Anesthesia wake-up state prediction method and prediction system

Abstract

This application discloses a method and system for predicting anesthesia recovery status. First, a standard prediction model is constructed. Then, an actual set of physiological parameters is obtained. Based on this set, an actual recovery status parameter set is generated, and an initial recovery curve is generated within the standard prediction model. A minimum validation circle is generated, and the initial recovery curve is corrected using this circle to obtain a corrected recovery curve. Next, the initial predicted recovery time is calculated based on the corrected recovery curve and the standard prediction model. A composite deviation rate is generated based on the actual recovery status parameter set and the corrected recovery curve. Finally, the initial predicted recovery time is corrected a second time based on the composite deviation rate to obtain the predicted recovery time. This application achieves precise positioning of the recovery point and provides timely warnings by using the intersection of the corrected recovery curve and the standard prediction model. This reduces the workload of medical staff while ensuring the orderly conduct of medical work, thus improving nursing quality and the operational efficiency of the recovery ward.

Description

Technical Field

[0001] This application relates to the field of medical data processing technology, specifically to a method and system for predicting anesthesia recovery status. Background Technology

[0002] Anesthesia is widely used in surgery and special examinations such as colonoscopy. Currently, patients waking from anesthesia need to be transferred to a dedicated recovery ward to await natural awakening. During this time, medical staff in the recovery ward conduct regular or irregular rounds to observe each patient's awakening status and promptly transfer those who have reached the required level of awakening out of the ward, ensuring its normal operation. However, because patients' beds are randomly distributed within the ward, and their awakening status varies greatly depending on individual physical conditions, when multiple patients are admitted to the recovery ward simultaneously, medical staff need to continuously move around and patrol. This not only hinders the timely identification of patients who have reached the required level of awakening but also increases the workload of medical staff, negatively impacting the quality of medical care. Summary of the Invention

[0003] The main purpose of this application is to provide a method and system for predicting anesthesia recovery status, aiming to solve the problem of high monitoring pressure in the existing technology.

[0004] This application achieves the above objectives through the following technical solutions:

[0005] A method for predicting anesthesia recovery status includes the following steps:

[0006] Construct a standard prediction model;

[0007] A set of actual recovery state parameters is generated based on the actual physiological parameter set. An initial recovery curve is generated in the standard prediction model based on the set of actual recovery state parameters. A minimum validation circle is generated, and the initial recovery curve is corrected using the minimum validation circle to obtain a corrected recovery curve.

[0008] The initial predicted awakening time is calculated based on the modified recovery curve and the standard prediction model.

[0009] A composite deviation rate is generated based on the actual recovery state parameter set and the corrected recovery curve;

[0010] The initial predicted awakening time is corrected a second time based on the composite deviation rate to obtain the predicted awakening time.

[0011] Optionally, a standard prediction model can be constructed, including the following steps:

[0012] Obtain standard basic information and standard state index calculation model;

[0013] The standard state index is calculated based on the aforementioned standard basic information and the standard state index calculation model; the calculation expression of the standard state index calculation model is as follows: Where A represents the age parameter, G represents the gender parameter, M represents the anesthesia depth parameter, D represents the anesthesia dose parameter, N represents the anesthesia time parameter, and a0-a5 represent the coefficients of each item, which are constants.

[0014] A standard two-dimensional coordinate system is constructed, and a standard prediction model is generated by combining the standard state index.

[0015] Optionally, an actual recovery state parameter set is generated based on the actual physiological parameter set, and an initial recovery curve is generated in the standard prediction model based on the actual recovery state parameter set, including the following steps:

[0016] Obtain the actual physiological parameters of each physiological indicator and preprocess them to generate the actual physiological parameter set {P} 1,1 P 1,2 , ..., P 1,t P 2,1 P 2,2 , ..., P 2,t , ..., P i,1 , ..., P i,t}; where i represents the physiological indicator number and t represents the timestamp number;

