A construction method and system for an anesthesia department risk prediction model

Abstract

The present invention relates to the field of medical technology, and particularly to a construction method and system of an anesthesia department risk prediction model. The method includes the following steps: obtaining historical anesthesia data of the anesthesia department and conducting adaptive analysis to obtain individual anesthesia adaptation data; obtaining historical medical record data and conducting intraoperative anesthesia risk assessment to obtain intraoperative anesthesia risk assessment data; conducting postoperative anesthesia risk assessment on anesthesia medical record data to obtain postoperative anesthesia risk assessment data; evaluating and correcting the postoperative anesthesia risk assessment data according to the individual anesthesia adaptation data to obtain corrected postoperative anesthesia risk assessment data; constructing an initial anesthesia risk prediction model based on the historical anesthesia data of the anesthesia department, and iteratively optimizing and adjusting the parameters of the initial anesthesia risk prediction model based on the corrected postoperative anesthesia risk assessment data and the intraoperative anesthesia risk assessment data, so as to obtain the anesthesia department risk prediction model. The present invention can more accurately predict the intraoperative and postoperative anesthesia risks of patients.

Description

Technical Field

[0001] The present invention relates to the field of medical technologies, and particularly to a construction method and system for an anesthesia department risk prediction model. Background Art

[0002] Anesthesia for surgery is an important part of the medical field, but accurate prediction of anesthesia risks during and after surgery is crucial for patient safety and surgical success. Traditional anesthesia risk assessment is mainly based on doctors' experience, which has the disadvantages of subjectivity and poor consistency. With the development of artificial intelligence and big data technologies, data-driven anesthesia risk prediction models have become an effective way to solve this problem. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a construction method and system for an anesthesia department risk prediction model to solve at least one of the above technical problems.

[0004] To achieve the above object, a construction method for an anesthesia department risk prediction model includes the following steps:

[0005] Step S1: Obtain historical anesthesia data of the anesthesia department, and perform individual anesthesia adaptability analysis based on the historical anesthesia data of the anesthesia department to obtain individual anesthesia adaptability data;

[0006] Step S2: Obtain historical medical record data, and extract anesthesia medical record data from the historical medical record data based on the historical anesthesia data of the anesthesia department to obtain anesthesia medical record data; perform intraoperative anesthesia risk assessment based on the anesthesia medical record data to obtain intraoperative anesthesia risk assessment data;

[0007] Step S3: Extract individual physiological characteristics from the anesthesia medical record data to obtain individual preoperative physiological data and individual postoperative physiological data, and perform postoperative anesthesia risk assessment on the individual preoperative physiological data and individual postoperative physiological data to obtain postoperative anesthesia risk assessment data;

[0008] Step S4: Evaluate the postoperative recovery offset value based on the individual anesthesia adaptability data for the individual preoperative physiological data and individual postoperative physiological data to obtain postoperative recovery offset value data; evaluate and correct the postoperative anesthesia risk assessment data based on the postoperative recovery offset value data to obtain postoperative anesthesia risk assessment correction data;

[0009] Step S5: Construct an initial anesthesia risk prediction model based on the historical anesthesia data of the anesthesia department, and perform iterative optimization and parameter adjustment on the initial anesthesia risk prediction model based on the postoperative anesthesia risk assessment correction data and the intraoperative anesthesia risk assessment data to obtain an anesthesia department risk prediction model.

[0010] By obtaining the historical anesthesia data of the anesthesiology department and conducting individual anesthesia adaptability analysis, the present invention can identify the characteristics and trends of individual patients from a large amount of data, providing data support for personalized anesthesia. It helps to determine the most suitable anesthesia plan for the patient according to the patient's physiological condition and historical data, reducing unnecessary risks and complications. Obtaining historical medical record data and extracting anesthesia medical record data can evaluate the intraoperative anesthesia risk, thereby discovering potential anesthesia risk factors in advance. It helps medical staff to take necessary measures in a timely manner during the operation to reduce the incidence of intraoperative complications and ensure the smooth progress of the operation. Extracting individual physiological characteristics from anesthesia medical record data and conducting postoperative anesthesia risk assessment helps to understand the physiological condition and anesthesia response of the patient after surgery, and adjust the treatment plan in a timely manner. It helps to improve the postoperative recovery rate of the patient, reduce the occurrence of postoperative complications, and ensure the long-term effect of surgical treatment. By evaluating the postoperative recovery deviation value data, the postoperative anesthesia risk of the patient can be more accurately evaluated, and risk correction and treatment plan adjustment can be carried out in a timely manner. It helps to avoid complications and adverse consequences caused by poor postoperative recovery, and improve the postoperative survival rate and quality of life of the patient. Based on the historical anesthesia data of the anesthesiology department, an initial anesthesia risk prediction model is constructed, and the model is iteratively optimized and parameter-tuned through postoperative anesthesia risk assessment correction data and intraoperative anesthesia risk assessment data, which can improve the accuracy and stability of the model. It can help medical staff to use the model for anesthesia risk prediction more reliably, provide more effective decision-making support for surgical treatment, and maximize the safety of the patient and the success of the operation.

[0011] Optionally, step S1 is specifically as follows:

[0012] Step S11: Obtain the historical anesthesia data of the anesthesiology department, and extract anesthesia adverse characteristics from the historical anesthesia data of the anesthesiology department to obtain anesthesia adverse data;

[0013] Step S12: Conduct statistical analysis on the number of individual anesthesias in the historical anesthesia data of the anesthesiology department to obtain high-frequency anesthesia individual data and low-frequency anesthesia individual data;

[0014] Step S13: Conduct high-frequency individual anesthesia adaptability analysis on the high-frequency anesthesia individual data according to the anesthesia adverse data to obtain high-frequency individual anesthesia adaptability data;

[0015] Step S14: Conduct low-frequency individual anesthesia adaptability analysis on the low-frequency anesthesia individual data according to the anesthesia adverse data to obtain low-frequency individual anesthesia adaptability data;

[0016] Step S15: Combine the high-frequency individual anesthesia adaptability data and the low-frequency individual anesthesia adaptability data to obtain individual anesthesia adaptability data.

[0017] The present invention obtains historical anesthesia data of the anesthesiology department and extracts adverse anesthesia characteristics, which can identify adverse events, complications, and other adverse situations existing during anesthesia. These adverse characteristics can provide valuable information to help identify potential anesthesia risk factors, guide clinical decision-making, improve anesthesia operation and management, and thus enhance surgical safety and the postoperative outcomes of patients. Statistical analysis of the individual anesthesia times in the historical anesthesia data of the anesthesiology department can distinguish high-frequency and low-frequency anesthesia operations of an individual in the anesthesiology department. It helps to understand the individual's anesthesia experience and exposure. For individuals with different frequencies of anesthesia, their anesthesia adaptability may vary, and targeted assessment and management are required. Analyzing the data of high-frequency anesthesia individuals based on the adverse anesthesia data can evaluate the anesthesia adaptability and risk status of high-frequency anesthesia individuals, help identify specific risk factors and potential problems of high-frequency anesthesia individuals, and provide personalized anesthesia strategies and management measures for them to improve the safety and effectiveness of the surgical process. Analyzing the data of low-frequency anesthesia individuals based on the adverse anesthesia data can evaluate the anesthesia adaptability and risk status of low-frequency anesthesia individuals, help discover special problems and potential risks of low-frequency anesthesia individuals, and provide more refined anesthesia management and precautions for them to avoid potential complications and side effects. Combining the anesthesia adaptability data of high-frequency and low-frequency individuals can obtain more comprehensive and integrated individual anesthesia adaptability data, which helps to comprehensively understand the anesthesia adaptability characteristics, risk status, and personalized needs of individuals, and provides more accurate data support for formulating individualized anesthesia plans and risk prediction models.

[0018] Optionally, step S13 is specifically as follows:

[0019] Step S131: Calculate the anesthesia intervals for the data of high-frequency anesthesia individuals to obtain anesthesia interval data, and perform low-interval clustering calculation on the anesthesia interval data to obtain short-interval anesthesia data;

[0020] Step S132: Obtain the anesthesia adverse reaction rules, and classify the adverse anesthesia data according to the anesthesia adverse reaction rules to obtain severe adverse data and mild adverse data;

[0021] Step S133: Perform an intersection operation on the severe adverse data and the data of high-frequency anesthesia individuals to obtain data of high-frequency anesthesia individuals with low adaptability, and perform misjudgment correction on the data of high-frequency anesthesia individuals with low adaptability according to the short-interval anesthesia data to obtain corrected data of high-frequency anesthesia individuals with low adaptability;

[0022] Step S134: Perform an intersection operation on the mild adverse data and the data of high-frequency anesthesia individuals to obtain data of high-frequency anesthesia individuals with medium adaptability, and perform misjudgment correction on the data of high-frequency anesthesia individuals with medium adaptability according to the short-interval anesthesia data to obtain corrected data of high-frequency anesthesia individuals with medium adaptability;

[0023] Step S135: Perform intersection data elimination on the high-frequency anesthesia individual data according to the high-frequency anesthesia adaptable individual correction data and the high-frequency anesthesia low-adaptability individual correction data, so as to obtain the high-frequency anesthesia highly adaptable individual data;

[0024] Step S136: Merge the high-frequency anesthesia highly adaptable individual data, the high-frequency anesthesia low-adaptability individual correction data, and the high-frequency anesthesia medium-adaptability individual correction data, so as to obtain the high-frequency individual anesthesia adaptability data.

