Anesthesia effect evaluation system based on big data

Abstract

The invention relates to the technical field of anesthesia effect evaluation, in particular to an anesthesia effect evaluation system based on big data, which comprises a physiological data processing module, a patient grouping and analysis module, an anesthesia risk evaluation module and an anesthesia adjustment support module. And collecting physiological indexes and types and doses of anesthetic drugs. According to the method, the real-time processing capacity of physiological data is enhanced by integrating historical health records and operation information, accurate evaluation of the instant effect of anesthetic is provided by deeply analyzing time sequence changes of heart rate and blood pressure, the response of different patient groups to the anesthetic can be recognized through grouping analysis, and the accuracy of the anesthetic is improved. Therefore, a targeted anesthesia scheme is realized, the anesthetic injection time and dosage are analyzed by using the LSTM network, the dynamic prediction of drug response is optimized, a doctor is supported to adjust the anesthetic dosage according to real-time data, the precision and safety of anesthesia management are greatly improved, and the anesthesia side effect risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of anesthesia effect evaluation, and in particular to an anesthesia effect evaluation system based on big data. Background Art

[0002] Anesthetic effect assessment technology involves monitoring and analyzing the effects of anesthetic drugs on patients to ensure the safety and comfort of patients during surgery or treatment. This field includes monitoring of multiple physiological parameters (such as level of consciousness, pain perception, vital signs, etc.), and using these data to evaluate the effectiveness and suitability of anesthetics, thereby optimizing anesthesia management.

[0003] Among them, the big data anesthesia effect evaluation system is a technology that uses large-scale data sets to analyze and predict the effects of anesthetic drugs. It collects and analyzes a large amount of data generated by patients during anesthesia (such as vital signs, drug dosage, surgical type, etc.), and uses data mining and machine learning techniques to predict the effects of anesthetic drugs and patients' responses to specific anesthetics. Its main purpose is to assist anesthesiologists in adjusting the anesthesia dosage in real time, improve surgical safety, and reduce anesthesia-related risks and side effects.

[0004] Although existing technologies can monitor basic vital signs, they lack the ability to conduct real-time and dynamic analysis of the rich data generated during the operation. Traditional technologies fail to achieve efficient data grouping and classification, resulting in the inability to adjust the anesthetic dose according to the patient's immediate response, limiting the accurate assessment of the effects of anesthetic drugs, increasing surgical risks and anesthetic side effects, which is particularly evident in complex or lengthy operations, resulting in difficulty and time-consuming recovery from anesthesia after surgery, affecting the overall effectiveness of the operation. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and propose an anesthesia effect evaluation system based on big data.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: an anesthetic effect evaluation system based on big data, the system comprising: The physiological data processing module collects physiological indicators, types and dosages of anesthetic drugs based on the patient's historical health records and surgical information, processes heart rate change and blood pressure change data, and analyzes time series changes to obtain an overview of physiological data; The patient grouping and analysis module analyzes the distribution characteristics of the patient data based on the physiological data overview, groups the patients according to the change trends of the heart rate and blood pressure data, analyzes the physiological response patterns of the different groups, and reclassifies the patient groups to obtain group classification indicators; The anesthesia risk assessment module uses the LSTM network to perform time series data analysis on the injection time and dosage of anesthetics based on the group classification indicators, calculates the response intensity of different patient groups to anesthetics, assesses the patient's anesthesia risk index, and compares it with the real-time monitoring data during the operation to obtain an anesthesia risk update record; The anesthesia adjustment support module evaluates the real-time effect of the current anesthesia based on the anesthesia risk update record and the physiological data collected in real time during the operation, and updates the adjustment suggestions for the anesthetic drug dosage to obtain the anesthesia management results.

