Intelligent processing method for anesthesia patient care data based on operating room

Abstract

The present application relates to the technical field of medical data processing, in particular to an intelligent processing method for anesthesia patient nursing data based on operating rooms, which obtains a vital sign unstable period of vital sign time series data of each vital sign of a plurality of historical patients, obtains an influence correlation evaluation of any two vital signs according to the correlation of each vital sign of each historical patient and other vital signs with respect to the same vital sign unstable period, obtains a dangerous time of each vital sign of each historical patient according to the influence correlation evaluation, vital sign data and its change amount corresponding to a historical nursing intervention time of each vital sign of each historical patient, and a preset dangerous threshold of each vital sign, and finally obtains a target early warning threshold combined with the historical nursing intervention time of each vital sign of each historical patient, which greatly improves the accuracy and reliability of the target early warning threshold compared with the traditional artificially set early warning threshold.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical data processing, in particular to an intelligent processing method for anesthesia patient nursing data based on an operating room. BACKGROUND

[0002] Anesthesia patient nursing in an operating room is the whole process management and care for a surgical patient receiving anesthesia, aiming to protect the patient's life safety, maintain physiological stability, and reduce intraoperative and postoperative complications. Intraoperative nursing needs to detect the patient's vital signs in real time, such as heart rate, blood pressure, blood oxygen saturation, respiratory rate, and body temperature, etc., for real-time monitoring of the patient's life condition, as well as operating room environment information, such as temperature, humidity, etc.

[0003] During the operation, if the patient's vital signs are abnormal, such as allergic reaction, low blood oxygen saturation, low body temperature, etc., the nursing staff needs to intervene in time to ensure the patient's life safety and ensure the smooth progress of the anesthesia process. Anesthesia patient nursing in an operating room requires nursing staff to have high acuity to identify any vital sign abnormalities in time to ensure safety during the anesthesia process. Using data intelligent processing method to monitor various data of the patient can remind the nursing staff to nurse the patient when an abnormality occurs. The usual processing method is to set a warning threshold close to the range boundary based on the normal range of each vital sign, and to issue a warning to remind the nursing staff when the actual vital sign data exceeds the warning threshold. However, the existing setting method of the warning threshold is artificial setting, which cannot be accurately determined. If the warning threshold is set too conservatively, it may trigger a false alarm, and if the warning threshold is set too close to the range boundary, it may cause the warning to be not timely enough, delaying the nursing intervention of the nursing staff, causing abnormality or sequelae during the patient's recovery period, and threatening the patient's health. SUMMARY

[0004] In order to solve the technical problem of inaccurate setting of the existing warning threshold, the purpose of the present application is to provide an intelligent processing method for anesthesia patient nursing data based on an operating room, and the technical solution adopted is as follows:

[0005] In the first aspect of the present application, an intelligent processing method for anesthesia patient nursing data based on an operating room is provided, comprising:

[0006] Obtaining vital sign time series data of each vital sign of a plurality of historical patients, and obtaining a vital sign unstable period of the vital sign time series data;

[0007] According to the correlation of each vital sign of each historical patient with other vital signs for the same vital sign unstable period, an influence correlation evaluation between any two vital signs is obtained;

[0008] According to the influence correlation evaluation, the historical nursing intervention time corresponding to the vital sign data and the vital sign data change amount of each vital sign of each historical patient, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained;

[0009] Based on the dangerous time of each vital sign of each historical patient, and the historical nursing intervention time of each vital sign of each historical patient, a target early warning threshold is obtained.

[0010] In an exemplary embodiment, the vital sign unstable period of the vital sign time series data comprises:

[0011] A stable state reference range centered on preset vital sign stable state data of a first vital sign is obtained, the first vital sign being any vital sign;

[0012] Abnormal vital sign data not in the stable state reference range is determined from the vital sign time series data of the first vital sign of the first historical patient, and a period formed by the abnormal vital sign data of the first vital sign of the first historical patient is taken as a vital sign unstable period of the first vital sign of the first historical patient; the first historical patient is any historical patient.

[0013] In an exemplary embodiment, the influence correlation evaluation between any two vital signs is obtained according to the correlation of each vital sign of each historical patient to the same vital sign unstable period of other vital signs, comprising:

[0014] The time series data of the first vital sign unstable period in the vital sign time series data of the first vital sign of the first historical patient is obtained to obtain first vital sign unstable time series data, and the time series data of the first vital sign unstable period in the vital sign time series data of the second vital sign of the first historical patient is obtained to obtain second vital sign unstable time series data; the first historical patient is any historical patient, the first vital sign and the second vital sign are any two different vital signs, and the first vital sign unstable period is the vital sign unstable period of the first vital sign of the first historical patient;

[0015] The correlation of the first vital sign unstable time series data and the second vital sign unstable time series data is obtained as the correlation of the first vital sign to the second vital sign;

[0016] The correlation of the first vital sign to the second vital sign of all historical patients is fused to obtain the influence correlation evaluation of the first vital sign to the second vital sign.

