A general anesthesia symptom prediction system based on patient data

Abstract

The present invention relates to the field of medical prediction technology, specifically a system for predicting general anesthesia symptoms based on patient data, the system comprising: a state monitoring module that collects the patient's heart rate and blood pressure parameters, records the continuous time point data during the operation, tracks the state changes, and obtains the state change tracking results. In the present invention, by analyzing the continuous physiological data of the patient during the operation, the accuracy of the prediction of anesthesia complications is improved, the slight changes in the physiological state are monitored in real time and the correlation analysis is performed with the effects of anesthetics, so that the ability to predict potential risks is enhanced, and through the real-time updated conditional probability calculation, not only the insight into key risk points is increased, but also data support is provided for the decision-making of the medical team, thereby effectively reducing the probability of patients encountering adverse events during the operation, and optimizing the risk assessment in combination with historical data, the overall safety of the operation process is guaranteed, and the treatment effect on the patient is significantly improved.

Description

Technical Field

[0001] The present invention relates to the field of medical prediction technology, and in particular to a general anesthesia symptom prediction system based on patient data. Background Art

[0002] The field of medical prediction technology is a technology that applies data analysis, machine learning and statistical methods to predict disease progression, treatment outcomes and possible health risks. This field covers a wide range of technologies from basic case data to complex bioinformatics analysis, aiming to improve the early diagnosis rate and treatment effect of diseases through the analysis of historical health records, real-time physiological data and genetic information. Medical prediction technology plays a key role in improving the quality of patient care and optimizing the allocation of medical resources, especially in formulating personalized treatment plans and conducting risk assessments.

[0003] Among them, the general anesthesia symptom prediction system is a specific example of the application of medical prediction technology. It predicts different symptoms that may occur after general anesthesia, such as respiratory depression and abnormal heart rate, by analyzing the patient's medical data. The purpose of this system is to provide targeted early warnings before surgery, so that the medical team can prepare countermeasures in advance and reduce anesthesia-related complications. Through real-time monitoring and data analysis, the system can help doctors adjust the dose and type of anesthetics more accurately to ensure patient safety. The use of this technology has greatly improved surgical safety and the overall care effect of patients.

[0004] Existing medical prediction technologies mainly focus on the analysis of basic data and lack the ability to comprehensively analyze real-time data during surgery, especially in terms of monitoring subtle changes in physiological parameters and responding immediately. These technologies fail to fully utilize continuous physiological monitoring data, limiting their effectiveness in emergency warning and risk assessment. In actual operations, this limitation may cause the medical team to respond to anesthesia complications in a less timely manner, increasing surgical risks. In addition, existing systems also show obvious limitations in the efficiency of data integration and the use of historical complication information for prediction. This lack of data-driven analysis may lead to the omission of key warning information and increase the health risks of patients. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a general anesthesia symptom prediction system based on patient data.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A system for predicting general anesthesia symptoms based on patient data comprises:

[0007] The state monitoring module collects the patient's heart rate and blood pressure parameters, and records the continuous time point data during the operation, tracks the state changes, and obtains the state change tracking results;

[0008] The causal analysis module uses the correlation analysis between the physiological parameters and the anesthetics based on the state change tracking results to determine the impact of the change in the physiological state on the anesthetics, generates parameter response analysis results, compares the parameter response analysis results with the historical data of anesthesia complications, and identifies the causal relationship between the anesthetics and the complications;

[0009] Based on the causal relationship, the risk assessment module uses data at continuous time points to adjust the conditional probability table, calculates the probability of occurrence of anesthetic complications under each state, combines historical complication data, estimates key risk points that may occur during the operation, and generates key risk point prediction results.

