardiographs to obtain heart-related information of patients after surgery, forming a collection data group HRQ;

[0123] The data preprocessing module performs outlier removal preprocessing, signal filtering preprocessing and normalization preprocessing on the collected data group HRQ to form a processed collected data group HRQ;

[0124] The feature extraction module extracts features from the processed data set HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability during QT, and obtains the RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the variability QTV during QT;

[0125] The fluctuation trend analysis module uses a sliding time window to perform time series analysis on the extracted RR interval standard deviation SDN, RR interval root mean square deviation RMS and QT period variability QTV, and obtains the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR to reflect the fluctuation trend of the patient's heart in the short and long term;

[0126] The patient status assessment and decision-making module constructs an arrhythmia risk assessment algorithm formula, substitutes the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR into the calculation, obtains the arrhythmia risk index ARI, and matches it with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ to obtain the patient's physical status assessment strategy plan, and prompts and notifies the patient and relevant medical staff to intervene.

[0127] 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 postoperative intelligent monitoring method, characterized in that: include: Use wearable devices, ECG monitors and electrocardiographs to obtain heart-related information from patients after surgery to form the collection data group HRQ; Performing outlier removal preprocessing, signal filtering preprocessing and normalization preprocessing on the collected data group HRQ to form a processed collected data group HRQ; Perform feature extraction on the processed data set HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability during QT, and obtain the RR interval standard deviation SDN, the RR interval root mean square deviation RMS and the QT period variability QTV; By applying a sliding time window to perform time series analysis on the extracted RR interval standard deviation SDN, RR interval root mean square deviation RMS and QT period variability QTV, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are obtained to reflect the fluctuation trend of the patient's heart in the short and long term; By constructing an arrhythmia risk assessment algorithm formula, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are substituted into the calculation to obtain the arrhythmia risk index ARI, which is then matched with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ to obtain the patient's physical condition assessment strategy plan, and prompt and notify the patient and relevant medical staff to intervene.

2. A postoperative intelligent monitoring method according to claim 1, characterized in that: The wearable device, the electrocardiogram monitor and the electrocardiograph are used to obtain the heart-related information of the patient after surgery, and the collected data group HRQ includes: During the patient's postoperative recovery period, wearable devices, hospital ECG monitors and electrocardiographs are installed and configured, and multi-dimensional cardiac physiological parameters of the patient are collected by attaching them to the patient's skin, including systolic blood pressure SBP, diastolic blood pressure DBP, heart rate HR, stroke volume SV, RR interval RR and QT interval QT, forming the initial data set CSJ of the patient; Among them, systolic blood pressure SBP and diastolic blood pressure DBP are collected and obtained by cuff-type sphygmomanometer and dynamic blood pressure monitor; stroke volume SV is collected and obtained by pulse wave analyzer; RR interval RR and QT interval QT are collected and obtained by electrocardiograph; The initial data set CSJ is initially processed to obtain the mean arterial pressure MAP and cardiac output CO, and then reorganized and integrated with the initial data set CSJ to form the acquisition data set HRQ = {HR, RR, QT, MAP, CO}; The mean arterial pressure MAP is Obtain calculation formula; The cardiac output CO is Obtain the calculation formula.

3. The postoperative intelligent monitoring method according to claim 1, characterized in that: The collected data group HRQ is subjected to preprocessing for removing outliers, preprocessing for signal filtering and preprocessing for normalization to form the processed collected data group HRQ, which includes: The collected data group HRQ is subjected to outlier removal preprocessing and signal filtering preprocessing. The outlier removal preprocessing includes using the triple standard deviation method to remove abnormal data that exceeds the normal range; the signal filtering preprocessing includes using the Kalman filter to smooth the data and adjust the noise and short-term fluctuations; Among them, the triple standard deviation method is used to eliminate abnormal data that exceeds the normal range. The specific elimination process is carried out through the following formula: ; In the formula, HRQi represents the i-th parameter in the collected data set HRQ, represents the average value of the i-th parameter in the collected data set HRQ, Indicates 3 times the standard deviation of the i-th parameter in the collected data set HRQ ; The standard deviation Obtained through the following calculation formula: ; In the formula, n represents the total number of data samples; The collected data group HRQ is normalized and preprocessed after outlier removal and signal filtering, including normalization using a Z-score normalization method, to form a processed collected data group HRQ.

4. The postoperative intelligent monitoring method according to claim 1, characterized in that: The feature extraction of the processed collected data group HRQ, obtaining the long-term and short-term fluctuation information of HRV and the variability during QT, and obtaining the RR interval standard deviation SDN, the RR interval root mean square deviation RMS and the QT period variability QTV include: Perform feature extraction on the processed data set HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability of QT during QT, and obtain the standard deviation SDN of the RR interval RR of the long-term fluctuation performance and the root mean square difference RMS of the RR interval RR of the short-term fluctuation performance; The RR interval RR standard deviation SDN is obtained by the following calculation formula: ; In the formula, n represents the total number of data samples, RRi represents the RR interval of the i-th sampling, represents the average value of RR interval; Among them, the average value of RR interval RR Obtained through the following calculation formula: ; The root mean square difference RMS of the RR interval RR is obtained by the following calculation formula: ; Perform feature extraction on the processed data set HRQ to obtain the variability QTV during the QT interval QT; The QT period variability QTV is obtained by the following calculation formula: ; In the formula, QTj represents the QT period QT of the i-th sampling, represents the QT interval QT mean; Among them, QT interval QT average value Obtained through the following calculation formula: .