[0017] The actual physiological parameter set is normalized to generate a normalized parameter set; the calculation expression for the normalization process is as follows: ,in This represents the maximum value of the i-th physiological indicator. This represents the minimum value of the i-th physiological indicator;

[0018] Based on the sampling time, the normalized parameter set is converted into several computational groups, wherein the expression for the computational group is N={P' 1,t , P' 2,t , ..., P' i,t};

[0019] Obtain the actual recovery status parameter calculation model, and generate the actual recovery status parameter set by combining the calculation groups described above;

[0020] Based on the actual physiological parameter set and sampling time, several calibration points are obtained in the standard prediction model;

[0021] Based on the calibration points, a nonlinear fitting is performed to generate an initial recovery curve.

[0022] Optionally, the expression for the actual recovery state parameter calculation model is as follows: The initial recovery curve is calculated using the following expression: , where L represents the maximum growth rate, and its calculation expression is L=S+q, where S represents the standard state exponent, q is a constant not less than 0; k, t0 and b are all fitting constants.

[0023] Optionally, a minimum validation circle is generated, and the initial recovery curve is corrected using the minimum validation circle to obtain a corrected recovery curve, including the following steps:

[0024] Obtain the sampling time and the minimum verification circle radius;

[0025] Based on the sampling time and the minimum verification circle radius, several minimum verification circles are generated on the initial recovery curve;

[0026] Mark each calibration point on the initial recovery curve;

[0027] Based on the sampling time, a one-to-one mapping relationship is established between each of the minimum verification circles and each of the calibration points;

[0028] According to the one-to-one mapping relationship, traverse each of the minimum verification circles and each of the calibration points. If the calibration point is located within the corresponding minimum verification circle, the calibration point is qualified and the number of qualified calibration points is counted.

[0029] The initial recovery curve is judged to be qualified based on the number of qualified calibration points and the judgment formula. If it is not qualified, the steps of obtaining a number of calibration points in the standard prediction model based on the actual physiological parameter set and sampling time are repeated until a corrected recovery curve is obtained.

[0030] Optionally, the initial predicted awakening time can be calculated using the following expression: , where L represents the maximum growth rate, and its calculation expression is L=S+q, where q is a constant not less than 0; k, t0 and b are all fitting constants.

[0031] Optionally, a composite deviation rate is generated based on the actual recovery state parameter set and the corrected recovery curve, including the following steps:

[0032] The single-point deviation rate of each actual recovery state parameter is calculated based on the actual recovery state parameter set and the corrected recovery curve.

[0033] Calculate the mean and standard deviation of the deviation rate for each point;

[0034] The composite deviation rate is calculated based on the mean and the standard deviation.

[0035] Optionally, the formula for calculating the single-point deviation rate is as follows: f(t) represents the actual recovery state parameter at time t, f'(t) represents the value of the corrected recovery curve at time t, and the formula for calculating the composite deviation rate is: ,in The mean of the form point deviation rate, The standard deviation represents the single-point deviation rate, and k represents the coefficient.

[0036] Optionally, the formula for calculating the predicted awakening time is: , where λ represents the correction coefficient and T is the initial predicted wake-up time.

[0037] Accordingly, this application also discloses a prediction system based on the above prediction method, including:

[0038] The model building module is used to build standard prediction models;

[0039] The first calculation module is used to generate an actual recovery state parameter set based on the actual physiological parameter set, and to generate an initial recovery curve in the standard prediction model by combining the actual recovery state parameter set.

[0040] The first correction module is used to generate a minimum verification circle, and to correct the initial recovery curve using the minimum verification circle to obtain a corrected recovery curve.

[0041] The second calculation module is used to calculate the initial predicted awakening time based on the corrected recovery curve and the standard prediction model.

[0042] The third calculation module is used to generate a composite deviation rate based on the actual recovery state parameter set and the corrected recovery curve.

[0043] The prediction module is used to perform a secondary correction on the initial predicted awakening time based on the composite deviation rate to obtain the predicted awakening time.