[0017] In an exemplary embodiment, the fusion of the correlation of the first vital sign to the second vital sign of all historical patients obtains an influence correlation evaluation of the first vital sign to the second vital sign, including:

[0018] Calculating the correlation of the first vital sign to the second vital sign of all historical patients to obtain a correlation mean value;

[0019] Obtaining the correlation variance of the first vital sign to the second vital sign of all historical patients;

[0020] According to the correlation mean value and the correlation variance, the influence correlation evaluation is obtained, which is proportional to the correlation mean value and inversely proportional to the correlation variance.

[0021] In an exemplary embodiment, according to the influence correlation evaluation, the historical nursing intervention time corresponding to the vital sign data and the vital sign data change amount of each historical patient, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained, including:

[0022] According to the influence correlation evaluation and the vital sign data change amount of each historical nursing intervention time of each vital sign of each historical patient, a vital sign change coefficient of each vital sign of each historical patient is obtained;

[0023] According to the vital sign change coefficient of each vital sign of each historical patient, the vital sign data corresponding to the historical nursing intervention time of each vital sign of each historical patient, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained.

[0024] In an exemplary embodiment, according to the influence correlation evaluation and the vital sign data change amount of each historical nursing intervention time of each vital sign of each historical patient, a vital sign change coefficient of each vital sign of each historical patient is obtained, including:

[0025] According to the influence correlation evaluation of the first vital sign to the first historical patient's reference vital sign, and the vital sign data change amount of the historical nursing intervention time of the first historical patient's reference vital sign, the vital sign data change feature of the first historical patient's reference vital sign is obtained; the first historical patient is any one historical patient, the first vital sign is any one vital sign, and the reference vital sign is other vital sign except the first vital sign;

[0026] The vital sign data change characteristics of each reference vital sign of the first historical patient are fused, and the vital sign data change amount corresponding to the historical nursing intervention moment of the first vital sign of the first historical patient is combined to obtain a vital sign change coefficient of the first vital sign of the first historical patient.

[0027] In an exemplary embodiment, the calculation formula of the vital sign change coefficient is as follows:

[0028] ;

[0029] Wherein, represents a vital sign change coefficient of the i-th vital sign of the k-th historical patient, represents a vital sign data change amount corresponding to the historical nursing intervention moment of the i-th vital sign of the k-th historical patient, represents an influence correlation evaluation of the j-th vital sign of the k-th historical patient on the i-th vital sign, represents a vital sign data change amount corresponding to the historical nursing intervention moment of the j-th vital sign of the k-th historical patient, and n represents the number of reference vital signs of the k-th historical patient.

[0030] In an exemplary embodiment, the vital sign change coefficient of each vital sign of each historical patient, the vital sign data corresponding to the historical nursing intervention moment of each vital sign of each historical patient, and the preset dangerous threshold of each vital sign are used to obtain the dangerous time of each vital sign of each historical patient, including:

[0031] The preset dangerous threshold of the first vital sign is calculated with the vital sign data difference of the vital sign data corresponding to the historical nursing intervention moment of the first vital sign of the first historical patient;

[0032] The ratio of the vital sign data difference and the vital sign change coefficient of the first vital sign of the first historical patient is calculated as the dangerous time of the first vital sign of the first historical patient.

[0033] In an exemplary embodiment, the target early warning threshold is obtained based on the dangerous time of each vital sign of each historical patient and the historical nursing intervention moment of each vital sign of each historical patient, including:

[0034] The time difference between the intervention effective moment of the first vital sign of the first historical patient and the historical nursing intervention moment is obtained, and the intervention effective moment is the moment of the first extreme point after the historical nursing intervention moment;

[0035] obtaining a lag characteristic of the first vital sign of the first historical patient according to the time difference and a size relationship between the first vital sign of the first historical patient and a dangerous time;

[0036] fusing the lag characteristics of the first vital sign of all the historical patients to obtain a comprehensive lag of the first vital sign;

[0037] obtaining a target early warning threshold of the first vital sign according to the comprehensive lag of the first vital sign, a preset dangerous threshold of the first vital sign, and historical nursing intervention time of the first vital sign of each historical patient.