[0025] The present invention can identify the data with a short anesthesia interval, that is, the short-interval anesthesia data, by calculating the anesthesia interval of high-frequency anesthesia individuals and performing low-interval clustering calculation. Short-interval anesthesia will increase the anesthesia risk of patients. Therefore, identifying and paying attention to these situations are crucial for improving anesthesia management and patient safety. Obtaining the anesthesia adverse reaction rules and classifying the bad data help distinguish severe and mild anesthesia adverse reactions. Such classification can help medical staff quickly identify and take corresponding treatment measures, thereby reducing the incidence and severity of adverse events of patients. Identifying the individuals with low adaptability among high-frequency anesthesia individuals through intersection operation and performing misjudgment correction on them according to the short-interval anesthesia data helps accurately evaluate the adaptability and risk of anesthesia individuals. Such correction can help improve the accuracy and personalization level of anesthesia management and reduce potential problems and complications caused by adaptability differences. Eliminating the high-frequency anesthesia individual data according to the adaptable individual correction data can remove the individual data with poor adaptability, so as to obtain the highly adaptable individual data, which helps ensure the accuracy and safety of anesthesia management, improve the smooth progress of the surgical procedure and the postoperative effect of patients. Merging the corrected highly adaptable data with the original data can obtain more comprehensive and accurate high-frequency individual anesthesia adaptability data, which can provide a more reliable basis for future anesthesia management, promote the optimization of anesthesia operations and the personalized treatment of patients.

[0026] Optionally, step S14 is specifically as follows:

[0027] Step S141: Perform an intersection operation on the severe bad data and the low-frequency anesthesia individual data, so as to obtain the low-frequency anesthesia low-adaptability individual data;

[0028] Step S142: Perform an intersection operation on the mild bad data and the low-frequency anesthesia individual data, so as to obtain the low-frequency anesthesia medium-adaptability individual data;

[0029] Step S143: Perform intersection data elimination on the low-frequency anesthesia individual data according to the low-frequency anesthesia medium-adaptability individual data and the low-frequency anesthesia low-adaptability individual data, so as to obtain the low-frequency anesthesia highly adaptable individual data;

[0030] Step S144: Merge the low-frequency anesthesia highly adaptable individual data, low-frequency anesthesia low adaptable individual data, and low-frequency anesthesia moderately adaptable individual data to obtain low-frequency individual anesthesia adaptability data.

[0031] Through the intersection operation of the severely bad data and the low-frequency anesthesia individual data, the present invention can identify the individuals with relatively low adaptability in low-frequency anesthesia, that is, the low-frequency anesthesia low adaptable individual data, which helps the anesthesia team identify the adaptability problems existing in low-frequency anesthesia, take measures in advance to reduce potential anesthesia risks, and ensure the safety of patients. By performing an intersection operation on the slightly bad data and the low-frequency anesthesia individual data, the individuals with better adaptability in low-frequency anesthesia can be identified, that is, the low-frequency anesthesia moderately adaptable individual data, which helps to determine which patients show better adaptability in low-frequency anesthesia, can provide valuable reference for anesthesia management, and help medical staff make more personalized treatment decisions. According to the low-frequency anesthesia moderately adaptable individual data and the low-frequency anesthesia low adaptable individual data, the intersection data of the low-frequency anesthesia individual data can be excluded to obtain the individuals with better adaptability in low-frequency anesthesia, that is, the low-frequency anesthesia highly adaptable individual data, which helps to exclude the individuals with poor adaptability in low-frequency anesthesia, improve the pertinence and effectiveness of anesthesia management, and reduce the incidence of adverse events. Merging the low-frequency anesthesia highly adaptable individual data, the low-frequency anesthesia low adaptable individual data, and the low-frequency anesthesia moderately adaptable individual data can obtain comprehensive low-frequency individual anesthesia adaptability data, which helps the medical team comprehensively understand the adaptability of low-frequency anesthesia patients and provides support and reference for the formulation of surgical plans and the personalization of anesthesia management.

[0032] Optionally, step S2 is specifically as follows:

[0033] Step S21: Obtain historical medical record data, and extract anesthesia medical record data from the historical medical record data according to the historical anesthesia data of the anesthesia department to obtain anesthesia medical record data;

[0034] Step S22: Extract intraoperative monitoring data and anesthetic drug data from the anesthesia medical record data to obtain intraoperative monitoring data and anesthetic drug data;

[0035] Step S23: Perform spectral transformation on the intraoperative monitoring data to obtain intraoperative monitoring spectrum;

[0036] Step S24: Perform intraoperative anesthesia risk assessment on the intraoperative monitoring spectrum and the anesthetic drug data to obtain intraoperative anesthesia risk assessment data.

[0037] The present invention obtains historical medical record data and extracts anesthesia-related information, which helps medical staff comprehensively understand important information such as the patient's medical history, allergic reactions, surgical history, etc. Extracting based on the historical anesthesia data in the anesthesia department can obtain more detailed and accurate anesthesia medical record data, providing basic data support for subsequent anesthesia management. Extracting intraoperative monitoring data and anesthetic drug data from the anesthesia medical record data helps medical staff monitor the patient's physiological parameters and the use of anesthetic drugs in real time. These data extractions can help the medical team understand the patient's physiological state and drug reactions during the operation, and timely adjust the dosage and type of anesthetic drugs to ensure the safe progress of the operation. By performing spectral transformation on the intraoperative monitoring data, the spectral information of the patient's physiological signals during the operation can be obtained, which helps medical staff more deeply understand the patient's physiological state, discover potential abnormal changes, and take timely measures to prevent the occurrence of complications. Combining the intraoperative monitoring spectrum and anesthetic drug data for intraoperative anesthesia risk assessment can evaluate the risk of anesthesia-related complications during the operation, helping the medical team timely identify and prevent intraoperative anesthesia risks, improve the safety of the operation, and reduce the incidence of adverse events.

[0038] Optionally, step S24 is specifically as follows:

[0039] Step S241: Perform frequency fluctuation statistics on the intraoperative monitoring spectrum to obtain frequency fluctuation data;

[0040] Step S242: Perform time series correlation analysis on the frequency fluctuation data and anesthetic drug data to obtain anesthesia fluctuation correlation data;

[0041] Step S243: Calculate the fluctuation peak value based on the anesthesia fluctuation correlation data to obtain the fluctuation peak value data, and perform threshold statistics on the fluctuation peak value data to obtain the abnormal fluctuation peak value threshold;

[0042] Step S244: Perform classification calculation on the anesthesia fluctuation correlation data according to the normal fluctuation peak value threshold to obtain high-risk anesthesia risk data and low-risk anesthesia risk data;

[0043] Step S245: Merge the high-risk anesthesia risk data and low-risk anesthesia risk data to obtain intraoperative anesthesia risk assessment data.

[0044] By statistically analyzing the frequency fluctuations of the intraoperative monitoring spectrum, the present invention can identify the frequency fluctuations of the patient's physiological signals, such as heart rate, respiratory rate, etc., which helps medical staff understand the changes in the patient's physiological state and detect abnormal conditions in a timely manner. Conducting time-series correlation analysis on the frequency fluctuation data and the anesthetic drug data can explore the impact of anesthetic drugs on the patient's physiological signals. Through this analysis, the degree of influence of the use of anesthetic drugs on the patient's physiological parameters can be understood, providing guidance for the use of anesthetic drugs. Calculating the peak value of fluctuations based on the anesthetic fluctuation correlation data and setting the threshold value of abnormal fluctuation peak value helps to determine the criteria for abnormal physiological signals, helps medical staff identify potential abnormal conditions, and take timely measures for intervention. Classifying and calculating the anesthetic fluctuation correlation data according to the normal fluctuation peak value threshold divides the data into high-risk and low-risk anesthetic risk data. This classification helps the medical team more effectively identify patients who may have anesthetic-related risks, prioritize the treatment of high-risk patients, and improve the safety of the surgery. Combining the high-risk and low-risk anesthetic risk data to form comprehensive intraoperative anesthetic risk assessment data helps medical staff comprehensively understand the anesthetic risk situation during the surgery, take corresponding measures to ensure the smooth progress of the surgery, and timely handle possible complications.

[0045] Optionally, step S3 is specifically as follows:

[0046] Step S31: Extract the individual physiological characteristics from the anesthetic medical record data to obtain the individual preoperative physiological data and the individual postoperative physiological data;

[0047] Step S32: Calculate the physiological coefficient differences between the individual preoperative physiological data and the individual postoperative physiological data to obtain the high-difference physiological data and the low-difference physiological data;

[0048] Step S33: Conduct statistical analysis on the individual postoperative physiological data to obtain the normal physiological coefficient range, and classify and calculate the postoperative physiological data according to the normal physiological coefficient range to obtain the normal postoperative physiological data and the abnormal postoperative physiological data;

[0049] Step S35: Perform an intersection operation on the high-difference physiological data and the abnormal postoperative physiological data to obtain the first postoperative anesthetic high-risk data; perform an intersection operation on the low-difference physiological data and the normal postoperative physiological data to obtain the first postoperative anesthetic low-risk data;

[0050] Step S36: Perform an intersection operation on the low-difference physiological data and the abnormal postoperative physiological data to obtain the second postoperative anesthetic high-risk data; perform an intersection operation on the high-difference physiological data and the normal postoperative physiological data to obtain the second postoperative anesthetic low-risk data;

[0051] Step S37: Combine the first postoperative anesthesia high-risk data, the second postoperative anesthesia high-risk data, the first postoperative anesthesia low-risk data, and the second postoperative anesthesia low-risk data to obtain postoperative anesthesia risk assessment data.

[0052] The present invention extracts individual physiological characteristics from anesthesia medical record data, which can obtain the preoperative and postoperative physiological data of each patient, helping to understand the physiological conditions of patients in a personalized manner and providing basic data for postoperative anesthesia management. By calculating the difference in physiological coefficients between individual preoperative and postoperative physiological data, the changes in physiological parameters can be identified. Through this calculation, the change range of the postoperative physiological state of each patient can be determined, further evaluating the risk after anesthesia. Statistical analysis of individual postoperative physiological data to determine the normal range of physiological coefficients, so as to classify postoperative physiological data as normal or abnormal, helps medical staff to promptly detect postoperative physiological abnormalities of patients and take necessary intervention measures. By performing an intersection operation, combining high-difference physiological data with abnormal postoperative physiological data to identify the first postoperative anesthesia high-risk data; combining low-difference physiological data with normal postoperative physiological data to identify the first postoperative anesthesia low-risk data; and identifying the second postoperative anesthesia high-risk data and the second postoperative anesthesia low-risk data, which helps to adopt personalized anesthesia management strategies for patients with different risk levels, reducing the occurrence of postoperative complications. Combining all the identified high- and low-risk data to form comprehensive postoperative anesthesia risk assessment data helps medical staff to comprehensively evaluate the anesthesia risk of patients and formulate more effective postoperative management plans, improving the quality of postoperative recovery of patients.