[0044] Compared with the prior art, this application has the following beneficial effects:

[0045] This application first constructs a standard prediction model, then generates an actual recovery state parameter set based on the actual physiological parameter set, and generates an initial recovery curve in the standard prediction model based on the actual recovery state parameter set; then generates a minimum validation circle, and corrects the initial recovery curve once using the minimum validation circle to obtain a corrected recovery curve; then calculates the initial predicted awakening time based on the corrected recovery curve and the standard prediction model; then generates a composite deviation rate based on the actual recovery state parameter set and the corrected recovery curve; finally, corrects the initial predicted awakening time a second time based on the composite deviation rate to obtain the predicted awakening time.

[0046] The standard prediction model defines the minimum criteria for full recovery or transfer out of the recovery ward. The modified recovery curve, generated based on the actual physiological parameter set, tracks the patient's recovery status in real time. The modified recovery curve will continuously approach the standard prediction model, and the intersection of the two is the recovery point. This enables precise positioning of the recovery time point, automatic monitoring of the patient's status, accurate prediction of the recovery threshold, and allows sufficient lead time for early warning of medical staff. This reduces the workload of medical staff, ensures the orderly conduct of medical and nursing work, improves the quality of nursing care, and allows for the timely transfer of eligible patients, effectively improving the operational efficiency of the recovery ward.

[0047] Secondly, during the calculation process, this application will also make two corrections, namely, correction by the minimum correction circle and correction by the composite deviation rate. The correction result of the minimum correction circle directly reflects whether the deviation of each point meets the requirements, that is, whether the distance between the fitted initial recovery curve and each point is in the optimal state, thereby ensuring that the initial recovery curve is the optimal solution, and thus ensuring the reliability of the basic parameters.

[0048] During the generation of the initial recovery curve, deviations from the actual locations are inevitable. A single correction can only minimize these deviations, but cannot eliminate them. The composite deviation rate reflects the overall offset of the initial recovery curve relative to each location. If the composite deviation rate is positive, it indicates that the predicted recovery time is earlier than the initial predicted recovery time; conversely, it indicates a certain lag. The composite deviation rate can supplement the unavoidable deviations into the final calculation results, thereby achieving a secondary correction and further revising the initial predicted recovery time, thus maximizing the accuracy of the prediction and correcting unavoidable systematic errors.

[0049] The two corrections mentioned above work together and synergistically to ensure the reliability of the trend, eliminate systematic biases as much as possible, and ensure the accuracy of the predicted awakening time.

[0050] Finally, this application enables real-time monitoring of the awakening state, that is, the actual physiological parameter set is updated in real time over time, further ensuring the matching degree with the real-time state, which is conducive to improving the accuracy of predicting the awakening time. Attached Figure Description

[0051] Figure 1 A flowchart of a method for predicting anesthesia recovery status provided in Embodiment 1 of this application;

[0052] Figure 2 The corrected schematic diagram for the minimum verification circle;

[0053] Figure 3This is a diagram illustrating the calculation principle of the composite deviation rate.

[0054] Figure 4 This is a schematic diagram of the structure of the prediction system provided in Embodiment 2 of this application;

[0055] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0057] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0058] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0059] Implementation Method 1

[0060] Reference Figure 1 This embodiment, as an optional embodiment of this application, discloses a method for predicting anesthesia recovery status.

[0061] S1. Construct a standard prediction model;

[0062] S11. Obtain standard basic information and standard state index calculation model;

[0063] Obtain the patient's standard basic information, which includes age, gender, depth of anesthesia, anesthetic dose, and anesthesia duration;

[0064] Then, the pre-stored lookup table and conversion formula are called to convert the standard basic information into the required parameters according to the proportion. For example, in the prediction lookup table, the value of male is 1 and female is 0, so the corresponding parameters are called according to gender.