[0007] The present invention is improved in that the processing steps of the heart rate change and blood pressure change data are specifically as follows: Based on the patient's historical health records and surgical information, physiological indicators, anesthetic drug types and dosage data were collected to obtain a preliminary physiological data set; Screening and preliminarily processing the heart rate and blood pressure data in the preliminary physiological data set, removing data points that deviate from a normal range, and obtaining screened heart rate and blood pressure data; The changing trends in the screened heart rate and blood pressure data were analyzed using the formula: ; Calculating heart rate variability ,in, Indicates heart rate data points, Represents the average value of heart rate data, is the total number of data points.

[0008] The present invention is improved in that the steps of obtaining the physiological data overview are specifically as follows: Analyzing the processed heart rate change and blood pressure change data, using an autoregressive model or a moving average model to identify long-term trends and periodic changes in the data, and obtaining time series analysis results; Based on the time series analysis results, key statistical indicators are extracted, including mean value, standard deviation, peak value, and stability or volatility indicators of heart rate and blood pressure, to obtain an overview of physiological data.

[0009] The present invention is improved in that the steps of analyzing the distribution characteristics of the patient data are specifically as follows: Based on the physiological data overview, the mean, median, standard deviation and skewness of each data are calculated to obtain preliminary statistical analysis results; The preliminary statistical analysis results were analyzed using the formula: ; Calculate the standard score for each data point, identify outliers in the data, and obtain the distribution characteristics of the filtered heart rate and blood pressure data, where Indicates The standard score of the data points, Represents heart rate or blood pressure data points, represents the average of the data points, represents the standard deviation of the data points; Based on the distribution characteristics of the screened heart rate and blood pressure data, cluster analysis is performed to identify different patient groups, evaluate the distribution consistency and differences among the patient groups, and obtain distribution characteristic analysis results.

[0010] The present invention is improved in that the step of obtaining the group classification index is specifically as follows: According to the change trends of the heart rate and blood pressure data, K-means clustering is used to classify the patients into different groups, and preliminary patient grouping results are obtained; Based on the preliminary patient grouping results, the formula was used: ; Evaluate the differences between multiple groups and obtain the results of physiological response pattern analysis, among which, is the within-group variance, Indicates the first heart rate or blood pressure value for each data point, is the mean value of the corresponding group data points, is the weight of each data point, defined as , is the sum of weights, is the number of data points in the population; Based on the physiological response pattern analysis results, the classification of the patient group is re-evaluated, the data differences within and between groups are checked, and the group classification indicators are obtained.

[0011] The present invention is improved in that the step of calculating the intensity of the reaction to the anesthetic is specifically: Based on the group classification indicators, the LSTM network is used to extract the time series data of the patient's anesthetic injection time and dosage for analysis, identify the patterns and trends in the data, and obtain the time series data features for preliminary analysis; Based on the time series data characteristics of the preliminary analysis, the formula is used: ; Calculate the intensity of response to anesthesia in different patient groups ,in, It is The anesthetic injection dose for each data point, Indicates The injection time of each data point, is the attenuation factor, which indicates the effect of time on the anesthetic effect. is the total number of data points, is the base of natural logarithms.

[0012] The present invention is improved in that the steps of obtaining the anesthesia risk update record are specifically as follows: Based on the intensity of the reaction to the anesthetic, analyzing the patient's anesthetic reaction characteristics during the operation to obtain anesthetic reaction data during the operation; Based on the anesthesia response data during the operation, the data is compared with the real-time monitoring data, and the change of anesthesia risk is determined by calculating the difference and trend analysis, using the formula: ; Get an updated record of anesthesia risk, where: is the total difference between the current and predicted anesthetic response intensity, Representative The intensity of the monitored anesthetic response for each data point, Indicates The predicted anesthetic response intensity for each data point, is the total number of data points.