[0010] As a further solution of the present invention, the step of obtaining the state change tracking result is specifically:

[0011] By collecting heart rate data at consecutive time points, the weighted average of the heart rate data is calculated using the formula:

[0012] ;

[0013] Generate average heart rate values ;

[0014] in, Representative The heart rate value at a time point, Representative The weight of the data at each time point, represents the total number of time points, represents the calculated weighted average of heart rate;

[0015] Using the average heart rate value , calculate the weighted standard deviation of heart rate fluctuations using the formula:

[0016] ;

[0017] Get heart rate fluctuation data;

[0018] in, represents the weighted standard deviation of heart rate, represents the total number of time points;

[0019] According to the average heart rate value and the heart rate fluctuation data, the formula is adopted:

[0020] ;

[0021] Calculate the heart rate status score and generate status change tracking results;

[0022] in, is the adjustment factor, is the normalization coefficient, Score your heart rate status.

[0023] As a further solution of the present invention, the step of obtaining the parameter response analysis result is specifically:

[0024] Collect the state change tracking results and data of multiple physiological parameters, using the formula:

[0025] ;

[0026] The correlation coefficient between each parameter and the anesthetic concentration was calculated to generate parameter-response relationship indices;

[0027] in, Representative The heart rate status score at each time point, Representative The physiological parameters are The value at a time point, , are the mean values ​​of heart rate state score and physiological parameters, is the correlation coefficient indicator;

[0028] Using the parameter response relationship index, the impact intensity analysis of each parameter is performed using the formula:

[0029] ;

[0030] Calculate the impact strength of each parameter to obtain the impact strength index;

[0031] in, To influence the intensity index, represents the adjustment factor, Indicates Correlation coefficients for time points.

[0032] Combined with the impact intensity index, the composite scoring model formula is adopted:

[0033] ;

[0034] Calculate and generate parameter response analysis results;

[0035] in, For the The importance weight of the parameters, Results for parameter response analysis.

[0036] As a further solution of the present invention, the step of identifying the causal relationship between the anesthetic and the complication is specifically:

[0037] The parameter response analysis results and historical anesthesia complication data were used to perform correlation analysis using the formula:

[0038] ;

[0039] Generate preliminary correlation coefficients;

[0040] in, Complication data at the same time point, is the weighting factor, is the preliminary correlation coefficient.

[0041] The preliminary correlation coefficient was analyzed, and the stability and reliability of the correlation coefficient were evaluated by statistical methods, using the formula:

[0042] ;

[0043] Get the standardized association strength;

[0044] in, is the mean, is the standard deviation, is the standardized correlation strength.

[0045] Based on the standardized strength of association, logistic regression analysis was used to determine the probability of a causal relationship using the formula:

[0046] ;

[0047] Identify causal relationships between anesthetics and complications;

[0048] in, is the regression coefficient, Indicates a causal relationship between the anesthetic and the complication.

[0049] As a further solution of the present invention, the step of calculating the probability of occurrence of anesthesia complications in each state is specifically as follows:

[0050] Based on the causal relationship between the anesthetics and complications, the anesthetic use at consecutive time points was integrated and the formula was used:

[0051] ;

[0052] Calculate the weighted average causal probability;

[0053] in, is the data weight at a time point, is the normalization coefficient, represents the weighted average causal probability;

[0054] Using the weighted causal probability, the conditional probability table is updated to reflect the effect of changes in anesthetic concentration on the probability of complications, applying the exponential adjustment formula:

[0055] ;

[0056] Generate an updated conditional probability table;

[0057] in, is the adjustment factor, is the updated conditional probability table;

[0058] Using the updated conditional probability table, combined with multi-state historical complication data, the formula is adopted:

[0059] ;

[0060] Calculate the probability of anesthetic complications under each condition;

[0061] in, For the The conditional probability after the state update is For the The historical frequency of occurrence of a state, The probability of occurrence of anesthetic complications under each condition.

[0062] As a further solution of the present invention, the step of obtaining the prediction result of the key risk point is specifically as follows:

[0063] Based on the probability of occurrence of anesthesia complications in each state, combined with real-time monitoring data, the risks are preliminarily classified and weighted using the formula:

[0064] ;

[0065] Produce preliminary risk assessment results;

[0066] in, is the risk weighting factor, is a small constant, Indicates the preliminary risk assessment results;

[0067] The preliminary risk assessment results are dynamically adjusted to reflect changes in real-time surgical conditions using the formula:

[0068] ;

[0069] Generate dynamic risk assessment results;

[0070] in, is the adjustment factor, is the variance of the probability of complications, It is the result of dynamic risk assessment;

[0071] Using the dynamic risk assessment results, combined with historical complication data and other clinical parameters, the formula is used:

[0072] ;

[0073] Generate prediction results of key risk points;

[0074] in, is the historical data weighting factor, is the normalization factor, Predict outcomes for key risk points.