[0065] S12. Calculate the standard state index based on the aforementioned standard basic information and standard state index calculation model;

[0066] The calculation expression for the standard state index calculation model is as follows: Where A represents age parameter, G represents gender parameter, M represents anesthesia depth parameter, D represents anesthesia dosage parameter, N represents anesthesia time parameter, and a0-a5 represent the coefficients of each item, which are constants;

[0067] The standard state index can be calculated by substituting the parameters obtained in step S11 into the track labeling state quality calculation model.

[0068] It should be noted that the standard state index is a specific awakening qualification line set according to the actual situation. The above calculation formula incorporates a large number of individual difference characteristics, such as gender, which can avoid the deviation of uniform indicators and thus ensure that the standard state index is closer to the actual situation of individuals.

[0069] S13. Construct a standard two-dimensional coordinate system and generate a standard prediction model by combining the standard state index.

[0070] First, a standard two-dimensional coordinate system is constructed, where the horizontal axis represents time and the vertical axis represents the state index.

[0071] By combining the standard state index with a horizontal line generated in the standard two-dimensional coordinate system, the standard prediction model can be obtained. See [link / reference] for details. Figure 2 ;

[0072] S2. Generate an actual recovery state parameter set based on the actual physiological parameter set, and generate an initial recovery curve in the standard prediction model using the actual recovery state parameter set; S21. Obtain the actual physiological parameters of each physiological indicator and preprocess them to generate the actual physiological parameter set {P} 1,1 P 1,2 , ..., P 1,t P 2,1 P 2,2 , ..., P 2,t , ..., P i,1 , ..., P i,t}; where i represents the physiological indicator number and t represents the timestamp number;

[0073] The detection parameters of various physiological indicators are obtained, and a unique code is generated for each parameter according to the physiological indicator number and sampling time.

[0074] Subsequently, the data were cleaned according to the 3σ principle to remove individual outlier parameters, which helps to improve the accuracy of the calculation.

[0075] For data missing after cleaning, adjacent parameters are extracted based on timestamp numbers, and then the difference is calculated using linear interpolation to fill in all missing data and obtain the actual physiological parameter set {P}. 1,1 P 1,2 , ..., P 1,t P 2,1 P 2,2 , ..., P 2,t , ..., P i,1 , ..., P i,t}; where i represents the physiological indicator number and t represents the timestamp number;

[0076] S22. Normalize the actual physiological parameter set to generate a normalized parameter set;

[0077] According to the normalization formula, each parameter in the actual physiological parameter set obtained in step S21 is normalized to obtain the normalized parameter set.

[0078] The calculation expression for the normalization process is as follows: ,in This represents the maximum value of the i-th physiological indicator. This represents the minimum value of the i-th physiological indicator;

[0079] It should be noted that the maximum and minimum values ​​of each physiological indicator can be preset or manually set.

[0080] S23. Based on the sampling time, the normalized parameter set is converted into several calculation groups, wherein the expression of the calculation group is N={P' 1,t , P' 2,t , ..., P' i,t};

[0081] Since all data from the same sampling time have the same timestamp number, normalized parameters with the same timestamp number can be grouped into the same set to obtain several calculation groups. Each calculation group corresponds to the parameters of all physiological indicators for one sampling time; the expression for the calculation array is N={P' 1,t , P'2,t , ..., P' i,t};

[0082] S24. Obtain the actual recovery status parameter calculation model, and generate the actual recovery status parameter set by combining each of the calculation groups;

[0083] The expression for the calculation model of the actual recovery state parameters is as follows: By substituting the parameters of each calculation group into the actual recovery state parameter calculation model, the actual recovery state parameters for each sampling time can be calculated, and then the actual recovery state parameter set can be obtained.

[0084] It should be noted that different weight values ​​can also be set for each physiological indicator, and then weighted calculations can be performed based on the weight values ​​to obtain the actual recovery status parameters;

[0085] S25. Obtain several calibration points in the standard prediction model based on the actual physiological parameter set and sampling time;

[0086] First, obtain the actual recovery state parameter set calculated in step S24, and then set the shift time window, which is a period of time, such as 10 minutes.