[0013] The present invention is improved in that the steps of obtaining the anesthesia management results are specifically as follows: Based on the anesthesia risk update record, extract the real-time anesthesia response data of the patient during the operation, including the intensity and time information of the patient's physiological response to the anesthetic, to obtain the real-time anesthesia response data; Combined with the physiological data collected in real time, and compared with the real-time anesthesia response data, the formula is used: ; Calculate recommended anesthetic adjustments , get the anesthesia management results, among which, Indicates The dose of anesthetic at each time point, It is The patient's real-time heart rate at a certain time point. is the target heart rate, is the sensitivity parameter, is the base of natural logarithms, is the total number of data points.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by integrating historical health records and surgical information, the real-time processing capability of physiological data is enhanced. By conducting in-depth analysis of the time series changes of heart rate and blood pressure, an accurate assessment of the immediate effects of anesthetic drugs is provided. Through cluster analysis, the responses of different patient groups to anesthetics can be identified, thereby achieving targeted anesthesia plans. The LSTM network is used to analyze the injection time and dosage of anesthetics, optimize the dynamic prediction of drug reactions, and support doctors to adjust the anesthetic dosage according to real-time data, which greatly improves the accuracy and safety of anesthesia management and reduces the risk of anesthetic side effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a system flow chart of the present invention; Figure 2 This is a flowchart of processing heart rate change and blood pressure change data in the present invention; Figure 3 A flowchart for obtaining an overview of physiological data in the present invention; Figure 4 is a flow chart for analyzing the distribution characteristics of patient data in the present invention; Figure 5 This is a flow chart for obtaining group classification indicators in the present invention; Figure 6 is a flow chart for calculating the intensity of the reaction to the anesthetic in the present invention; Figure 7 A flowchart for obtaining anesthesia risk update records in the present invention; Figure 8 This is a flow chart for obtaining anesthesia management results in the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0018] Example: See also Figure 1The present invention provides a technical solution: an anesthetic effect evaluation system based on big data includes: The physiological data processing module collects physiological indicators, types and dosages of anesthetic drugs based on the patient's historical health records and surgical information, processes heart rate change and blood pressure change data, and analyzes time series changes to obtain an overview of physiological data; The patient clustering and analysis module analyzes the distribution characteristics of patient data based on the physiological data overview, clusters patients according to the changing trends of heart rate and blood pressure data, analyzes the physiological response patterns of different groups, and reclassifies patient groups to obtain group classification indicators; The anesthesia risk assessment module uses the LSTM network to analyze the time and dosage of anesthetic injections based on group classification indicators, calculates the intensity of the response of different patient groups to anesthetics, evaluates the patient's anesthesia risk index, and compares it with the real-time monitoring data during surgery to obtain an anesthesia risk update record; The anesthesia adjustment support module updates records of anesthesia risks and combines physiological data collected in real time during surgery to evaluate the real-time effect of current anesthesia, update the adjustment recommendations for anesthetic drug dosages, and obtain anesthesia management results.

[0019] The physiological data overview includes heart rate records, blood pressure records, and medication records. The group classification indicators are specifically heart rate trends, blood pressure trends, and response pattern differences. The anesthesia risk update records include risk index, time series analysis results, and real-time data comparison results. The anesthesia management results are specifically dose adjustment recommendations, effect evaluation results, and real-time monitoring feedback information.

[0020] See also Figure 2 , the processing steps of heart rate change and blood pressure change data are as follows: Based on the patient's historical health records and surgical information, physiological indicators, anesthetic drug types and dosage data were collected to obtain a preliminary physiological data set; Through the medical database interface and patient file retrieval system, key data is extracted, including continuously recorded physiological indicators such as heart rate, blood pressure, body temperature, blood oxygen saturation, and the type, dosage and duration of anesthetic drugs detailed in the surgical anesthesia record. Unstructured text data is automatically parsed, and the required numerical information is extracted and classified into structured data tables. At the same time, data from multiple sources are compared and verified to ensure the consistency of the same indicators in different records, and the continuity of data recording time is checked to eliminate records with duplicate or abnormal timestamps. After the sorting is completed, a physiological data set in a unified format is generated. This data set covers the patient's historical health records and surgery-related information.