[0053] Optionally, step S4 is specifically as follows:

[0054] Step S41: Classify the individual preoperative physiological data and the individual postoperative physiological data according to the individual anesthesia adaptability data to obtain anesthesia high-adaptability individual physiological data;

[0055] Step S42: Extract anesthesia-related physiological characteristics based on historical medical record data and anesthesia medical record data to obtain anesthesia-related physiological data;

[0056] Step S43: Extract physiological data from the anesthesia high-adaptability individual physiological data and the anesthesia low-adaptability individual physiological data respectively according to the anesthesia-related physiological data to obtain high-adaptability individual-related physiological data;

[0057] Step S44: Calculate the postoperative recovery offset value based on the normal physiological coefficient range for the high-adaptability individual-related physiological data to obtain postoperative recovery offset value data;

[0058] Step S45: Evaluate and correct the postoperative anesthesia risk assessment data based on the postoperative recovery offset value data, so as to obtain the corrected postoperative anesthesia risk assessment data.

[0059] The present invention classifies the preoperative and postoperative physiological data of an individual through the individual anesthesia adaptability data, and can identify individuals with high adaptability in terms of anesthesia, which helps to determine which patients have high adaptability to anesthesia, can more accurately predict their postoperative recovery conditions, and thus provides a basis for personalized postoperative management. By extracting anesthesia-related physiological characteristics from historical medical record data and anesthesia medical record data, a more comprehensive understanding of the patient's anesthesia history and related physiological indicators can be obtained, which helps to identify potential risk factors related to anesthesia and provides basic data for subsequent steps. According to the anesthesia-related physiological data, the physiological data of high-adaptability individuals and low-adaptability individuals are extracted to distinguish patients with higher and lower adaptabilities, which helps to further refine the postoperative management strategy and take personalized postoperative anesthesia management measures for patients with different adaptabilities. By calculating the postoperative recovery offset value of the physiological data of high-adaptability individuals within the normal physiological range, the deviation degree of the postoperative recovery situation can be evaluated, which helps to timely detect patients with abnormal postoperative recovery and take targeted intervention measures to reduce the occurrence of complications. According to the postoperative recovery offset value data, the postoperative anesthesia risk assessment data is corrected to more accurately evaluate the postoperative anesthesia risk of the patient, which helps to improve the accuracy and effectiveness of anesthesia management and ensure the safety and comfort of the patient after surgery.

[0060] Optionally, step S45 is specifically as follows:

[0061] Step S451: Perform correlation analysis on the postoperative anesthesia risk assessment data based on the postoperative recovery offset value data, so as to obtain the anesthesia risk assessment data to be corrected;

[0062] Step S452: Extract high-risk data from the anesthesia risk assessment data to be corrected, so as to obtain the postoperative high-risk data to be corrected;

[0063] Step S453: Evaluate and correct the postoperative high-risk data to be corrected based on the postoperative recovery offset value data, so as to obtain the corrected postoperative high-risk data;

[0064] Step S454: Replace the postoperative anesthesia risk assessment data with the corrected postoperative high-risk data, so as to obtain the corrected postoperative anesthesia risk assessment data.

[0065] The present invention performs correlation analysis on the postoperative recovery offset value data and the postoperative anesthesia risk assessment data, which can discover the potential relationship between postoperative recovery offset and anesthesia risk, help identify the impact degree of postoperative recovery offset on anesthesia risk assessment, and provide a basis for subsequent correction. Extracting high-risk data from the anesthesia risk assessment data to be corrected, that is, the data of patients who may have complications or high anesthesia risk after surgery, helps to focus on the patient groups that need key attention and prioritize their postoperative management and intervention. According to the postoperative recovery offset value data, the postoperative high-risk data to be corrected is evaluated and corrected, that is, the high-risk data is adjusted and corrected according to the actual postoperative recovery situation, which helps to more accurately evaluate the true situation of postoperative high-risk patients and improve the accuracy of risk assessment. According to the corrected postoperative high-risk data, the postoperative anesthesia risk assessment data is replaced, and the corrected data is used for the final risk assessment, which helps to ensure that the postoperative risk assessment data is more objective and accurate, provides a more reliable decision-making basis for doctors, and safeguards the safety and health of patients.

[0066] Optionally, the present invention also provides a construction system based on an anesthesia department risk prediction model for implementing a construction method based on an anesthesia department risk prediction model as described above. The construction system based on the anesthesia department risk prediction model includes:

[0067] An adaptability analysis module for obtaining historical anesthesia data of the anesthesia department and performing individual anesthesia adaptability analysis based on the historical anesthesia data of the anesthesia department to obtain individual anesthesia adaptability data;

[0068] An intraoperative risk assessment module for obtaining historical medical record data and extracting anesthesia medical record data from the historical medical record data according to the historical anesthesia data of the anesthesia department to obtain anesthesia medical record data; performing intraoperative anesthesia risk assessment based on the anesthesia medical record data to obtain intraoperative anesthesia risk assessment data;

[0069] A postoperative risk assessment module for extracting individual physiological characteristics from the anesthesia medical record data to obtain individual preoperative physiological data and individual postoperative physiological data, and performing postoperative anesthesia risk assessment on the individual preoperative physiological data and individual postoperative physiological data to obtain postoperative anesthesia risk assessment data;

[0070] A postoperative risk assessment correction module for performing postoperative recovery offset value assessment on the individual preoperative physiological data and individual postoperative physiological data according to the individual anesthesia adaptability data to obtain postoperative recovery offset value data; evaluating and correcting the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data to obtain corrected postoperative anesthesia risk assessment data;

[0071] A risk prediction model construction module is used to construct an initial anesthesia risk prediction model based on the historical anesthesia data of the anesthesiology department, and iteratively optimize and adjust the parameters of the initial anesthesia risk prediction model based on the postoperative anesthesia risk assessment correction data and the intraoperative anesthesia risk assessment data, so as to obtain the anesthesiology department risk prediction model.

[0072] The construction system of the anesthesiology department risk prediction model of the present invention can implement any construction method of the anesthesiology department risk prediction model of the present invention, and is used as a medium for coordinating the operations and signal transmissions between various modules to complete the construction method of the anesthesiology department risk prediction model. The internal modules of the system cooperate with each other, so as to more accurately predict the intraoperative and postoperative anesthesia risks of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, objects and advantages of the present invention will become more obvious:

[0074] Figure 1 It is a schematic flow chart of the steps of the construction method of the anesthesiology department risk prediction model of the present invention;

[0075] Figure 2 It is a detailed schematic flow chart of step S1 in the present invention;

[0076] Figure 3 It is a detailed schematic flow chart of step S2 in the present invention.

[0077] The realization, functional characteristics and advantages of the object of the present invention will be further described with reference to the embodiments and the drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] The technical method of the present invention will be clearly and completely described below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those skilled in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0079] In addition, the drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated descriptions will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0080] It should be understood that although terms such as "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly, the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0081] To achieve the above object, please refer to Figures 1 to 3 , the present invention provides a method for constructing an anesthesiology risk prediction model, and the method includes the following steps:

[0082] Step S1: Obtain the historical anesthetic data of the anesthesiology department, and perform individual anesthetic adaptability analysis based on the historical anesthetic data of the anesthesiology department, so as to obtain individual anesthetic adaptability data;

[0083] In this embodiment, the historical anesthetic data of the anesthesiology department is collected through the anesthesiology department database, including information such as the type of surgery of the patient, the use of anesthetic drugs, and the duration of the surgery. Then, individual anesthetic adaptability analysis is performed on these data, considering factors such as the patient's age, gender, and basic health status, to determine the adaptability of each patient to anesthesia. For example, based on the performance of similar patients in the historical data, the risk and adaptability of the current patient in a specific anesthetic situation can be evaluated, and then individual anesthetic adaptability data can be obtained.

[0084] Step S2: Obtain the historical medical record data, and extract the anesthetic medical record data from the historical medical record data according to the historical anesthetic data of the anesthesiology department, so as to obtain the anesthetic medical record data; perform intraoperative anesthetic risk assessment according to the anesthetic medical record data, so as to obtain intraoperative anesthetic risk assessment data;

[0085] In this embodiment, the historical medical record data is obtained, including the disease diagnosis, surgical record, medication situation, etc. of the patient. Then, the historical medical record data is extracted according to the historical anesthetic data of the anesthesiology department, and the information related to anesthesia is screened out to form the anesthetic medical record data. Next, these data are used for intraoperative anesthetic risk assessment, considering factors such as the patient's medical history and type of surgery, to evaluate the anesthetic risk of the patient during the surgery and generate intraoperative anesthetic risk assessment data.

[0086] Step S3: Extract the individual physiological characteristics from the anesthetic medical record data, so as to obtain the individual preoperative physiological data and the individual postoperative physiological data, and perform postoperative anesthetic risk assessment on the individual preoperative physiological data and the individual postoperative physiological data, so as to obtain postoperative anesthetic risk assessment data;

[0087] In this embodiment, individual physiological characteristics are extracted from the anesthesia medical record data, including physiological parameters of the patient such as blood pressure, heart rate, blood oxygen saturation, etc. Then, these data are used to obtain the preoperative physiological data and postoperative physiological data of each patient. Next, postoperative anesthesia risk assessment is performed based on these physiological data, considering the physiological changes of the patient after surgery, assessing the anesthesia risk of the patient after surgery, and generating postoperative anesthesia risk assessment data.

[0088] Step S4: Evaluate the postoperative recovery offset value of the individual preoperative physiological data and individual postoperative physiological data according to the individual anesthesia adaptability data, so as to obtain postoperative recovery offset value data; evaluate and correct the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data, so as to obtain postoperative anesthesia risk assessment correction data;

[0089] In this embodiment, the individual anesthesia adaptability data is used to evaluate the postoperative recovery offset value of the preoperative physiological data and postoperative physiological data of each patient. By comparing the actual postoperative recovery situation of the patient with the expected physiological state, the degree of physiological offset of the patient after surgery is evaluated, so as to obtain postoperative recovery offset value data. Then, the postoperative anesthesia risk assessment data is evaluated and corrected according to these data, the original assessment result is corrected, and more accurate postoperative anesthesia risk assessment correction data is obtained.