[0087] Since the data corresponding to the largest timestamp number is the latest data, we can start from the last data and reverse the process to obtain the parameters within the time period corresponding to the moving time window.

[0088] Then, based on the extracted parameters and their corresponding sampling times, several calibration points can be calibrated in the standard two-dimensional coordinate system.

[0089] Moving the time window can reduce the amount of data to be calculated, thereby improving computational efficiency.

[0090] Secondly, as the anesthesia wears off, the reference value of early data will continue to decrease. By moving the time window, early parameters can be cleared in a timely manner, thereby reducing their interference with the calculation results, ensuring that the calculation truly reflects the actual state, and improving the accuracy of the early warning.

[0091] Finally, when the later correction fails, the amount of data can be controlled by adjusting the size of the moving time window, and then the number of calibration points can be adjusted to achieve parameter adjustment over a larger range, ensuring the stable operation of the entire calculation method and avoiding the calculation process from falling into an infinite loop.

[0092] S26. Perform nonlinear fitting based on each of the calibration points to generate an initial recovery curve.

[0093] Based on each calibration point, a nonlinear fitting is performed using computer fitting to obtain an initial recovery curve. The calculation expression for the initial recovery curve is as follows: , where L represents the maximum growth rate, and its calculation expression is L=S+q, where S represents the standard state exponent, q is a constant not less than 0; k, t0 and b are all fitting constants.

[0094] Preferably, the value of q is 5-10. Since the calculation expression for the maximum growth rate is L=S+q, it can raise the upper limit based on the standard state exponent, ensuring that the two curves will definitely intersect, thereby achieving accurate prediction.

[0095] S3. Generate a minimum verification circle, and use the minimum verification circle to correct the initial recovery curve to obtain the corrected recovery curve;

[0096] S31. Obtain the sampling time and the minimum verification circle radius;

[0097] The minimum verification circle radius can be determined by a preset value or by manual assignment; then, the minimum verification circle is generated based on the minimum verification circle radius.

[0098] S32. Generate several minimum verification circles on the initial recovery curve according to the sampling time and the minimum verification circle radius;

[0099] Reference Figure 2 First, based on the sampling time, several vertical calibration lines are generated in the standard two-dimensional coordinate system. The intersection of each calibration line with the initial recovery curve is the center of the minimum verification circle. Then, several minimum verification circles can be generated.

[0100] S33. Mark each calibration point on the initial recovery curve;

[0101] S34. Establish a one-to-one mapping relationship between each of the minimum verification circles and each of the calibration points according to the sampling time;

[0102] Since each calibration point corresponds to a timestamp number, a one-to-one mapping relationship can be established between the minimum verification circle and each calibration point based on the sampling time; that is, each minimum verification circle corresponds to a unique calibration point, and the calibration point must be on the calibration line generated in step S32.

[0103] S35. According to the one-to-one mapping relationship, traverse each of the minimum verification circles and each of the calibration points. If the calibration point is located in the corresponding minimum verification circle, the calibration point is qualified and the number of qualified calibration points is counted.

[0104] Verify the positional relationship between each calibration point and its corresponding minimum verification circle. If the calibration point is located within the corresponding minimum verification circle, the calibration point is qualified; otherwise, the calibration point is deemed unqualified.

[0105] Each calibration point was compared one by one, and finally the number of calibration points that were deemed qualified was counted.

[0106] S36. Determine whether the initial recovery curve is qualified based on the number of qualified calibration points and the judgment formula. If it is not qualified, repeat the step of obtaining several calibration points in the standard prediction model based on the actual physiological parameter set and sampling time until the corrected recovery curve is obtained.

[0107] Obtain the judgment formula and determine whether the initial recovery curve is qualified based on the number of qualified calibration points. If the judgment formula is met, the initial recovery curve is qualified and is directly output as the corrected recovery curve.

[0108] If the condition is not met, the process is deemed unqualified. In this case, the process returns to step S25 and a new initial recovery curve is output again until the condition is met.