5. The postoperative intelligent monitoring method according to claim 1, characterized in that: The method of applying a sliding time window to perform time series analysis on the extracted RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval, and the QT period variability QTV to obtain the HRV change rate HVR, the RMS change rate RVR, and the QTV change rate QTR to reflect the fluctuation trend of the patient's heart in the short and long term includes: Set a sliding time window and divide the time series data into multiple continuous time windows to obtain the outstanding performance of short-term and long-term trends, where the multiple continuous time windows include the RR interval RR standard deviation SDN and the RR interval RR root mean square difference RMS of the short-term volatility performance, divide the time period, and obtain the multi-window data set W={W1, W2, ..., Wz}, Wz={SDNz, RMSz, QTVz}; Among them, SDNz={SDN1, SDN2, ..., SDNz}; RMSz={RMS1, RMS2, ..., RMSz}; QTVz{QTV1, QTV2, ..., QTVz}; The standard deviation SDN of the RR interval RR in each time window and the root mean square difference RMS of the RR interval RR of the short-term fluctuation performance are calculated together with the change rate of the adjacent time windows to reflect the trend performance of short-term and long-term fluctuations, and obtain the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR; The HRV change rate HVR is obtained by the following calculation formula: ; Where T represents the current observation time point, specifically the end time of the current time window, z represents the length of the time window, and SDNt represents the standard deviation SDN of the RR interval at time t; The RMS change rate RVR is obtained by the following calculation formula: ;Wherein, RMSt represents the RMS rate of change RVR at time t; The QTV change rate QTR is obtained by the following calculation formula: ; In the formula, QTVt represents the QTV change rate QTR at time t, Indicates the average value of QTV change rate QTR within the elapsed time window; Among them, the average pass Formula calculation acquisition; The long-term trend of heart rhythm fluctuations is analyzed by acquiring the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR, including HVR data trend set = {HVR1, HVR2, ..., HVRz}, RMS change rate RVR data trend set = {RVR1, RVR2, ..., RVRz}, QTV change rate QTR data trend set = {QTR1, QTR2, ..., QTRz}, and a trend graph of the transformation rate is generated, including a broken line trend graph and a multiple overlapping trend graph.

6. The postoperative intelligent monitoring method according to claim 1, characterized in that: The arrhythmia risk assessment algorithm formula is constructed, and the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR are substituted into the calculation to obtain the arrhythmia risk index ARI, and matched with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ, to obtain the patient's physical condition assessment strategy, and prompt and notify the patient and relevant medical staff to intervene, including: By constructing an arrhythmia risk assessment algorithm formula, the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR are substituted into the calculation to obtain the arrhythmia risk index ARI of the fluctuation trend of the calculated change rate; The arrhythmia risk index ARI is obtained by the following calculation formula: ; In the formula, , and Respectively represent the preset weight values ​​of HRV change rate HVR, RMS change rate RVR and QTV change rate QTR, the specific values ​​are set by the user, and , , , ; Match the arrhythmia risk index ARI with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ, obtain the patient's physical condition assessment strategy, and prompt and notify the patient and relevant medical staff to intervene; The patient's physical condition assessment strategy is obtained by the following matching method: When the arrhythmia risk index ARI is less than the lower threshold value XZ, the patient's physical condition assessment result is obtained as qualified, and a qualified strategy plan for the patient's physical condition is generated, including prompting that the heart condition is normal and following the doctor's advice; When the lower threshold value XZ≤arrhythmia risk index ARI≤upper threshold value SZ, the patient's physical condition assessment result is obtained as unqualified, and a strategy plan for unqualified patient physical condition is generated. The short-term and long-term fluctuation trends of the current patient are associated with the patient information and sent to the task list of doctors and nurses to be rounded, and the task list is processed through SMS and pop-up windows on the interactive software page; When the upper threshold value SZ is less than the arrhythmia risk index ARI, the patient's physical condition assessment result is obtained as a warning, and a patient physical condition warning strategy is generated, including a voice prompt to the patient's associated attending physician to handle it, and the patient's arrhythmia risk index ARI and fluctuation trend are simultaneously associated with the patient's information, and sent to the task list of doctors and nurses to be inspected, and the task list is pinned and marked as a warning, and then a pop-up window flashes through SMS and interactive software pages to prompt the processing of the task list.

7. A postoperative intelligent monitoring system, used to implement a postoperative intelligent monitoring method according to any one of claims 1 to 6, characterized in that: It includes data acquisition module, data preprocessing module, feature extraction module, fluctuation trend analysis module and patient status assessment and decision-making module; The data acquisition module obtains heart-related information of the patient after surgery by using a wearable device, an electrocardiogram monitor and an electrocardiograph to form a collection data group HRQ; The data preprocessing module performs outlier removal preprocessing, signal filtering preprocessing and normalization preprocessing on the collected data group HRQ to form a processed collected data group HRQ; The feature extraction module extracts features from the processed collected data group HRQ to obtain the long-term and short-term fluctuation information of HRV and the variability during QT, and obtains the RR interval standard deviation SDN, the root mean square deviation RMS of the RR interval and the variability QTV during QT; The fluctuation trend analysis module performs time series analysis on the extracted RR interval standard deviation SDN, RR interval root mean square deviation RMS and QT period variability QTV by applying a sliding time window to obtain the HRV change rate HVR, RMS change rate RVR and QTV change rate QTR to reflect the fluctuation trend of the patient's heart in the short and long term; The patient status assessment and decision-making module constructs an arrhythmia risk assessment algorithm formula, substitutes the HRV change rate HVR, the RMS change rate RVR and the QTV change rate QTR into the calculation, obtains the arrhythmia risk index ARI, and matches it with the preset arrhythmia risk assessment upper threshold SZ and lower threshold XZ, obtains the patient's physical status assessment strategy plan, and prompts and notifies the patient and relevant medical staff to intervene.

Images