[0021] Screening and preliminarily processing the heart rate and blood pressure data in the preliminary physiological data set, removing data points that deviate from the normal range, and obtaining screened heart rate and blood pressure data; First, conditional filtering is performed on the heart rate data to eliminate data points that are not in the range of 60 to 100 bpm. Filtering conditions are set for the blood pressure data according to the upper and lower limits of systolic and diastolic pressures to eliminate data points that are out of the normal range. At the same time, the timestamps of the heart rate and blood pressure data are checked to ensure continuity and logical consistency. Data points with too many abnormal fluctuations or missing records within a certain period of time are eliminated. All data that meet the filtering conditions are reintegrated into the filtered data set to finally obtain the filtered heart rate and blood pressure data.

[0022] Analyze the changing trends in the filtered heart rate and blood pressure data using the formula: ; Calculating heart rate variability , used to indicate the fluctuation degree of heart rate data, where Indicates A heart rate data point is each specific heart rate value obtained from the filtered heart rate data set. Represents the average value of the heart rate data, which is obtained by calculating the average of all the filtered heart rate data points. is the total number of data points, i.e., the total number of data points in the filtered heart rate dataset; There is a small sample data, where The heart rate reading is , the average heart rate Calculated as , substitute into the formula to calculate the heart rate variability: ; ; This result shows that the calculated heart rate variability A value of 100 indicates that in this sample, the heart rate fluctuates greatly, reflecting that the instability of the heart rate requires further evaluation and intervention by the doctor.

[0023] See also Figure 3 ,The specific steps for obtaining the physiological data overview are: Analyze the processed heart rate change and blood pressure change data, use autoregressive model or moving average model to identify the long-term trend and periodic changes in the data, and obtain time series analysis results; Time series data is extracted from the cleaned data set. The data is arranged in chronological order of heart rate and blood pressure. For the heart rate data, time windows such as hours or days can be set to divide subsets, and the average value and change range of each time window can be calculated. The blood pressure data is processed separately according to high pressure and low pressure, and the two types of data are simultaneously mapped to the time axis. Then, the autoregressive model (AR) is used to identify the long-term trend by calculating the correlation between the current data point and multiple time points in the past. The moving average model (MA) calculates the average value of the data in the moving window and smoothes the fluctuations. At the same time, model parameters such as the time delay order are selected to adapt to the data characteristics. Finally, after completing the model training, the data is iterated multiple times to obtain the time series analysis results reflecting the changing trends of heart rate and blood pressure.

[0024] Based on the time series analysis results, key statistical indicators are extracted, including mean, standard deviation, peak, and stability or volatility indicators of heart rate and blood pressure, to obtain an overview of physiological data; Extract key statistical indicators, including calculating the average value by summing up the heart rate and blood pressure data of each time period and dividing it by the number of data points to obtain the average value; calculating the standard deviation by taking the square root of the square of the deviation between the data points and the average value; marking the peak points in the time series; identifying the highest and lowest values ​​by comparing the size relationship between each data point and the surrounding data; and finally extracting the stability or volatility indicators of heart rate and blood pressure through the degree of variation of statistical data. The analysis results are integrated into a set of key statistical indicators to generate a complete overview of physiological data.

[0025] See also Figure 4 , the specific steps for analyzing the distribution characteristics of patient data are: Based on the physiological data overview, the mean, median, standard deviation and skewness of each data were calculated to obtain preliminary statistical analysis results; By calculating the mean, median, standard deviation and skewness of heart rate and blood pressure data, the calculation of statistical indicators requires checking the original data first, removing outliers and incomplete records, ensuring data quality, and then calculating the mean as the central trend of the data, calculating the median to measure the symmetry and skewness of the data, the standard deviation is used to measure the degree of dispersion of the data, and the skewness is used to indicate the asymmetry of the data distribution. All calculation results are organized into a complete set of analysis results, and finally form preliminary statistical analysis results, providing basic data for subsequent distribution characteristics and group analysis.