[0109] It should be noted that if the second-order fitting still does not meet the requirements, the data can be adjusted by changing the size of the shift time window;

[0110] The calculation expression of the judgment formula is B≥b, where B represents the number of qualified calibration points and b represents the minimum judgment threshold; that is, the initial recovery curve can only be considered qualified when the number of qualified calibration points is greater than the minimum judgment threshold.

[0111] By correcting the initial recovery curve using the minimum verification circle, the overall deviation rate of the initial recovery curve can be effectively judged, which directly reflects whether the deviation of each point meets the requirements, that is, whether the distance between the fitted initial recovery curve and each point is in the optimal state, thereby ensuring that the initial recovery curve is the optimal solution, and thus ensuring the reliability of the basic parameters.

[0112] S4. Calculate the initial predicted awakening time based on the modified recovery curve and the standard prediction model;

[0113] Since the modified recovery curve trends towards the standard prediction model, the intersection of the two is the initial predicted wake-up time. Therefore, the calculation expression for the initial predicted wake-up time is: , where L represents the maximum growth rate, and its calculation expression is L=S+q, where q is a constant not less than 0; k, t0 and b are all fitting constants;

[0114] By substituting the standard state index into the expression of the modified recovery curve during the calculation, the time can be calculated in reverse, thus obtaining the initial predicted awakening time.

[0115] S5. Generate a composite deviation rate based on the actual recovery state parameter set and the corrected recovery curve;

[0116] S51. Calculate the single-point deviation rate of each actual recovery state parameter based on the actual recovery state parameter set and the corrected recovery curve;

[0117] First, obtain the actual recovery status parameter set and the corrected recovery curve;

[0118] Then, each sampling time is substituted into the calculation formula of the corrected recovery curve to obtain the value of the corrected recovery curve at each sampling time.

[0119] Based on the same sampling time, a one-to-one mapping relationship will be established between the two sets of parameters mentioned above, as detailed in the following reference. Figure 3 Finally, the deviation rate of each single point is calculated according to the formula for calculating the single-point deviation rate.

[0120] The formula for calculating the single-point deviation rate is as follows: f(t) represents the actual recovery state parameter at time t, and f'(t) represents the value of the corrected recovery curve at time t.

[0121] The single-point deviation rate truly reflects the deviation of the actual recovery state parameters at each point from the corrected recovery curve.

[0122] S52. Calculate the mean and standard deviation of the deviation rate for each single point;

[0123] Calculate the mean and standard deviation of each single-point deviation rate obtained in step S51;

[0124] S53. Calculate the composite deviation rate based on the mean and the standard deviation.

[0125] Using the formula for calculating the composite deviation rate, the composite deviation rate is calculated based on the mean and standard deviation obtained in step S52; the expression for calculating the composite deviation rate is as follows: ,in The mean of the form point deviation rate, The standard deviation represents the single-point deviation rate, and k represents the coefficient.

[0126] S6. The initial predicted awakening time is corrected a second time based on the composite deviation rate to obtain the predicted awakening time.

[0127] The initial predicted awakening time is then corrected a second time by combining the composite deviation rate calculated in step S5 and the formula for calculating the predicted awakening time, thereby obtaining the final predicted awakening time; wherein the formula for calculating the predicted awakening time is as follows: , where λ represents the correction coefficient.

[0128] It should be noted that the correction factor ranges from 0.8 to 1.2. By introducing the correction factor, the influence of the composite deviation rate can be controlled, thereby further adjusting the correction accuracy.

[0129] As can be seen from the above calculation formula, when the composite deviation rate is negative, that is, the overall deviation is negative, the awakening time is delayed; conversely, the awakening time is advanced.

[0130] In the technical solution described in this application, the standard prediction model describes the minimum standard for full awakening or being able to be transferred out of the recovery ward. The modified recovery curve generated based on the actual physiological parameter set tracks the patient's recovery status in real time. That is, the modified recovery curve will continuously approach the standard prediction model, and the intersection of the two is the awakening point, thereby achieving precise positioning of the awakening point. This enables automatic monitoring of the patient's status and accurate prediction of the awakening threshold, while reserving sufficient lead time to provide early warning to medical staff. This reduces the workload of medical staff, ensures the orderly conduct of medical and nursing work, improves the quality of nursing care, and allows for the timely transfer of eligible patients, effectively improving the operational efficiency of the recovery ward.