[0090] Step S5: Construct an initial anesthesia risk prediction model based on the historical anesthesia data of the anesthesia department, and perform iterative optimization and parameter tuning on the initial anesthesia risk prediction model based on the postoperative anesthesia risk assessment correction data and the intraoperative anesthesia risk assessment data, so as to obtain the anesthesia department risk prediction model.

[0091] In this embodiment, through machine learning algorithms such as random forest and logistic regression, an initial anesthesia risk prediction model is constructed according to the historical anesthesia data of the anesthesia department, considering factors such as the anesthesia situation of the patient and the type of surgery, and predicting the anesthesia risk of the patient. Then, the postoperative anesthesia risk assessment correction data and the intraoperative anesthesia risk assessment data are used to add new features to the initial anesthesia risk prediction model to better capture the relevant information of the postoperative and intraoperative anesthesia risks. Then, the model with new features added is iteratively optimized to make the model more accurately predict the anesthesia risk of the patient, so as to obtain the anesthesia department risk prediction model.

[0092] By obtaining historical anesthesia data in the anesthesiology department and conducting individual anesthesia adaptability analysis, the present invention can identify the characteristics and trends of individual patients from a large amount of data, providing data support for personalized anesthesia. It helps to determine the most suitable anesthesia plan for the patient based on the patient's physiological condition and historical data, reducing unnecessary risks and complications. Obtaining historical medical record data and extracting anesthesia medical record data can evaluate the intraoperative anesthesia risk, thereby detecting potential anesthesia risk factors in advance. It helps medical staff in the operation to take necessary measures in a timely manner, reducing the incidence of intraoperative complications and ensuring the smooth progress of the operation. Extracting individual physiological characteristics from anesthesia medical record data and conducting postoperative anesthesia risk assessment helps to understand the physiological condition and anesthesia response of the patient after surgery, and adjust the treatment plan in a timely manner. It helps to improve the postoperative recovery rate of the patient, reduce the occurrence of postoperative complications, and ensure the long-term effect of surgical treatment. By evaluating the postoperative recovery deviation value data, the postoperative anesthesia risk of the patient can be more accurately evaluated, and risk correction and treatment plan adjustment can be carried out in a timely manner. It helps to avoid complications and adverse consequences caused by poor postoperative recovery, and improve the postoperative survival rate and quality of life of the patient. Based on the historical anesthesia data in the anesthesiology department, an initial anesthesia risk prediction model is constructed, and the model is iteratively optimized and parameter-tuned through postoperative anesthesia risk assessment correction data and intraoperative anesthesia risk assessment data, which can improve the accuracy and stability of the model. It can help medical staff to use the model for anesthesia risk prediction more reliably, provide more effective decision-making support for surgical treatment, and maximize the safety of the patient and the success of the operation.

[0093] Optionally, step S1 is specifically as follows:

[0094] Step S11: Obtain historical anesthesia data in the anesthesiology department, and extract anesthesia adverse characteristics from the historical anesthesia data in the anesthesiology department to obtain anesthesia adverse data;

[0095] In this embodiment, historical anesthesia data is obtained from the database of the anesthesiology department or the electronic medical record system of the hospital, including the type of surgery, the use of anesthetic drugs, the operation time, the basic information of the patient, etc. Then, anesthesia adverse characteristics are extracted from these data, such as adverse events such as postoperative nausea and vomiting, hypotension during anesthesia, etc., to form an anesthesia adverse data set.

[0096] Step S12: Conduct statistical analysis on the number of individual anesthesias in the historical anesthesia data in the anesthesiology department to obtain high-frequency anesthesia individual data and low-frequency anesthesia individual data;

[0097] In this embodiment, statistical analysis on the number of individual anesthesias in the historical anesthesia data in the anesthesiology department can calculate the number of anesthesias for each patient. Grouping according to the number of anesthesias, high-frequency anesthesia individual data and low-frequency anesthesia individual data are obtained for subsequent analysis.

[0098] Step S13: Perform high-frequency individual anesthesia adaptability analysis on high-frequency anesthesia individual data based on anesthesia adverse data, so as to obtain high-frequency individual anesthesia adaptability data;

[0099] In this embodiment, anesthesia adverse data is used to perform high-frequency individual anesthesia adaptability analysis on high-frequency anesthesia individual data. This includes analyzing whether there are specific adverse events in high-frequency individual patients during multiple anesthesias, and whether there is a correlation between these events and the individual characteristics of the patients.

[0100] Step S14: Perform low-frequency individual anesthesia adaptability analysis on low-frequency anesthesia individual data based on anesthesia adverse data, so as to obtain low-frequency individual anesthesia adaptability data;

[0101] In this embodiment, anesthesia adverse data is used to perform low-frequency individual anesthesia adaptability analysis on low-frequency anesthesia individual data. The purpose of this step is to explore the possible adverse events in low-frequency individual patients during anesthesia and find the potential connections between them and individual characteristics.

[0102] Step S15: Merge the high-frequency individual anesthesia adaptability data and the low-frequency individual anesthesia adaptability data to obtain individual anesthesia adaptability data.

[0103] In this embodiment, the high-frequency individual anesthesia data and the low-frequency individual anesthesia data after adaptability analysis are merged to form a complete individual anesthesia adaptability data set. This data set will contain the anesthesia adaptability information of patients at different frequencies, providing a basis for further model training and optimization.

[0104] The present invention obtains historical anesthesia data in the anesthesiology department and extracts adverse anesthesia characteristics, which can identify adverse events, complications, and other adverse conditions during anesthesia. These adverse characteristics can provide valuable information to help identify potential anesthesia risk factors, guide clinical decision-making, improve anesthesia operation and management, and thus enhance surgical safety and the postoperative outcomes of patients. Conducting statistical analysis on the number of individual anesthesia times in the historical anesthesia data of the anesthesiology department can distinguish high-frequency and low-frequency anesthesia operations of an individual in the anesthesiology department. This helps to understand the anesthesia experience and exposure of an individual. For individuals with different frequencies of anesthesia, their anesthesia adaptabilities may vary, and targeted assessments and management are required. Analyzing the data of high-frequency anesthesia individuals based on adverse anesthesia data can evaluate the anesthesia adaptability and risk status of high-frequency anesthesia individuals, help identify specific risk factors and potential problems of high-frequency anesthesia individuals, provide personalized anesthesia strategies and management measures for them, and improve the safety and effectiveness of the surgical process. Analyzing the data of low-frequency anesthesia individuals based on adverse anesthesia data can evaluate the anesthesia adaptability and risk status of low-frequency anesthesia individuals, help discover special problems and potential risks of low-frequency anesthesia individuals, provide more refined anesthesia management and precautions for them, and avoid potential complications and side effects. Combining the anesthesia adaptability data of high-frequency and low-frequency individuals can obtain more comprehensive and integrated individual anesthesia adaptability data, which helps to comprehensively understand the anesthesia adaptability characteristics, risk status, and personalized needs of an individual, and provide more accurate data support for formulating individualized anesthesia plans and risk prediction models.

[0105] Optionally, step S13 is specifically as follows:

[0106] Step S131: Calculate the anesthesia intervals for the data of high-frequency anesthesia individuals to obtain anesthesia interval data, and perform low-interval clustering calculation on the anesthesia interval data to obtain short-interval anesthesia data;

[0107] In this embodiment, for the data of high-frequency anesthesia individuals, first, the calculation method of anesthesia intervals needs to be determined, which can adopt timestamp or date-time format to calculate the time difference between adjacent anesthesia events. Then, using a clustering algorithm such as the K-means algorithm, group the data points according to the characteristics of anesthesia intervals, and find the data points with shorter intervals. These data points may represent short-interval anesthesia data.

[0108] Step S132: Obtain anesthesia adverse reaction rules, and classify the degree of adverse effects of the anesthesia adverse data according to the anesthesia adverse reaction rules to obtain severe adverse data and mild adverse data;

[0109] In this embodiment, rules for anesthetic adverse reactions are obtained, such as postoperative nausea and vomiting, allergic reactions, etc. After obtaining the rules for anesthetic adverse reactions, rule extraction and modeling can be performed through natural language processing techniques or expert knowledge. For example, for the rule of postoperative nausea and vomiting, it may include situations such as having nausea and vomiting more than 2 times within 24 hours after surgery. Then, historical anesthetic data is screened and classified, and the adverse events among them are classified as severe adverse data or minor adverse data. For example, severe adverse reactions may include respiratory depression, etc.

[0110] Step S133: Perform an intersection operation on the severe adverse data and the high-frequency anesthetic individual data to obtain high-frequency anesthetic low-adaptability individual data, and perform misjudgment correction on the high-frequency anesthetic low-adaptability individual data according to the short-interval anesthetic data, so as to obtain high-frequency anesthetic low-adaptability individual corrected data;

[0111] In this embodiment, when performing the intersection operation, it is first necessary to determine the data formats and data fields of the severe adverse data and the high-frequency anesthetic individual data to ensure effective matching. Then, through the intersection operation, individual data with both severe adverse reactions and high-frequency anesthetic characteristics are found, that is, high-frequency anesthetic low-adaptability individual data. Next, the short-interval anesthetic data is used to perform misjudgment correction on these individual data. Machine learning methods such as logistic regression models or support vector machines can be used to exclude adverse reactions caused by too short anesthetic intervals, so as to obtain high-frequency anesthetic low-adaptability individual corrected data.

[0112] Step S134: Perform an intersection operation on the minor adverse data and the high-frequency anesthetic individual data to obtain high-frequency anesthetic medium-adaptability individual data, and perform misjudgment correction on the high-frequency anesthetic medium-adaptability individual data according to the short-interval anesthetic data, so as to obtain high-frequency anesthetic medium-adaptability individual corrected data;

[0113] In this embodiment, an intersection operation is performed on the minor adverse data and the high-frequency anesthetic individual data to find high-frequency anesthetic medium-adaptability individual data. Then, the short-interval anesthetic data is used to perform misjudgment correction on these individual data. Machine learning methods or statistical methods are also used to exclude adverse reactions caused by too short anesthetic intervals to obtain high-frequency anesthetic medium-adaptability individual corrected data.