[0075] Compared with the prior art, the advantages and positive effects of the present invention are:

[0076] In the present invention, the accuracy of anesthesia complication prediction is improved by analyzing the continuous physiological data of the patient during surgery. Real-time monitoring of small changes in physiological status and correlation analysis with the effects of anesthetics enhances the ability to predict potential risks. Real-time updated conditional probability calculations not only increase insight into key risk points, but also provide data support for the medical team's decision-making, thereby effectively reducing the probability of patients encountering adverse events during surgery. Optimization of risk assessment combined with historical data ensures the overall safety of the surgical process and significantly improves the treatment effect on patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 is a system flow chart of the present invention;

[0078] Figure 2 A flowchart of the steps for obtaining the state change tracking result of the present invention;

[0079] Figure 3 A flow chart of the steps for obtaining the parameter response analysis results of the present invention;

[0080] Figure 4 A flow chart of the steps for identifying a causal relationship between an anesthetic and a complication for the present invention;

[0081] Figure 5 A flow chart of the steps for calculating the probability of occurrence of anesthesia complications under each state in the present invention;

[0082] Figure 6 This is a flow chart of the steps for obtaining the key risk point prediction results of the present invention. DETAILED DESCRIPTION

[0083] 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.

[0084] 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" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the 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, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0085] Example 1: Please refer to Figure 1 , a general anesthesia symptom prediction system based on patient data includes:

[0086] The state monitoring module collects the patient's heart rate and blood pressure parameters, and records the continuous time point data during the operation, tracks the state changes, and obtains the state change tracking results;

[0087] The causal analysis module uses the correlation analysis between physiological parameters and anesthetics based on the state change tracking results to determine the impact of changes in physiological state on anesthetics, generate parameter response analysis results, compare the parameter response analysis results with the historical data of anesthesia complications, and identify the causal relationship between anesthetics and complications;

[0088] The risk assessment module is based on causality and uses data at continuous time points to adjust the conditional probability table, calculate the probability of occurrence of anesthetic complications under each state, combine historical complication data, estimate the key risk points that may occur during the operation, and generate key risk point prediction results.

[0089] The state change tracking results include heart rate parameters, blood pressure parameters, and time series data. The parameter response analysis results include physiological parameter association, anesthetic effect judgment, and complication history comparison. The key risk point prediction results include conditional probability adjustment, complication probability estimation, and key risk point identification.

[0090] See also Figure 2 , the specific steps for obtaining the state change tracking results are:

[0091] By collecting heart rate data at consecutive time points, the weighted average of the heart rate data is calculated using the formula:

[0092] ;

[0093] Generate average heart rate values ;

[0094] in, Representative The heart rate value at a time point, Representative The weight of the data at each time point, represents the total number of time points, represents the calculated weighted average of heart rate;

[0095] Using average heart rate values , calculate the weighted standard deviation of heart rate fluctuations using the formula:

[0096] ;

[0097] Get heart rate fluctuation data;

[0098] in, represents the weighted standard deviation of heart rate, represents the total number of time points;

[0099] According to the average heart rate value and heart rate fluctuation data, the formula is used:

[0100] ;

[0101] Calculate the heart rate status score and generate status change tracking results;

[0102] in, is the adjustment factor, is the normalization coefficient, Score your heart rate status.

[0103] Step 1: Calculate your weighted average heart rate ;

[0104] formula:

[0105] ;

[0106] Parameter explanation:

[0107] : No. Heart rate value at a time point.

[0108] : No. The weight of a data point indicates the importance of the data point.

[0109] : The total number of time points.