[0026] The preliminary statistical analysis results were analyzed using the formula: ; Calculate the standard score for each data point, identify outliers in the data, and obtain the distribution characteristics of the filtered heart rate and blood pressure data, where Indicates The standard score of a data point is used to measure the deviation of the data point from the average value of the overall data. Represents heart rate or blood pressure data points are the heart rate or blood pressure values ​​obtained from the processed data set. Represents the average value of the data point, which is obtained by calculating the average of all heart rate or blood pressure data. Represents the standard deviation of the data points, which is used to measure the dispersion of the data, that is, the distribution range of the data points around the mean value; The following data was collected, the heart rate data points are , group average heart rate , standard deviation , substitute the value into the formula to calculate: ; The calculated standard score is 0.57, indicating that the heart rate data point has a low degree of deviation from the average value and can be considered within the normal fluctuation range clinically. Further data analysis should focus on data points with high deviation scores.

[0027] Based on the distribution characteristics of the screened heart rate and blood pressure data, cluster analysis is performed to identify different patient groups, evaluate the distribution consistency and differences among patient groups, and obtain distribution characteristic analysis results; Patients are divided into different groups through cluster analysis. The implementation of cluster analysis first requires standardization of the screened heart rate and blood pressure data, using standardized formulas to adjust the data points to a uniform dimension, and then selecting an appropriate clustering method, such as calculating the similarity between data points based on Euclidean distance, and further determining the optimal number of clusters by calculating the elbow point. The positions of the cluster centers are adjusted iteratively for multiple times until the clustering results are stable, and through statistical descriptions of the data characteristics within each group, patient groups with significant differences are identified to obtain distribution characteristic analysis results.

[0028] See also Figure 5 , the specific steps for obtaining group classification indicators are: According to the changing trends of heart rate and blood pressure data, K-means clustering was used to classify patients into different groups, and preliminary patient grouping results were obtained; K-means clustering is used to classify patients into different groups. The process includes standardized preprocessing of the data to eliminate the influence of heart rate and blood pressure units. The initial group affiliation is determined by calculating the distance from the data point to the cluster center. The data points are then reallocated and the positions of the cluster centers are updated using an iterative adjustment method until the change in the positions of all cluster centers is less than the preset convergence threshold or the set maximum number of iterations is reached. The entire clustering result needs to be evaluated by means of silhouette coefficient or elbow rule to judge the rationality of the clustering. Finally, group classification is completed based on the similarity of the patient's heart rate and blood pressure data to obtain preliminary patient clustering results.

[0029] Based on the preliminary patient grouping results, the formula was used: ; Evaluate the differences between multiple groups and obtain the results of physiological response pattern analysis, among which, is the within-group variance, which quantifies how much the heart rate or blood pressure data points of members within a particular group differ from the group mean. Indicates the first The heart rate or blood pressure values ​​of each data point are used to evaluate the value of each individual data point. is the average value of the corresponding group data points, which is used to measure the deviation of each data point from the group average level. is the weight of each data point, defined as , is the sum of weights, is the number of data points in the population; There are three data points in the group with heart rates of 70, 72, and 80 bpm respectively. The average heart rate of the group is The weight of each data point is calculated as , the sum of weights , substitute it into the formula and get: ; ; The results show that the calculated intra-group variance is approximately 13.17, which reflects the actual variability of the heart rate of patients within the group and shows that the heart rate of patients within the group has a certain volatility.

[0030] Based on the results of the physiological response pattern analysis, the classification of the patient groups was re-evaluated, the data differences within and between groups were checked, and the group classification indicators were obtained; Re-evaluate the classification of patient groups by checking the data differences within and between groups, including calculating the internal standard deviation of each group and the mean difference between groups. By comparing statistical characteristics, determine whether there is significant heterogeneity within the group and whether the distribution between groups has statistically significant differences. In addition, further evaluate the distribution consistency between groups and the degree of concentration within groups. Statistical tests such as t-tests or ANOVA tests can be used to verify the reliability of group classification. Adjust the group classification rules based on the statistical results obtained from the analysis. Finally, complete the reclassification of the patient group and obtain the group classification indicators.