[0114] Step S135: Perform intersection data elimination on the high-frequency anesthetic individual data according to the high-frequency anesthetic medium-adaptability individual corrected data and the high-frequency anesthetic low-adaptability individual corrected data, so as to obtain high-frequency anesthetic high-adaptability individual data;

[0115] In this embodiment, when removing intersection data from the individual data of high-frequency anesthesia, it is necessary to consider the consistency and integrity of the data to ensure that the removed data is logical. Database queries or data processing software can be used for operations to remove the corrected data of adaptable individuals and the corrected data of low-adaptability individuals in high-frequency anesthesia from the individual data of high-frequency anesthesia, obtaining the high-adaptability individual data of high-frequency anesthesia.

[0116] Step S136: Merge the high-adaptability individual data of high-frequency anesthesia, the corrected data of low-adaptability individuals in high-frequency anesthesia, and the corrected data of medium-adaptability individuals in high-frequency anesthesia to obtain the high-frequency individual anesthesia adaptability data.

[0117] In this embodiment, the obtained high-adaptability individual data of high-frequency anesthesia, the corrected data of low-adaptability individuals in high-frequency anesthesia, and the corrected data of medium-adaptability individuals in high-frequency anesthesia can be merged through data connection or merging operations to form a high-frequency individual anesthesia adaptability data set, providing a basis for further analysis and application.

[0118] The present invention can identify data with short anesthesia intervals, i.e., short-interval anesthesia data, by calculating the anesthesia intervals of high-frequency anesthesia individuals and performing low-interval clustering calculations. Short-interval anesthesia increases the anesthesia risk of patients. Therefore, identifying and paying attention to these situations are crucial for improving anesthesia management and patient safety. Obtaining anesthesia adverse reaction rules and classifying bad data help distinguish severe and mild anesthesia adverse reactions. This classification can help medical staff quickly identify and take corresponding treatment measures, thereby reducing the incidence and severity of adverse events in patients. Identifying individuals with low adaptability among high-frequency anesthesia individuals through intersection operations and correcting their misjudgments based on short-interval anesthesia data help accurately evaluate the adaptability and risk of anesthesia individuals. Such correction can help improve the accuracy and personalization level of anesthesia management, reducing potential problems and complications caused by adaptability differences. Removing the individual data of low-adaptability individuals from the high-frequency anesthesia individual data according to the corrected data of adaptable individuals can obtain high-adaptability individual data, which helps ensure the accuracy and safety of anesthesia management, improve the smooth progress of the surgical procedure, and the postoperative effect of patients. Merging the corrected high-adaptability data with the original data can obtain more comprehensive and accurate high-frequency individual anesthesia adaptability data, which can provide a more reliable basis for future anesthesia management, promoting the optimization of anesthesia operations and the personalized treatment of patients.

[0119] Optionally, step S14 is specifically as follows:

[0120] Step S141: Perform an intersection operation on the severe bad data and the individual data of low-frequency anesthesia to obtain the low-adaptability individual data of low-frequency anesthesia;

[0121] In this embodiment, data query language or specialized data processing tools are used to screen out individual data of low-frequency anesthesia from severely adverse data. Then, an intersection operation is performed on these individual data of low-frequency anesthesia and the severely adverse data to identify the individual data that coexist between the two. Through this intersection operation, individual data of low adaptability in low-frequency anesthesia can be obtained, that is, those individuals who show low adaptability in low-frequency anesthesia.

[0122] Step S142: Perform an intersection operation on the slightly adverse data and the individual data of low-frequency anesthesia to obtain individual data of medium adaptability in low-frequency anesthesia;

[0123] In this embodiment, similar operations are performed on the slightly adverse data and the individual data of low-frequency anesthesia. First, the individual data of low-frequency anesthesia are extracted from the slightly adverse data, and then an intersection operation is performed on these data and the individual data of low-frequency anesthesia. In this way, the individuals who show medium adaptability in low-frequency anesthesia can be determined, and thus the individual data of medium adaptability in low-frequency anesthesia can be obtained.

[0124] Step S143: Perform intersection data elimination on the individual data of low-frequency anesthesia according to the individual data of medium adaptability in low-frequency anesthesia and the individual data of low adaptability in low-frequency anesthesia to obtain individual data of high adaptability in low-frequency anesthesia;

[0125] In this embodiment, after obtaining the individual data of low adaptability in low-frequency anesthesia and the individual data of medium adaptability in low-frequency anesthesia, the individual data of low-frequency anesthesia can be further processed. By taking the intersection of the individual data of medium adaptability in low-frequency anesthesia and the individual data of low adaptability in low-frequency anesthesia, those individuals who show low adaptability in low-frequency anesthesia can be eliminated, so as to obtain the individual data of high adaptability in low-frequency anesthesia.

[0126] Step S144: Merge the individual data of high adaptability in low-frequency anesthesia, the individual data of low adaptability in low-frequency anesthesia, and the individual data of medium adaptability in low-frequency anesthesia to obtain individual data of low-frequency anesthesia adaptability.

[0127] In this embodiment, the obtained individual data of high adaptability in low-frequency anesthesia, the individual data of low adaptability in low-frequency anesthesia, and the individual data of medium adaptability in low-frequency anesthesia are merged. Through the data merging operation, a complete dataset of individual low-frequency anesthesia adaptability can be constructed, which includes individual data at different adaptability levels and provides a basis for subsequent analysis and application.

[0128] By performing an intersection operation on severely adverse data and individual data of low-frequency anesthesia, the present invention can identify individuals with low adaptability in low-frequency anesthesia, that is, individual data of low adaptability in low-frequency anesthesia, which helps the anesthesia team identify adaptability problems existing in low-frequency anesthesia, take measures in advance to reduce potential anesthesia risks, and ensure the safety of patients. By performing an intersection operation on slightly adverse data and individual data of low-frequency anesthesia, individuals with better adaptability in low-frequency anesthesia can be identified, that is, individual data of medium adaptability in low-frequency anesthesia, which helps to determine which patients show better adaptability in low-frequency anesthesia, can provide valuable reference for anesthesia management, and help medical staff make more personalized treatment decisions. By performing intersection data elimination on individual data of low-frequency anesthesia according to individual data of medium adaptability and individual data of low adaptability in low-frequency anesthesia, individuals with better adaptability in low-frequency anesthesia can be obtained, that is, individual data of high adaptability in low-frequency anesthesia, which helps to exclude individuals with poor adaptability in low-frequency anesthesia, improve the pertinence and effectiveness of anesthesia management, and reduce the incidence of adverse events. By combining individual data of high adaptability in low-frequency anesthesia, individual data of low adaptability in low-frequency anesthesia, and individual data of medium adaptability in low-frequency anesthesia, comprehensive individual anesthesia adaptability data of low frequency can be obtained, which helps the medical team fully understand the adaptability of patients undergoing low-frequency anesthesia and provides support and reference for the formulation of surgical plans and the personalization of anesthesia management.

[0129] Optionally, step S2 is specifically as follows:

[0130] Step S21: Obtain historical medical record data, and extract anesthetic medical record data from the historical medical record data according to the historical anesthetic data of the anesthesiology department, so as to obtain anesthetic medical record data;

[0131] In this embodiment, the historical medical record data of the patient is collected, which may include information such as medical records, surgical records, and diagnostic reports. Then, using the historical anesthetic data of the anesthesiology department, such as anesthetic record sheets and anesthetic plans, the historical medical record data is screened and extracted to obtain information related to anesthesia, such as the anesthetic method, anesthetic drug usage, and anesthetic effect of the patient, so as to form complete anesthetic medical record data.

[0132] Step S22: Extract intraoperative monitoring data and anesthetic drug data from the anesthetic medical record data, so as to obtain intraoperative monitoring data and anesthetic drug data;

[0133] In this embodiment, the obtained anesthetic medical record data is further processed to extract intraoperative monitoring data and anesthetic drug data therefrom. The intraoperative monitoring data may include the patient's vital sign monitoring data, such as heart rate, blood pressure, respiratory rate, etc., and various monitoring indicators during the surgical process. At the same time, anesthetic drug data is extracted, including information such as the type, dosage, and administration route of anesthetic drugs, for subsequent analysis and evaluation.

[0134] Step S23: Perform spectral transformation on the intraoperative monitoring data to obtain the intraoperative monitoring spectrum;

[0135] In this embodiment, for the extracted intraoperative monitoring data, spectral transformation processing is performed. This can be achieved by using Fourier transform or other frequency-domain analysis methods to convert the time-domain monitoring data into frequency-domain data, thereby obtaining the intraoperative monitoring spectrum. This spectrum can reflect the physiological state changes of the patient in different frequency bands during the operation, providing a basis for subsequent risk assessment.

[0136] Step S24: Perform intraoperative anesthesia risk assessment on the intraoperative monitoring spectrum and anesthesia drug data to obtain intraoperative anesthesia risk assessment data.

[0137] In this embodiment, the intraoperative monitoring spectrum data and anesthesia drug data are combined for intraoperative anesthesia risk assessment. This may involve using machine learning algorithms, statistical analysis methods, or professional medical evaluation models, comprehensively considering the changes in physiological indicators in the monitoring data and the effects of anesthesia drugs, and evaluating the risk level of the patient during anesthesia, thereby providing timely reference and decision support for clinicians.