[0110] Assumptions:

[0111] Heart rate data ;

[0112] Weight ;

[0113] Total data points ;

[0114] Calculation process:

[0115] Calculate the sum of the weighted sums of the heart rates:

[0116] ;

[0117] Calculate the weighted sum:

[0118] ;

[0119] Calculate weighted average heart rate:

[0120] ;

[0121] result:

[0122] Weighted average heart rate Approx. . This means that for a given weight, the average heart rate is .

[0123] Step 2: Calculate the weighted standard deviation of heart rate ( );

[0124] formula:

[0125] ;

[0126] Calculation process:

[0127] Compute the weighted sum of the sums of squared differences:

[0128] ;

[0129] Using the previously calculated weight sum:

[0130] ;

[0131] result:

[0132] Weighted standard deviation It is about 11.55. This shows how much the heart rate data fluctuates around the average.

[0133] Step 3: Calculate your heart rate status score ;

[0134] formula:

[0135] ;

[0136] Assuming coefficients:

[0137] ;

[0138] ;

[0139] (normalization factor);

[0140] Calculation process:

[0141] ;

[0142] result:

[0143] Heart Rate Status Score The score is 52.62. This score can be used to measure the health of an individual's heart rate status, where a higher score indicates a more stable heart rate or a desired health status.

[0144] See also Figure 3 , the steps to obtain the parameter response analysis results are as follows:

[0145] Collect state change tracking results and data of multiple physiological parameters using the formula:

[0146] ;

[0147] The correlation coefficient between each parameter and the anesthetic concentration was calculated to generate parameter-response relationship indices;

[0148] in, Representative The heart rate status score at each time point, Representative The physiological parameters are The value at a time point, , are the mean values ​​of heart rate state score and physiological parameters, is the correlation coefficient indicator;

[0149] Use the parameter response relationship index to analyze the impact intensity of each parameter using the formula:

[0150] ;

[0151] Calculate the impact strength of each parameter to obtain the impact strength index;

[0152] in, To influence the intensity index, represents the adjustment factor, Indicates Correlation coefficients for time points.

[0153] Combined with the impact intensity index, the composite scoring model formula is adopted:

[0154] ;

[0155] Calculate and generate parameter response analysis results;

[0156] in, For the The importance weight of the parameters, Results for parameter response analysis.

[0157] Formula analysis and examples for step 1

[0158] formula:

[0159] ;

[0160] Parameter explanation:

[0161] : No. Heart rate status score at each time point;

[0162] : The average value of heart rate state score;

[0163] : No. The physiological parameters are The value at each time point;

[0164] : No. The average value of physiological parameters;

[0165] Correlation coefficients between parameters and heart rate scores;

[0166] Calculation process and examples:

[0167] Suppose there are data at three time points , heart rate state score and physiological parameters They are as follows:

[0168] ;

[0169] ;

[0170] Calculate the average:

[0171] ;

[0172] ;

[0173] Calculate the correlation coefficient :

[0174] molecular: ;

[0175] Left part of the denominator: ;

[0176] Right part of the denominator: ;

[0177] Denominator: ;

[0178] therefore, , indicating that there is no correlation between heart rate state score and physiological parameters.

[0179] Formula analysis and examples for step 2;

[0180] formula:

[0181] ;

[0182] Parameter explanation:

[0183] : No. Correlation coefficients of time points;

[0184] : No. Physiological parameter values ​​at time points;

[0185] : adjustment coefficient;

[0186] Calculation process and examples:

[0187] Assumptions And using the previous step calculated and value:

[0188] ;

[0189] Formula analysis and examples for step 3;

[0190] formula:

[0191] ;

[0192] Parameter explanation:

[0193] No. The intensity of the response of each physiological parameter;

[0194] : No. The importance weight of each parameter;

[0195] Calculation process and examples:

[0196] Assume there is a physiological parameter and :

[0197] ;

[0198] The results Represents the comprehensive parameter response analysis results. A value of 1085 indicates the overall impact strength. Higher values ​​indicate that physiological parameters have a significant impact on anesthetic response.

[0199] See also Figure 4 The steps to identify the causal relationship between anesthetics and complications are as follows:

[0200] The results of parameter response analysis and historical anesthesia complication data were used for correlation analysis using the formula:

[0201] ;

[0202] Generate preliminary correlation coefficients;

[0203] in, Complication data at the same time point, is the weighting factor, is the preliminary correlation coefficient.