[0131] Secondly, during the calculation process, this application will also make two corrections, namely, correction by the minimum correction circle and correction by the composite deviation rate. The correction result of the minimum correction circle directly reflects whether the deviation of each point meets the requirements, that is, whether the distance between the fitted initial recovery curve and each point is in the optimal state, thereby ensuring that the initial recovery curve is the optimal solution, and thus ensuring the reliability of the basic parameters.

[0132] The composite deviation rate reflects the overall deviation of the initial recovery curve relative to each point. If the composite deviation rate is positive, it indicates that the predicted recovery time is earlier than the initial predicted recovery time, and vice versa. This achieves secondary correction, thereby correcting the initial predicted recovery time and improving the accuracy of the prediction as much as possible. In other words, it corrects unavoidable systematic errors.

[0133] The two corrections mentioned above work together and synergistically to ensure the reliability of the trend, eliminate systematic biases as much as possible, and ensure the accuracy of the predicted awakening time.

[0134] Finally, this application enables real-time monitoring of the awakening state, that is, the actual physiological parameter set is updated in real time over time, further ensuring the matching degree with the real-time state, which is conducive to improving the accuracy of predicting the awakening time.

[0135] Implementation Method 2

[0136] Reference Figure 4This embodiment, as another optional embodiment of this application, discloses a prediction model, including a model building module and a first calculation module. The model building module is used to build a standard prediction model. The output of the model building module is electrically connected to a first calculation module for generating an initial recovery curve. The output of the first calculation module is electrically connected to a first correction module. The first correction module is used to generate a minimum validation circle and correct the initial recovery curve once using the minimum validation circle.

[0137] The output of the first correction module is electrically connected to the second calculation module and the third calculation module; the prediction system also includes a prediction module, and the outputs of the second calculation module and the third calculation module are respectively electrically connected to the prediction module;

[0138] The second calculation module calculates the initial predicted wake-up time based on the modified recovery curve and the standard prediction model. The third calculation module generates a composite deviation rate based on the actual recovery state parameter set and the modified recovery curve. The prediction module then performs a secondary correction on the initial predicted wake-up time based on the composite deviation rate to obtain the predicted wake-up time.

[0139] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. An anesthesia wake-up state prediction method characterized by comprising: Includes the following steps: Obtain standard basic information and standard state index calculation model; The standard state index is calculated based on the aforementioned standard basic information and the standard state index calculation model; the calculation expression of the standard state index calculation model is as follows: Where A represents age parameter, G represents gender parameter, M represents anesthesia depth parameter, D represents anesthesia dosage parameter, N represents anesthesia time parameter, and a0-a5 represent the coefficients of each item, which are constants; Construct a standard two-dimensional coordinate system and generate a standard prediction model by combining the standard state index; An actual recovery state parameter set is generated based on the actual physiological parameter set, and an initial recovery curve is generated in the standard prediction model based on the actual recovery state parameter set. A minimum verification circle is generated, and the initial recovery curve is corrected using the minimum verification circle to obtain the corrected recovery curve. The time corresponding to the intersection of the modified recovery curve and the standard prediction model is taken as the initial predicted awakening time; the calculation expression for the initial predicted awakening time is as follows: , where L represents the maximum growth rate, and its calculation expression is L=S+q, where q is a constant not less than 0; k, t0 and b are all fitting constants; A composite deviation rate is generated based on the actual recovery state parameter set and the corrected recovery curve; The initial predicted awakening time is corrected a second time based on the composite deviation rate to obtain the predicted awakening time.