[0138] The present invention obtains historical medical record data and extracts anesthesia-related information, which helps medical staff comprehensively understand important information such as the patient's medical history, allergic reactions, and surgical history. By extracting according to the historical anesthesia data in the anesthesia department, more detailed and accurate anesthesia medical record data can be obtained, providing basic data support for subsequent anesthesia management. Extracting the intraoperative monitoring data and anesthesia drug data in the anesthesia medical record data helps medical staff monitor the patient's physiological parameters and the use of anesthesia drugs in real time. These data extractions can help the medical team understand the patient's physiological state and drug reactions during the operation, timely adjust the dosage and type of anesthesia drugs, and ensure the safe progress of the operation. By performing spectral transformation on the intraoperative monitoring data, the physiological signal spectrum information of the patient during the operation can be obtained, which helps medical staff more deeply understand the patient's physiological state, discover potential abnormal changes, and take timely measures to prevent the occurrence of complications. Combining the intraoperative monitoring spectrum and anesthesia drug data for intraoperative anesthesia risk assessment can evaluate the risk of anesthesia-related complications during the operation, helping the medical team timely identify and prevent intraoperative anesthesia risks, improve the safety of the operation, and reduce the incidence of adverse events.

[0139] Optionally, step S24 is specifically:

[0140] Step S241: Perform frequency fluctuation statistics on the intraoperative monitoring spectrum to obtain frequency fluctuation data;

[0141] In this embodiment, when performing frequency fluctuation statistics on the intraoperative monitoring spectrum, the spectrum data is first segmented according to a certain time window, and then the frequency fluctuation of the spectrum data within each time window is calculated. This may include calculating statistical indicators such as the standard deviation and root mean square value of the frequency to reflect the temporal variation of the spectrum, thereby obtaining frequency fluctuation data.

[0142] Step S242: Perform time series correlation analysis on the frequency fluctuation data and the anesthetic drug data to obtain anesthetic fluctuation correlation data;

[0143] In this embodiment, after obtaining the frequency fluctuation data, it is subjected to time series correlation analysis with the anesthetic drug data. By comparing and analyzing the frequency fluctuation data with the recorded anesthetic drug usage at the same time, the correlation between them can be explored. This may involve using time series analysis methods such as autocorrelation function and cross-correlation function to determine the association between frequency fluctuation and anesthetic drug usage, thereby obtaining anesthetic fluctuation correlation data.

[0144] Step S243: Calculate the fluctuation peak value based on the anesthetic fluctuation correlation data to obtain the fluctuation peak value data, and perform threshold statistics on the fluctuation peak value data to obtain the abnormal fluctuation peak value threshold;

[0145] In this embodiment, after obtaining the anesthetic fluctuation correlation data, the calculation of the fluctuation peak value is performed. First, the peak values in the anesthetic fluctuation data are identified, and these peak values may represent the abnormal physiological states of the patient in certain situations. Then, statistical analysis is performed on these peak values to determine the threshold of the abnormal fluctuation peak value. The setting of this threshold can be based on the statistical characteristics of the normal population data to help distinguish normal fluctuations from abnormal fluctuations.

[0146] Step S244: Perform classification calculation on the anesthetic fluctuation correlation data according to the normal fluctuation peak value threshold to obtain high-risk anesthetic risk data and low-risk anesthetic risk data;

[0147] In this embodiment, based on the determined normal fluctuation peak value threshold, classification calculation is performed on the anesthetic fluctuation correlation data, which is divided into high-risk anesthetic risk data and low-risk anesthetic risk data. By comparing the peak values in the anesthetic fluctuation data with the set threshold, the anesthetic risk level of the patient can be judged. This will help doctors promptly identify potential risks during anesthesia and take corresponding measures for treatment.

[0148] Step S245: Merge the high-risk anesthetic risk data and the low-risk anesthetic risk data to obtain intraoperative anesthetic risk assessment data.

[0149] In this embodiment, high-risk anesthesia risk data and low-risk anesthesia risk data are merged to obtain complete intraoperative anesthesia risk assessment data. These assessment data will provide comprehensive anesthesia risk information for doctors, helping them make more accurate clinical judgments and decisions.

[0150] By statistically analyzing the frequency fluctuations of the intraoperative monitoring spectrum, the present invention can identify the frequency fluctuations of the patient's physiological signals, such as heart rate, respiratory rate, etc., which helps medical staff understand the changes in the patient's physiological state and detect abnormal conditions in a timely manner. Performing time-series correlation analysis on the frequency fluctuation data and the anesthesia drug data can explore the impact of anesthesia drugs on the patient's physiological signals. Through this analysis, the degree of influence of the use of anesthesia drugs on the patient's physiological parameters can be understood, providing guidance for the use of anesthesia drugs. Calculating the peak value of the fluctuation and setting the abnormal fluctuation peak value threshold based on the anesthesia fluctuation correlation data helps determine the criteria for abnormal physiological signals, helps medical staff identify potential abnormal conditions, and take timely measures for intervention. Classifying and calculating the anesthesia fluctuation correlation data according to the normal fluctuation peak value threshold divides the data into high-risk and low-risk anesthesia risk data. This classification helps the medical team more effectively identify patients who may have anesthesia-related risks, prioritize the treatment of high-risk patients, and improve the safety of the surgery. Merging high-risk and low-risk anesthesia risk data to form comprehensive intraoperative anesthesia risk assessment data helps medical staff comprehensively understand the anesthesia risk situation during the operation, take corresponding measures to ensure the smooth progress of the operation, and timely handle possible complications.

[0151] Optionally, step S3 is specifically as follows:

[0152] Step S31: Extract individual physiological characteristics from the anesthesia medical record data to obtain individual preoperative physiological data and individual postoperative physiological data;

[0153] In this embodiment, individual physiological characteristics are extracted from the anesthesia medical record data. For example, preoperative physiological information such as the patient's age, weight, height, etc. is extracted from the medical record, and the preoperative physiological data is recorded. (For example: preoperative physiological data such as the patient's age, weight, blood pressure, heart rate, etc. are obtained through the anesthesia record sheet and the medical record system, such as: age is 45 years old, weight is 70 kg, blood pressure is 120 / 80 mmHg, heart rate is 70 beats per minute)

[0154] Step S32: Calculate the physiological coefficient difference between the individual preoperative physiological data and the individual postoperative physiological data to obtain high-difference physiological data and low-difference physiological data;

[0155] In this embodiment, the physiological coefficient differences between the preoperative physiological data and the postoperative physiological data of an individual are calculated. For example, the difference between the postoperative physiological data and the preoperative physiological data is calculated, such as the change value of blood pressure. Then, based on these difference data, the threshold of high difference values is statistically obtained, and the difference data are classified into high difference values and low difference values by using the threshold (for example: calculate the difference between the postoperative blood pressure and the preoperative blood pressure, and the change value is the postoperative blood pressure minus the preoperative blood pressure, resulting in a blood pressure change value of 10 / 5 mmHg).

[0156] Step S33: Statistically analyze the postoperative physiological data of the individual to obtain the normal physiological coefficient range, and classify and calculate the postoperative physiological data according to the normal physiological coefficient range to obtain the normal postoperative physiological data and the abnormal postoperative physiological data;

[0157] In this embodiment, the postoperative physiological data of the individual is statistically analyzed to obtain the normal physiological coefficient range, and the postoperative physiological data is classified and calculated according to it to obtain the normal postoperative physiological data and the abnormal postoperative physiological data. (For example: statistically analyze the postoperative blood pressure data, calculate the average value and standard deviation of the postoperative blood pressure, and determine the normal physiological blood pressure range as 90 - 130 / 60 - 90 mmHg according to the statistical characteristics of the normal population. Classify according to whether the blood pressure data is within this range to obtain the normal postoperative blood pressure data and the abnormal postoperative blood pressure data.)

[0158] Step S35: Perform an intersection operation on the high-difference physiological data and the abnormal postoperative physiological data to obtain the first postoperative anesthesia high-risk data; perform an intersection operation on the low-difference physiological data and the normal postoperative physiological data to obtain the first postoperative anesthesia low-risk data;

[0159] In this embodiment, an intersection operation is performed on the high-difference physiological data and the abnormal postoperative physiological data to obtain the first postoperative anesthesia high-risk data; an intersection operation is performed on the low-difference physiological data and the normal postoperative physiological data to obtain the first postoperative anesthesia low-risk data. (For example: screen out the high-difference physiological data, such as the data with blood pressure change exceeding the normal range, and perform an intersection operation with the abnormal postoperative physiological data to obtain the first postoperative anesthesia high-risk data. In addition, screen out the low-difference physiological data, such as the data with blood pressure change within the normal range, and perform an intersection operation with the normal postoperative physiological data to obtain the first postoperative anesthesia low-risk data.)

[0160] Step S36: Perform an intersection operation on the low-difference physiological data and the abnormal postoperative physiological data to obtain the second postoperative anesthesia high-risk data; perform an intersection operation on the high-difference physiological data and the normal postoperative physiological data to obtain the second postoperative anesthesia low-risk data;

[0161] In this embodiment, the intersection operation is performed on the low-difference physiological data and the abnormal postoperative physiological data to obtain the second high-risk data for postoperative anesthesia; the intersection operation is performed on the high-difference physiological data and the normal postoperative physiological data to obtain the second low-risk data for postoperative anesthesia. (For example: screening the low-difference physiological data, such as the data with blood pressure changes within the normal range, and performing the intersection operation with the abnormal postoperative physiological data to obtain the second high-risk data for postoperative anesthesia. In addition, screening the high-difference physiological data, such as the data with blood pressure changes exceeding the normal range, and performing the intersection operation with the normal postoperative physiological data to obtain the second low-risk data for postoperative anesthesia.)

[0162] Step S37: Merge the first high-risk data for postoperative anesthesia, the second high-risk data for postoperative anesthesia, the first low-risk data for postoperative anesthesia, and the second low-risk data for postoperative anesthesia to obtain the risk assessment data for postoperative anesthesia.

[0163] In this embodiment, the first high-risk data for postoperative anesthesia, the second high-risk data for postoperative anesthesia, the first low-risk data for postoperative anesthesia, and the second low-risk data for postoperative anesthesia are merged to obtain the risk assessment data for postoperative anesthesia, forming a complete risk assessment data for postoperative anesthesia, so that doctors can comprehensively evaluate and make decisions on the anesthesia risk of patients.