[0204] The preliminary correlation coefficient was analyzed, and the stability and reliability of the correlation coefficient were evaluated by statistical methods, using the formula:

[0205] ;

[0206] Get the standardized association strength;

[0207] in, is the mean, is the standard deviation, is the standardized correlation strength.

[0208] Logistic regression analysis was used to determine the probability of a causal relationship based on the standardized strength of association, using the formula:

[0209] ;

[0210] Identify causal relationships between anesthetics and complications;

[0211] in, is the regression coefficient, Indicates a causal relationship between the anesthetic and the complication.

[0212] Step 1: Preliminary correlation coefficients ;

[0213] formula:

[0214] ;

[0215] Parameter details:

[0216] :Indicates the The weight of each time point is assumed to be equal and is set to 1.

[0217] :Indicates the Complication data at each time point were obtained from historical records.

[0218] Calculation example:

[0219] Suppose there are data at three time points:

[0220] data: ;

[0221] data: ;

[0222] Weight : ;

[0223] calculate:

[0224] ;

[0225] explain:

[0226] The calculated result 33.5 represents the overall correlation between anesthetics and complication data, taking into account the weight of each time point.

[0227] Step 2: Normalize the association strength ;

[0228] formula:

[0229] ;

[0230] Parameter details:

[0231] : The correlation coefficient calculated in step 1.

[0232] mean The mean of .

[0233] The standard deviation of .

[0234] Calculation example:

[0235] Use the previous step :

[0236] Assume that the data set Other values ​​for ,but and The calculation is as follows:

[0237] Mean ;

[0238] The standard deviation can be calculated using the standard formula ;

[0239] ;

[0240] explain:

[0241] A value of -0.25 indicates that the correlation coefficient is below the average level, reflecting the weak association between anesthetics and complications.

[0242] Step 3: Causality Assessment Results ;

[0243] formula:

[0244] ;

[0245] Parameter details:

[0246] : The normalized association strength calculated in step 2.

[0247] , : Parameters of the logistic regression model, assuming To simplify the model.

[0248] Calculation example:

[0249] Use step 2 :

[0250] ;

[0251] explain:

[0252] A result of 0.44 indicates a low probability (44%) that there is a causal relationship between the anesthetic and the complication.

[0253] See also Figure 5 ,The steps to calculate the probability of occurrence of anesthesia complications under each state are as follows:

[0254] Based on the causal relationship between anesthetics and complications, the use of anesthetics at consecutive time points was integrated and the formula was used:

[0255] ;

[0256] Calculate the weighted average causal probability;

[0257] in, is the data weight at a time point, is the normalization coefficient, represents the weighted average causal probability;

[0258] Using the weighted causal probability, the conditional probability table is updated to reflect the effect of changes in anesthetic concentration on the probability of complications, applying the exponential adjustment formula:

[0259] ;

[0260] Generate an updated conditional probability table;

[0261] in, is the adjustment factor, is the updated conditional probability table;

[0262] Using the updated conditional probability table and combining the historical complication data of multiple states, the formula is adopted:

[0263] ;

[0264] Calculate the probability of anesthetic complications under each condition;

[0265] in, For the The conditional probability after the state update is For the The historical frequency of occurrence of a state, The probability of occurrence of anesthetic complications under each condition.

[0266] Formula for step 1:

[0267] ;

[0268] Parameters and data settings:

[0269] Suppose there are data at three time points.

[0270] (Strength of causal relationship): -Indicates the strength of the causal relationship at different points in time.

[0271] (Data weight): - indicates the importance of a time point, with the second time point having higher importance.

[0272] - Normalization coefficients to ensure that the results are within an appropriate range.

[0273] Calculation process:

[0274] ;

[0275] This calculation result Represents the causal relationship strength after weighted average.

[0276] Formula for step 2:

[0277] ;

[0278] in, is the weighted causality probability obtained from step 1, is the adjustment coefficient, which is used to control the adjustment range.