2. The method of claim 1, wherein, The step of generating an actual recovery state parameter set based on the actual physiological parameter set, and generating an initial recovery curve in the standard prediction model based on the actual recovery state parameter set, includes the following steps: Obtain the actual physiological parameters of each physiological indicator and preprocess them to generate the actual physiological parameter set {P} 1,1 P 1,2 , ..., P 1,t P 2,1 P 2,2 , ..., P 2,t , ..., P i,1 , ..., P i,t }; where i represents the physiological indicator number and t represents the timestamp number; normalizing the actual physiological parameter set to generate a normalized parameter set; the calculation expression of the normalization processing is wherein maxi represents a maximum value of the i-th physiological index, mini represents a minimum value of the i-th physiological index; According to the sampling time, the set of normalization parameters is converted into a number of calculation groups, where the expression of the calculation groups is N = {P 1,t , P 2,t ,..., P i,t} Obtain the actual recovery status parameter calculation model, and generate the actual recovery status parameter set by combining the calculation groups described above; Based on the actual physiological parameter set and sampling time, several calibration points are obtained in the standard prediction model; Based on the calibration points, a nonlinear fitting is performed to generate an initial recovery curve.

3. The method of claim 2, wherein, The expression for the calculation model of the actual recovery state parameters is as follows: The initial recovery curve is calculated using the following expression: , where L represents the maximum growth rate, and its calculation expression is L=S+q, where S represents the standard state exponent, q is a constant not less than 0; k, t0 and b are all fitting constants.

4. The method of claim 1, wherein, The process of generating a minimum validation circle and then using that minimum validation circle to correct the initial recovery curve to obtain a corrected recovery curve includes the following steps: Obtain the sampling time and the minimum verification circle radius; Based on the sampling time and the minimum verification circle radius, several minimum verification circles are generated on the initial recovery curve; Mark each calibration point on the initial recovery curve; Based on the sampling time, a one-to-one mapping relationship is established between each of the minimum verification circles and each of the calibration points; According to the one-to-one mapping relationship, traverse each of the minimum verification circles and each of the calibration points. If the calibration point is located within the corresponding minimum verification circle, the calibration point is qualified and the number of qualified calibration points is counted. The initial recovery curve is judged to be qualified based on the number of qualified calibration points and the judgment formula. If it is not qualified, the steps of obtaining a number of calibration points in the standard prediction model based on the actual physiological parameter set and sampling time are repeated until a corrected recovery curve is obtained.

5. The method of claim 1, wherein, The step of generating the composite deviation rate based on the actual recovery state parameter set and the corrected recovery curve includes the following steps: The single-point deviation rate of each actual recovery state parameter is calculated based on the actual recovery state parameter set and the corrected recovery curve. Calculate the mean and standard deviation of the deviation rate for each point; The composite deviation rate is calculated based on the mean and the standard deviation.

6. The method of claim 5, wherein, The formula for calculating the single-point deviation rate is as follows: f(t) represents the actual recovery state parameter at time t, f'(t) represents the value of the corrected recovery curve at time t, and the formula for calculating the composite deviation rate is: ,in The mean of the form point deviation rate, The standard deviation represents the single-point deviation rate, and k represents the coefficient.

7. The method of claim 1, wherein, The calculation expression of the predicted wake-up time is where λ represents a correction coefficient.

8. The prediction system of the anesthetic emergence state prediction method according to any one of claims 1 to 7, characterized by, include: The model building module constructs standard prediction models; The first calculation module is used to generate an actual recovery state parameter set based on the actual physiological parameter set, and to generate an initial recovery curve in the standard prediction model by combining the actual recovery state parameter set. The first correction module is used to generate a minimum verification circle, and to correct the initial recovery curve using the minimum verification circle to obtain a corrected recovery curve. The second calculation module is used to calculate the initial predicted awakening time based on the corrected recovery curve and the standard prediction model. The third calculation module is used to generate a composite deviation rate based on the actual recovery state parameter set and the corrected recovery curve. A prediction module is configured to perform secondary correction on the initial predicted wake-up time according to the composite deviation rate to obtain a predicted wake-up time.

Images