[0164] The present invention extracts individual physiological characteristics from anesthesia medical record data, can obtain the preoperative and postoperative physiological data of each patient, helps to understand the physiological conditions of patients in a personalized manner, and provides basic data for postoperative anesthesia management. Calculating the physiological coefficient difference of individual preoperative and postoperative physiological data can identify the changes in physiological parameters. Through this calculation, the change range of the postoperative physiological state of each patient can be determined, and the risk after anesthesia can be further evaluated. Statistical analysis of individual postoperative physiological data to determine the normal physiological coefficient range, so as to classify the postoperative physiological data as normal or abnormal, helps medical staff to timely detect postoperative physiological abnormalities of patients and take necessary intervention measures. Through the intersection operation, combining the high-difference physiological data with the abnormal postoperative physiological data to identify the first high-risk data for postoperative anesthesia; combining the low-difference physiological data with the normal postoperative physiological data to identify the first low-risk data for postoperative anesthesia; and identifying the second high-risk data for postoperative anesthesia and the second low-risk data for postoperative anesthesia, which helps to adopt personalized anesthesia management strategies for patients with different risk levels, reduce the occurrence of postoperative complications. Merging all the identified high- and low-risk data to form comprehensive risk assessment data for postoperative anesthesia helps medical staff to comprehensively evaluate the anesthesia risk of patients and formulate more effective postoperative management plans, improving the postoperative recovery quality of patients.

[0165] Optionally, step S4 is specifically as follows:

[0166] Step S41: Classify the preoperative physiological data and postoperative physiological data of the individual according to the individual anesthesia adaptability data, so as to obtain the physiological data of highly adaptable individuals for anesthesia;

[0167] In this embodiment, according to the individual anesthesia adaptability data, such as the patient's basic health status, drug allergy history, etc., the preoperative physiological data and postoperative physiological data of the individual are classified. For example, for a certain patient, if his health status is good and he has no drug allergy history, he is classified as an individual with high anesthesia adaptability, and his relevant physiological data is recorded.

[0168] Step S42: Extract anesthesia-related physiological characteristics according to the historical medical record data and anesthesia medical record data, so as to obtain anesthesia-related physiological data;

[0169] In this embodiment, anesthesia-related physiological characteristics are extracted according to the historical medical record data and anesthesia medical record data, including physiological indicators before surgery, physiological changes during surgery, etc. For example, data such as the patient's blood pressure and heart rate before surgery are extracted from the medical record, and the drug use situation during the surgery is recorded.

[0170] Step S43: Respectively extract physiological data from the physiological data of highly adaptable individuals for anesthesia and the physiological data of individuals with low anesthesia adaptability according to the anesthesia-related physiological data, so as to obtain relevant physiological data of highly adaptable individuals;

[0171] In this embodiment, according to the anesthesia-related physiological data, physiological data are respectively extracted from highly adaptable individuals for anesthesia and individuals with low anesthesia adaptability. For example, for highly adaptable individuals for anesthesia, data such as blood pressure and heart rate before, during, and after surgery are extracted; for individuals with low anesthesia adaptability, similar physiological data are also extracted.

[0172] Step S44: Calculate the postoperative recovery offset value according to the relevant physiological data of highly adaptable individuals within the range of normal physiological coefficients, so as to obtain postoperative recovery offset value data;

[0173] In this embodiment, according to the range of normal physiological coefficients, the postoperative recovery offset value is calculated for the relevant physiological data of highly adaptable individuals for anesthesia. For example, the offset values of indicators such as postoperative blood pressure and heart rate of the patient from the normal range are calculated to evaluate his postoperative physiological recovery situation.

[0174] Step S45: Evaluate and correct the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data, so as to obtain postoperative anesthesia risk assessment correction data.

[0175] In this embodiment, the postoperative anesthesia risk assessment data is evaluated and corrected according to the postoperative recovery offset value data. For example, by combining the offset of the patient's postoperative physiological data from the normal range, the postoperative anesthesia risk assessment is corrected to more accurately evaluate the patient's postoperative anesthesia risk.

[0176] The present invention classifies the preoperative and postoperative physiological data of an individual through the individual anesthesia adaptability data, and can identify individuals with high adaptability in terms of anesthesia, which helps to determine which patients have a higher adaptability to anesthesia, can more accurately predict their postoperative recovery, and thus provides a basis for personalized postoperative management. By extracting anesthesia-related physiological characteristics from historical medical record data and anesthesia medical record data, a more comprehensive understanding of the patient's anesthesia history and related physiological indicators can be obtained, which helps to identify potential risk factors related to anesthesia and provides basic data for subsequent steps. According to the anesthesia-related physiological data, the physiological data of high-adaptability individuals and low-adaptability individuals are extracted to distinguish patients with higher and lower adaptabilities, which helps to further refine the postoperative management strategy and take personalized postoperative anesthesia management measures for patients with different adaptabilities. By calculating the postoperative recovery offset value of the physiological data related to high-adaptability individuals within the normal physiological range, the deviation degree of the postoperative recovery can be evaluated, which helps to timely detect patients with abnormal postoperative recovery and take targeted intervention measures to reduce the occurrence of complications. According to the postoperative recovery offset value data, the postoperative anesthesia risk assessment data is corrected to more accurately evaluate the patient's postoperative anesthesia risk, which helps to improve the accuracy and effectiveness of anesthesia management and ensure the safety and comfort of the patient after surgery.

[0177] Optionally, step S45 is specifically as follows:

[0178] Step S451: Perform correlation analysis on the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data to obtain the anesthesia risk assessment data to be corrected;

[0179] In this embodiment, the postoperative recovery offset value data is correlated with the postoperative anesthesia risk assessment data to determine the relationship between the postoperative recovery and the anesthesia risk. For example, through statistical methods or machine learning algorithms, the correlation between the postoperative recovery offset value and the incidence of anesthesia complications is analyzed to obtain the anesthesia risk assessment data to be corrected.

[0180] Step S452: Extract high-risk data from the anesthesia risk assessment data to be corrected to obtain the postoperative high-risk data to be corrected;

[0181] In this embodiment, high-risk data is extracted from the anesthesia risk assessment data to be corrected, and these data may indicate possible complications or high-risk situations that may occur after the operation for the patient. For example, factors that may lead to postoperative complications are identified from the anesthesia record, such as abnormal physiological changes during the operation or poor postoperative recovery.

[0182] Step S453: Evaluate and correct the postoperative high-risk data to be corrected according to the postoperative recovery offset value data, so as to obtain the corrected postoperative high-risk data;

[0183] In this embodiment, the postoperative high-risk data to be corrected is evaluated and corrected according to the postoperative recovery offset value data. This involves taking the postoperative recovery offset value data, that is, the postoperative recovery situation, into account in the high-risk data to more accurately evaluate the postoperative risk of the patient. For example, the high-risk data is re-evaluated in combination with the postoperative physiological recovery situation to correct possible risk predictions.

[0184] Step S454: Replace the postoperative anesthesia risk assessment data with the corrected postoperative high-risk data, so as to obtain the corrected postoperative anesthesia risk assessment data.

[0185] In this embodiment, the original postoperative anesthesia risk assessment data is replaced with the corrected postoperative high-risk data to obtain the corrected postoperative anesthesia risk assessment data. This ensures the accuracy and reliability of the postoperative anesthesia risk assessment, enabling the medical team to better formulate postoperative management strategies and monitoring plans to ensure the safety and health of the patient.

[0186] The present invention conducts a correlation analysis between the postoperative recovery offset value data and the postoperative anesthesia risk assessment data, and can discover the potential relationship between the postoperative recovery offset and the anesthesia risk, which helps to identify the degree of influence of the postoperative recovery offset on the anesthesia risk assessment and provides a basis for subsequent correction. Extracting high-risk data from the anesthesia risk assessment data to be corrected, that is, the data of patients who may have complications or high anesthesia risk after the operation, helps to focus on the patient group that needs to be focused on and prioritize their postoperative management and intervention. Evaluating and correcting the postoperative high-risk data to be corrected according to the postoperative recovery offset value data, that is, adjusting and correcting the high-risk data according to the actual postoperative recovery situation, helps to more accurately evaluate the true situation of postoperative high-risk patients and improve the accuracy of risk assessment. Replacing the postoperative anesthesia risk assessment data with the corrected postoperative high-risk data and using the corrected data for the final risk assessment helps to ensure that the postoperative risk assessment data is more objective and accurate, providing a more reliable decision-making basis for doctors to ensure the safety and health of the patient.

[0187] Optionally, the present invention further provides a construction system for an anesthesiology risk prediction model, which is used to execute the construction method of the anesthesiology risk prediction model as described above. The construction system for the anesthesiology risk prediction model includes:

[0188] An adaptability analysis module, configured to obtain historical anesthesiology data of the anesthesiology department, and perform individual anesthesia adaptability analysis based on the historical anesthesiology data of the anesthesiology department, so as to obtain individual anesthesia adaptability data;

[0189] An intraoperative risk assessment module, configured to obtain historical medical record data, and extract anesthetic medical record data from the historical medical record data according to the historical anesthesiology data of the anesthesiology department, so as to obtain anesthetic medical record data; perform intraoperative anesthesia risk assessment based on the anesthetic medical record data, so as to obtain intraoperative anesthesia risk assessment data;

[0190] A postoperative risk assessment module, configured to extract individual physiological characteristics from the anesthetic medical record data, so as to obtain individual preoperative physiological data and individual postoperative physiological data, and perform postoperative anesthesia risk assessment on the individual preoperative physiological data and the individual postoperative physiological data, so as to obtain postoperative anesthesia risk assessment data;

[0191] A postoperative risk assessment correction module, configured to perform postoperative recovery offset value assessment on the individual preoperative physiological data and the individual postoperative physiological data according to the individual anesthesia adaptability data, so as to obtain postoperative recovery offset value data; perform assessment and correction on the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data, so as to obtain postoperative anesthesia risk assessment correction data;

[0192] A risk prediction model construction module, configured to construct an initial anesthesia risk prediction model based on the historical anesthesiology data of the anesthesiology department, and perform iterative optimization and parameter adjustment on the initial anesthesia risk prediction model based on the postoperative anesthesia risk assessment correction data and the intraoperative anesthesia risk assessment data, so as to obtain an anesthesiology risk prediction model.