[0279] Parameter settings:

[0280] -Adjustment factor, higher values ​​indicate larger adjustments to the original data.

[0281] Calculation process:

[0282] ;

[0283] This result Represents the updated conditional probability value.

[0284] Formula for step 3:

[0285] ;

[0286] Parameters and data settings:

[0287] Assume there are 3 states of data.

[0288] -Updated conditional probabilities.

[0289] (Historical frequency): -Indicates the historical frequency of each state.

[0290] Calculation process:

[0291] ;

[0292] This result In all cases, the average probability of complications is .

[0293] See also Figure 6 , the specific steps for obtaining the prediction results of key risk points are:

[0294] Based on the probability of occurrence of anesthesia complications in each state and combined with real-time monitoring data, the risks are preliminarily classified and weighted using the formula:

[0295] ;

[0296] Produce preliminary risk assessment results;

[0297] in, is the risk weighting factor, is a small constant, Indicates the preliminary risk assessment results;

[0298] The initial risk assessment results are dynamically adjusted to reflect changes in real-time surgical conditions using the formula:

[0299] ;

[0300] Generate dynamic risk assessment results;

[0301] in, is the adjustment factor, is the variance of the probability of complications, It is the result of dynamic risk assessment;

[0302] Using the dynamic risk assessment results, combined with historical complication data and other clinical parameters, the formula is used:

[0303] ;

[0304] Generate prediction results of key risk points;

[0305] in, is the historical data weighting factor, is the normalization factor, Predict outcomes for key risk points.

[0306] Step 1: Initial risk assessment;

[0307] formula:

[0308] ;

[0309] Parameter explanation:

[0310] : No. Risk weight factors for each state are based on clinical prospective analysis.

[0311] : No. The probability of complications in each state.

[0312] : Small constant used to avoid numerical errors in logarithmic calculations.

[0313] : Indicates preliminary risk assessment results.

[0314] Derivation process:

[0315] Assume there are three states, corresponding to the probability of complications They are , 0.5, and the corresponding risk weight for , 1.5, take :

[0316] For the first state:

[0317] ;

[0318] For the second state:

[0319] ;

[0320] For the third state:

[0321] ;

[0322] Overall risk assessment:

[0323] ;

[0324] This result value indicates that, based on the set parameters and given probability of complications, the initial risk assessment indicates that the risk is low.

[0325] Step 2: Dynamic risk adjustment;

[0326] formula:

[0327] ;

[0328] Parameter explanation:

[0329] Initial risk assessment from step 1.

[0330] : Variance adjustment factor, used to adjust the sensitivity of surgical dynamics.

[0331] The variance in the probability of complications.

[0332] : Dynamic risk assessment results.

[0333] Derivation process:

[0334] Continuing with the above assumed values, calculate The variance of :

[0335] ;

[0336] ;

[0337] This resulting value indicates that the dynamically adjusted risk assessment shows a slightly increased risk after taking into account real-time data.

[0338] Step 3: Prediction of key risk points;

[0339] formula:

[0340] ;

[0341] Parameter explanation:

[0342] : No. Dynamic risk assessment results for each state.

[0343] : Historical data weighting factor.

[0344] : Normalization factor.

[0345] : Predicted key risk point results.

[0346] Derivation process:

[0347] Assuming historical data weighting factors and the normalization factor :

[0348] ;

[0349] This result value represents the final forecast of key risk points after comprehensive dynamic risk adjustment and historical data, indicating that the existing risk is relatively low.