[0031] See also Figure 6 , the specific steps for calculating the intensity of the reaction to anesthetics are: Based on the group classification indicators, the LSTM network is used to extract the time series data of the patient's anesthetic injection time and dosage for analysis, identify the patterns and trends in the data, and obtain the time series data features for preliminary analysis; The LSTM network is used to analyze the time series data of patients' anesthetic injection time and dosage. The injection time and dosage are extracted from the patient's historical records. The extracted data are sorted in chronological order to form a time series data set. Data cleaning includes removing outliers and filling missing values. Mean filling and standardization are used to ensure the integrity and consistency of the data. The corresponding relationship between injection time and dosage is captured through the input layer of the network. The hidden layer extracts the implicit patterns and trends between sequences through repeated training. The output layer of the network maps the time series into a structured feature set. The features include the injection frequency, dosage distribution and trend changes of anesthetics, and generate preliminary time series data features.

[0032] Based on the time series data characteristics of the preliminary analysis, the formula is used: ; Calculate the intensity of response to anesthesia in different patient groups ,in, It is The dose of anesthetic injected for each data point, Indicates The injection time of each data point is used to calculate the time interval from injection to the evaluation moment. is the attenuation factor, which indicates the effect of time on the anesthetic effect. is the total number of data points, is the base of natural logarithms; There are three patients with anesthesia data, each of which has an anesthetic injection dose The doses are 50mg, 45mg and 55mg respectively, and the corresponding injection times are For 1 hour, 2 hours and 1.5 hours, the decay factor Set it to 0.5, and calculate the dose effect after attenuation at each time point as: ; ; ; The average response intensity of the group for: ; The results showed that the average anesthetic response intensity of the different patient groups was 24.29 mg, indicating that the comprehensive dosage and time factors under various conditions can provide a more accurate basis for anesthesia management in clinical practice.

[0033] See also Figure 7 , the specific steps for obtaining anesthesia risk update records are: Based on the intensity of the response to the anesthetic, the patient's anesthetic response characteristics during surgery were analyzed to obtain the anesthetic response data during surgery; The time and dose of anesthetic injections for patients were extracted from medical records. The data included dose values ​​of multiple injections and their corresponding time points. The data were first standardized to ensure that the injection time was converted into a time interval in a unified format and the injection dose was normalized to a reasonable range. The data were then sorted in chronological order and the time difference between each data point was calculated to generate time series feature data with the time interval as the horizontal axis and the injection dose as the vertical axis. Finally, the dose change at each time point was converted into the corresponding anesthetic reaction feature through a calculation model based on reaction intensity, and the anesthetic reaction data during surgery for further analysis was obtained.

[0034] Based on the anesthesia response data during surgery, compared with the real-time monitoring data, the change of anesthesia risk is determined by calculating the difference and trend analysis, using the formula: ; Get an updated record of anesthesia risk, where: is the total difference between the current and predicted anesthetic response intensity, Representative The monitored anesthetic response intensity of the data points comes from the real-time monitoring system during the operation. Indicates The predicted anesthetic response intensity for each data point is based on the results of LSTM network analysis. is the total number of data points; There are three data points of anesthetic response intensity for each patient, and the actual monitored values are 50, 60 and 55 respectively, while the predicted value They are 48, 63 and 53 respectively. Substitute them into the formula to calculate each difference: ; ; The results showed that the total difference value of the anesthesia risk update record was 17, reflecting that there was a certain deviation between the actual anesthetic reaction intensity and the prediction. This quantified deviation can help anesthesiologists more accurately monitor and manage anesthesia risks during surgery.