[0193] The construction system for the anesthesiology risk prediction model of the present invention can implement any construction method of the anesthesiology risk prediction model of the present invention, and is used as a medium for coordinating the operations and signal transmissions between each module to complete the construction method of the anesthesiology risk prediction model. The internal modules of the system cooperate with each other, so as to more accurately predict the intraoperative and postoperative anesthesia risks of patients.

[0194] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be included in the present invention.

[0195] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A construction method of an anesthesia department risk prediction model, characterized in that, Including the following steps: Step S1: Obtain the historical anesthesia data of the anesthesiology department, and conduct individual anesthesia adaptability analysis based on the historical anesthesia data of the anesthesiology department to obtain individual anesthesia adaptability data; Step S2: Obtain the historical medical record data, and extract the anesthesia medical record data from the historical medical record data according to the historical anesthesia data of the anesthesiology department to obtain the anesthesia medical record data; conduct intraoperative anesthesia risk assessment based on the anesthesia medical record data to obtain intraoperative anesthesia risk assessment data; Step S3: Extract individual physiological characteristics from the anesthesia medical record data to obtain individual preoperative physiological data and individual postoperative physiological data, and conduct postoperative anesthesia risk assessment on the individual preoperative physiological data and individual postoperative physiological data to obtain postoperative anesthesia risk assessment data; Step S4: Conduct postoperative recovery offset value assessment on the individual preoperative physiological data and individual postoperative physiological data according to the individual anesthesia adaptability data to obtain postoperative recovery offset value data; conduct assessment and correction on the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data to obtain postoperative anesthesia risk assessment correction data; Step S4 is specifically: Step S41: Conduct classification of high-adaptability individual data on the individual preoperative physiological data and individual postoperative physiological data according to the individual anesthesia adaptability data to obtain anesthetic high-adaptability individual physiological data; Step S42: Extract anesthesia-related physiological characteristics according to the historical medical record data and the anesthesia medical record data to obtain anesthesia-related physiological data; Step S43: Extract physiological data from the anesthetic high-adaptability individual physiological data according to the anesthesia-related physiological data to obtain high-adaptability individual-related physiological data; Step S44: Calculate the postoperative recovery offset value for the high-adaptability individual-related physiological data according to the normal physiological coefficient range to obtain the postoperative recovery offset value data; Step S45: Conduct assessment and correction on the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data to obtain postoperative anesthesia risk assessment correction data; Step S45 is specifically: Step S451: Conduct correlation analysis on the postoperative anesthesia risk assessment data according to the postoperative recovery offset value data to obtain the anesthesia risk assessment data to be corrected; Step S452: Extract high-risk data from the anesthesia risk assessment data to be corrected to obtain the postoperative high-risk data to be corrected; Step S453: Conduct assessment and correction on the postoperative high-risk data to be corrected according to the postoperative recovery offset value data to obtain postoperative high-risk correction data; Step S454: Conduct data replacement on the postoperative anesthesia risk assessment data according to the postoperative high-risk correction data to obtain postoperative anesthesia risk assessment correction data; Step S5: Construct an initial anesthesia risk prediction model based on the historical anesthesia data of the anesthesiology department, and conduct iterative optimization and parameter adjustment on the initial anesthesia risk prediction model based on the postoperative anesthesia risk assessment correction data and the intraoperative anesthesia risk assessment data to obtain the anesthesiology department risk prediction model.

2. The construction method of the risk prediction model based on the anesthesiology department according to claim 1, wherein, Step S1 is specifically: Step S11: Obtain the historical anesthesia data of the anesthesiology department, and extract anesthesia adverse characteristics from the historical anesthesia data of the anesthesiology department to obtain anesthesia adverse data; Step S12: Statistically analyze the individual anesthesia times of the historical anesthesia data in the anesthesiology department to obtain high-frequency anesthesia individual data and low-frequency anesthesia individual data; Step S13: Conduct a high-frequency individual anesthesia adaptability analysis on the high-frequency anesthesia individual data according to the anesthesia adverse data to obtain high-frequency individual anesthesia adaptability data; Step S14: Conduct a low-frequency individual anesthesia adaptability analysis on the low-frequency anesthesia individual data according to the anesthesia adverse data to obtain low-frequency individual anesthesia adaptability data; Step S15: Combine the high-frequency individual anesthesia adaptability data and the low-frequency individual anesthesia adaptability data to obtain individual anesthesia adaptability data.

3. The construction method of the risk prediction model based on the anesthesiology department according to claim 2, wherein, Step S13 specifically includes: Step S131: Calculate the anesthesia intervals for the high-frequency anesthesia individual data to obtain anesthesia interval data, and conduct a low-interval clustering calculation on the anesthesia interval data to obtain short-interval anesthesia data; Step S132: Obtain the anesthesia adverse reaction rules, and classify the anesthesia adverse data according to the anesthesia adverse reaction rules to obtain severe adverse data and minor adverse data; Step S133: Perform an intersection operation on the severe adverse data and the high-frequency anesthesia individual data to obtain high-frequency anesthesia low-adaptability individual data, and correct the misjudgment of the high-frequency anesthesia low-adaptability individual data according to the short-interval anesthesia data to obtain high-frequency anesthesia low-adaptability individual corrected data; Step S134: Perform an intersection operation on the minor adverse data and the high-frequency anesthesia individual data to obtain high-frequency anesthesia medium-adaptability individual data, and correct the misjudgment of the high-frequency anesthesia medium-adaptability individual data according to the short-interval anesthesia data to obtain high-frequency anesthesia medium-adaptability individual corrected data; Step S135: Eliminate the intersection data of the high-frequency anesthesia individual data according to the high-frequency anesthesia medium-adaptability individual corrected data and the high-frequency anesthesia low-adaptability individual corrected data to obtain high-frequency anesthesia high-adaptability individual data; Step S136: Combine the high-frequency anesthesia high-adaptability individual data, the high-frequency anesthesia low-adaptability individual corrected data, and the high-frequency anesthesia medium-adaptability individual corrected data to obtain high-frequency individual anesthesia adaptability data.

4. The construction method of the risk prediction model based on the anesthesiology department according to claim 3, characterized in that Step S14 specifically includes: Step S141: Perform an intersection operation on the severe adverse data and the low-frequency anesthesia individual data to obtain low-frequency anesthesia low-adaptability individual data; Step S142: Perform an intersection operation on the minor adverse data and the low-frequency anesthesia individual data to obtain low-frequency anesthesia medium-adaptability individual data; Step S143: Eliminate the intersection data of the low-frequency anesthesia individual data according to the low-frequency anesthesia medium-adaptability individual data and the low-frequency anesthesia low-adaptability individual data to obtain low-frequency anesthesia high-adaptability individual data; Step S144: Combine the low-frequency anesthesia high-adaptability individual data, the low-frequency anesthesia low-adaptability individual data, and the low-frequency anesthesia medium-adaptability individual data to obtain low-frequency individual anesthesia adaptability data.

5. The construction method of the risk prediction model based on the anesthesiology department according to claim 1, characterized in that Step S2 specifically includes: Step S21: Obtain the historical medical record data, and extract the anesthesia medical record data from the historical medical record data according to the historical anesthesia data in the anesthesiology department to obtain anesthesia medical record data; Step S22: Extract intraoperative monitoring data and anesthetic drug data from the anesthetic medical record data, so as to obtain intraoperative monitoring data and anesthetic drug data; Step S23: Perform spectral transformation on the intraoperative monitoring data, so as to obtain the intraoperative monitoring spectrum; Step S24: Perform intraoperative anesthetic risk assessment on the intraoperative monitoring spectrum and anesthetic drug data, so as to obtain intraoperative anesthetic risk assessment data.

6. The construction method of the risk prediction model based on the anesthesiology department according to claim 5, characterized in that Step S24 specifically includes: Step S241: Perform frequency fluctuation statistics on the intraoperative monitoring spectrum, so as to obtain frequency fluctuation data; Step S242: Perform time series correlation analysis on the frequency fluctuation data and anesthetic drug data, so as to obtain anesthetic fluctuation correlation data; Step S243: Calculate the fluctuation peak value according to the anesthetic fluctuation correlation data, so as to obtain the fluctuation peak value data, and perform threshold statistics on the fluctuation peak value data, so as to obtain the abnormal fluctuation peak value threshold; Step S244: Perform classification calculation on the anesthetic fluctuation correlation data according to the normal fluctuation peak value threshold, so as to obtain high-risk anesthetic risk data and low-risk anesthetic risk data; Step S245: Merge the high-risk anesthetic risk data and low-risk anesthetic risk data, so as to obtain intraoperative anesthetic risk assessment data.

7. The construction method of the risk prediction model based on the anesthesiology department according to claim 1, wherein Step S3 specifically includes: Step S31: Extract individual physiological characteristics from the anesthetic medical record data, so as to obtain individual preoperative physiological data and individual postoperative physiological data; Step S32: Calculate the physiological coefficient difference between the individual preoperative physiological data and the individual postoperative physiological data, so as to obtain high-difference physiological data and low-difference physiological data; Step S33: Perform statistical analysis on the individual postoperative physiological data, so as to obtain the normal physiological coefficient range, and perform classification calculation on the postoperative physiological data according to the normal physiological coefficient range, so as to obtain normal postoperative physiological data and abnormal postoperative physiological data; Step S34: Perform intersection operation according to the high-difference physiological data and the abnormal postoperative physiological data, so as to obtain the first postoperative anesthetic high-risk data; Perform intersection operation according to the low-difference physiological data and the normal postoperative physiological data, so as to obtain the first postoperative anesthetic low-risk data; Step S35: Perform intersection operation according to the low-difference physiological data and the abnormal postoperative physiological data, so as to obtain the second postoperative anesthetic high-risk data; Perform intersection operation according to the high-difference physiological data and the normal postoperative physiological data, so as to obtain the second postoperative anesthetic low-risk data; Step S36: Merge the first postoperative anesthetic high-risk data, the second postoperative anesthetic high-risk data, the first postoperative anesthetic low-risk data and the second postoperative anesthetic low-risk data, so as to obtain postoperative anesthetic risk assessment data.

Images