[0350] 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. A system for predicting general anesthesia symptoms based on patient data, characterized in that: The system comprises: The state monitoring module collects the patient's heart rate and blood pressure parameters, and records the continuous time point data during the operation, tracks the state changes, and obtains the state change tracking results; The causal analysis module uses the correlation analysis between the physiological parameters and the anesthetics based on the state change tracking results to determine the impact of the change in the physiological state on the anesthetics, generates parameter response analysis results, compares the parameter response analysis results with the historical data of anesthesia complications, and identifies the causal relationship between the anesthetics and the complications; Based on the causal relationship, the risk assessment module uses the data at continuous time points to adjust the conditional probability table, calculates the probability of occurrence of anesthesia complications under each state, combines historical complication data, estimates the key risk points that occur during the operation, and generates key risk point prediction results; The steps for obtaining the state change tracking result are specifically as follows: By collecting heart rate data at consecutive time points, the weighted average of the heart rate data is calculated using the formula: ； Generate average heart rate values ; in, Representative The heart rate value at a time point, Representative The weight of the data at each time point, represents the total number of time points, represents the calculated weighted average of heart rate; Using the average heart rate value , calculate the weighted standard deviation of heart rate fluctuations using the formula: ； Get heart rate fluctuation data; in, represents the weighted standard deviation of heart rate, represents the total number of time points; According to the average heart rate value and the heart rate fluctuation data, the formula is adopted: ； Calculate the heart rate status score and generate status change tracking results; in, , is the adjustment factor, is the normalization coefficient, Score your heart rate status; The steps for obtaining the parameter response analysis results are specifically as follows: Collect the state change tracking results and data of multiple physiological parameters, using the formula: ； The correlation coefficient between each parameter and the anesthetic concentration was calculated to generate parameter-response relationship indices; in, Representative The heart rate status score at each time point, Representative The physiological parameters are The value at a time point, , are the mean values ​​of heart rate state score and physiological parameters, is the correlation coefficient indicator; Using the parameter response relationship index, the impact intensity analysis of each parameter is performed using the formula: ； Calculate the impact strength of each parameter to obtain the impact strength index; in, To influence the intensity index, , represents the adjustment factor, Indicates Correlation coefficients of time points; Combined with the impact intensity index, the composite scoring model formula is adopted: ； Calculate and generate parameter response analysis results; in, For the The importance weight of the parameters, is the parameter response analysis result; The steps of identifying the causal relationship between the anesthetic and the complication are specifically: The parameter response analysis results and historical anesthesia complication data were used to perform correlation analysis using the formula: ； Generate preliminary correlation coefficients; in, Complication data at the same time point, is the weighting factor, is the preliminary correlation coefficient; The preliminary correlation coefficient was analyzed, and the stability and reliability of the correlation coefficient were evaluated by statistical methods, using the formula: ； Get the standardized association strength; in, is the mean, is the standard deviation, is the standardized association strength; Based on the standardized strength of association, logistic regression analysis was used to determine the probability of a causal relationship using the formula: ； Identify causal relationships between anesthetics and complications; in, , is the regression coefficient, Indicates a causal relationship between the anesthetic and the complication; The steps of calculating the probability of occurrence of anesthesia complications under each state are specifically as follows: Based on the causal relationship between the anesthetics and complications, the anesthetic use at consecutive time points was integrated and the formula was used: ； Calculate the weighted average causal probability; in, is the data weight at a time point, is the normalization coefficient, represents the weighted average causal probability; Using the weighted average causal probability, the conditional probability table is updated to reflect the effect of changes in anesthetic concentration on the probability of complications, and the exponential adjustment formula is applied: ； Generate an updated conditional probability table; in, is the adjustment factor, is the updated conditional probability table; Using the updated conditional probability table, combined with multi-state historical complication data, the formula is adopted: ； Calculate the probability of anesthetic complications under each condition; in, For the The conditional probability after the state update is For the The historical frequency of occurrence of a state, is the probability of occurrence of anesthetic complications under each condition; The steps for obtaining the prediction results of the key risk points are specifically as follows: Based on the probability of occurrence of anesthesia complications in each state, combined with real-time monitoring data, the risks are preliminarily classified and weighted using the formula: ； Produce preliminary risk assessment results; in, is the risk weighting factor, is a small constant, Indicates the preliminary risk assessment results; The preliminary risk assessment results are dynamically adjusted to reflect changes in real-time surgical conditions using the formula: ； Generate dynamic risk assessment results; in, is the adjustment factor, is the variance of the probability of complications, It is the result of dynamic risk assessment; Using the dynamic risk assessment results, combined with historical complication data and other clinical parameters, the formula is used: ； Generate prediction results of key risk points; in, is the historical data weighting factor, is the normalization factor, Predict outcomes for key risk points.

Images