[0035] See also Figure 8 , the specific steps for obtaining anesthesia management results are: Based on the anesthesia risk update record, extract the patient's real-time anesthesia response data during surgery, including the patient's physiological response intensity and time information to anesthetics, and obtain real-time anesthesia response data; Extract the patient's real-time anesthesia response data during surgery, including the patient's physiological response intensity and time information to anesthetics. This process collects key physiological indicators such as heart rate, blood pressure and respiratory rate from the real-time monitoring system, matches them with the anesthetic injection time and dosage records, and ensures the synchronization of data through timestamp association. Then, the original data is preprocessed to remove outliers and fill in missing data to form complete continuous time series data. The cleaned data is then segmented by time point to extract the patient's immediate response intensity to anesthetics. The response intensity data is stored together with the corresponding time point information to generate real-time anesthesia response data for further analysis.

[0036] Combined with the real-time collected physiological data, and compared with the real-time anesthesia response data, the formula is used: ; Calculate recommended anesthetic adjustments , get the anesthesia management results, among which, Indicates The recommended dose of anesthetic at a specific time point. It is The patient's real-time heart rate at a certain time point. The target heart rate is the ideal heart rate level during surgery to ensure that the patient is stable. is the sensitivity parameter, is the base of natural logarithms, is the total number of data points; There are 3 data points, the anesthetic doses are 50mg, 45mg and 55mg respectively, the patient's real-time heart rate is 70bpm, 75bpm and 65bpm respectively, the target heart rate is 72bpm, the sensitivity parameter GA is set to 0.5, and the formula is substituted for calculation: ; ; ; The results showed that the calculated adjusted recommended anesthetic dose was 17.29 mg, indicating that adjusting the anesthetic dose based on the difference between the real-time heart rate and the target heart rate was reasonable and helpful in maintaining the patient's stable state during surgery.

[0037] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. Anesthesia effect evaluation system based on big data, characterized in that: The system comprises: The physiological data processing module collects physiological indicators, types and dosages of anesthetic drugs based on the patient's historical health records and surgical information, processes heart rate change and blood pressure change data, and analyzes time series changes to obtain an overview of physiological data; The patient grouping and analysis module analyzes the distribution characteristics of the patient data based on the physiological data overview, groups the patients according to the change trends of the heart rate and blood pressure data, analyzes the physiological response patterns of the different groups, and reclassifies the patient groups to obtain group classification indicators; The anesthesia risk assessment module uses the LSTM network to perform time series data analysis on the injection time and dosage of anesthetics based on the group classification indicators, calculates the response intensity of different patient groups to anesthetics, assesses the patient's anesthesia risk index, and compares it with the real-time monitoring data during the operation to obtain an anesthesia risk update record; The anesthesia adjustment support module evaluates the real-time effect of the current anesthesia based on the anesthesia risk update record and the physiological data collected in real time during the operation, and updates the adjustment suggestions for the anesthetic drug dosage to obtain the anesthesia management results.

2. The anesthesia effect evaluation system based on big data according to claim 1, characterized in that: The processing steps of the heart rate change and blood pressure change data are specifically as follows: Based on the patient's historical health records and surgical information, physiological indicators, anesthetic drug types and dosage data were collected to obtain a preliminary physiological data set; Screening and preliminarily processing the heart rate and blood pressure data in the preliminary physiological data set, removing data points that deviate from a normal range, and obtaining screened heart rate and blood pressure data; The changing trends in the screened heart rate and blood pressure data were analyzed using the formula: ； Calculating heart rate variability ,in, Indicates heart rate data points, Represents the average value of heart rate data, is the total number of data points.

3. The anesthesia effect evaluation system based on big data according to claim 1, characterized in that: The steps for obtaining the physiological data overview are specifically as follows: Analyzing the processed heart rate change and blood pressure change data, using an autoregressive model or a moving average model to identify long-term trends and periodic changes in the data, and obtaining time series analysis results; Based on the time series analysis results, key statistical indicators are extracted, including mean value, standard deviation, peak value, and stability or volatility indicators of heart rate and blood pressure, to obtain an overview of physiological data.

4. The anesthesia effect evaluation system based on big data according to claim 1, characterized in that: The steps for analyzing the distribution characteristics of the patient data are specifically as follows: Based on the physiological data overview, the mean, median, standard deviation and skewness of each data are calculated to obtain preliminary statistical analysis results; The preliminary statistical analysis results were analyzed using the formula: ； Calculate the standard score for each data point, identify outliers in the data, and obtain the distribution characteristics of the filtered heart rate and blood pressure data, where Indicates The standard score of the data points, Represents the heart rate or blood pressure data points, represents the average of the data points, represents the standard deviation of the data points; Based on the distribution characteristics of the screened heart rate and blood pressure data, cluster analysis is performed to identify different patient groups, evaluate the distribution consistency and differences among the patient groups, and obtain distribution characteristic analysis results.

5. The anesthesia effect evaluation system based on big data according to claim 1, characterized in that: The steps for obtaining the group classification index are specifically as follows: According to the change trends of the heart rate and blood pressure data, the patients are classified into different groups using K-means clustering to obtain preliminary patient grouping results; Based on the preliminary patient grouping results, the formula was used: ； Evaluate the differences between multiple groups and obtain the results of physiological response pattern analysis, among which, is the within-group variance, Indicates the first heart rate or blood pressure value for each data point, is the mean value of the corresponding group data points, is the weight of each data point, defined as , is the sum of weights, is the number of data points in the population; Based on the physiological response pattern analysis results, the classification of the patient group is re-evaluated, the data differences within and between groups are checked, and the group classification indicators are obtained.

6. The anesthesia effect evaluation system based on big data according to claim 1, characterized in that: The calculation steps of the intensity of the reaction to the anesthetic are specifically as follows: Based on the group classification indicators, the LSTM network is used to extract the time series data of the patient's anesthetic injection time and dosage for analysis, identify the patterns and trends in the data, and obtain the time series data features for preliminary analysis; Based on the time series data characteristics of the preliminary analysis, the formula is used: ； Calculate the intensity of response to anesthesia in different patient groups ,in, It is The dose of anesthetic injected for each data point, Indicates The injection time of each data point, is the attenuation factor, which indicates the effect of time on the anesthetic effect. is the total number of data points, is the base of natural logarithms.

7. The anesthesia effect evaluation system based on big data according to claim 1, characterized in that: The steps for obtaining the anesthesia risk update record are specifically as follows: Based on the intensity of the reaction to the anesthetic, analyzing the patient's anesthetic reaction characteristics during the operation to obtain anesthetic reaction data during the operation; Based on the anesthesia response data during the operation, the data is compared with the real-time monitoring data, and the change of anesthesia risk is determined by calculating the difference and trend analysis, using the formula: ； Get an updated record of anesthesia risk, where: is the total difference between the current and predicted anesthetic response intensity, Representative The intensity of the monitored anesthetic response for each data point, Indicates The predicted anesthetic response intensity for each data point, is the total number of data points.

8. The anesthesia effect evaluation system based on big data according to claim 1, characterized in that: The steps for obtaining the anesthesia management results are specifically as follows: Based on the anesthesia risk update record, extract the real-time anesthesia response data of the patient during the operation, including the intensity and time information of the patient's physiological response to the anesthetic, to obtain the real-time anesthesia response data; Combined with the physiological data collected in real time, and compared with the real-time anesthesia response data, the formula is used: ； Calculate recommended anesthetic adjustments , get the anesthesia management results, among which, Indicates The dose of anesthetic at each time point, It is The patient's real-time heart rate at a certain time point. is the target heart rate, is the sensitivity parameter, is the base of natural logarithms, is the total number of data points.

Images