[0038] In an exemplary embodiment, the obtaining of the target early warning threshold of the first vital sign according to the comprehensive lag of the first vital sign, the preset dangerous threshold of the first vital sign, and the vital sign data corresponding to the historical nursing intervention time of the first vital sign of each historical patient comprises:

[0039] obtaining a vital sign data average value of the vital sign data corresponding to the historical nursing intervention time of the first vital sign of all the historical patients;

[0040] when the comprehensive lag of the first vital sign is greater than or equal to a comprehensive lag threshold of the first vital sign, the target early warning threshold is calculated by using the following calculation formula:

[0041] ;

[0042] wherein, the target early warning threshold of the i-th vital sign, the preset dangerous threshold of the i-th vital sign, the comprehensive lag of the i-th vital sign, the preset comprehensive lag threshold of the i-th vital sign, the vital sign data average value of the vital sign data corresponding to the historical nursing intervention time of the i-th vital sign of all the historical patients.

[0043] The intelligent processing method of anesthesia patient nursing data based on an operating room provided by the present application has the following beneficial effects: the intelligent processing method of anesthesia patient nursing data based on an operating room provided by the present application performs data comprehensive analysis on multiple vital signs of multiple historical patients, and obtains early warning thresholds closely related to each vital sign according to data correlation between vital sign data of each vital sign of each historical patient, so that the accuracy and reliability are greatly improved compared with the traditional artificial early warning threshold. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 is a flowchart of an intelligent processing method of anesthesia patient nursing data based on an operating room provided by an embodiment of the present application;

[0045] Figure 2 is a flowchart of acquiring a vital sign unstable period according to an embodiment of the present application;

[0046] Figure 3 is a flowchart of acquiring an impact-related evaluation according to an embodiment of the present application;

[0047] Figure 4 is a flowchart of a specific calculation of an impact-related evaluation according to an embodiment of the present application;

[0048] Figure 5 is a flowchart of acquiring a dangerous time according to an embodiment of the present application;

[0049] Figure 6 is a flowchart of acquiring a vital sign change coefficient according to an embodiment of the present application;

[0050] Figure 7 is a flowchart of a specific calculation of a dangerous time according to an embodiment of the present application;

[0051] Figure 8 is a flowchart of acquiring a target early warning threshold according to an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to further clarify the technical means and effects of the present application taken to achieve the predetermined object of the application, the specific embodiments, structures, features and effects of the present application are described in detail below in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The data information collected in this application is obtained with full authorization, and the collection, use and processing of relevant information need to comply with relevant regulations.

[0054] As shown in Figure 1 The present embodiment provides an intelligent processing method for anesthesia patient care data based on an operating room, which comprises the following steps:

[0055] Step 1: acquiring vital sign time series data of various vital signs of a plurality of historical patients, and acquiring a vital sign unstable period of the vital sign time series data;

[0056] Step 2: Obtain the influence correlation evaluation between any two vital signs according to the correlation between each vital sign of each historical patient and other vital signs for the same vital sign unstable period;

[0057] Step 3: Obtain the risk time of each vital sign of each historical patient according to the influence correlation evaluation, the vital sign data corresponding to the historical nursing intervention time of each vital sign of each historical patient and the vital sign data change, and the preset risk threshold of each vital sign.

[0058] Step 4: Obtain the target early warning threshold based on the risk time of each vital sign of each historical patient and the historical nursing intervention time of each vital sign of each historical patient.

[0059] The steps of the intelligent processing method for anesthesia patient nursing data in the operating room provided by the embodiment will be described in detail below with reference to the accompanying drawings.

[0060] Step 1: Obtain the vital sign time series data of each vital sign of a plurality of historical patients, and obtain the vital sign unstable period of the vital sign time series data.

[0061] The vital sign time series data of each vital sign of a plurality of historical patients is obtained. It should be understood that each historical patient is a plurality of patients who have completed the same operation in the operating room, and the anesthesia method of the patients is also the same. The number of historical patients is set by actual conditions. The number of types and specific items of vital signs of historical patients are set by actual needs, such as heart rate, blood pressure, blood oxygen saturation, respiratory rate, and body temperature. Each vital sign is collected at a preset collection frequency, and the collection frequency is set by actual conditions, such as 1 time per second. According to the set collection frequency, discrete vital sign data of each vital sign of each historical patient is obtained, and then the discrete vital sign data is arranged in time series to obtain vital sign time series data of each vital sign of each historical patient. For example, the heart rate time series data is composed of the heart rate values collected at each time.

[0062] Since each historical patient is a patient who has undergone the same operation, and the anesthesia method of the patient is also the same, the vital sign time series data of each vital sign of each historical patient is obtained from the beginning of anesthesia, and data is continuously obtained for a certain period of time. The length of the specific time period is set by actual needs, such as ending the acquisition at the end of the operation. The length of the acquisition time period of each vital sign of each historical patient is the same. Further, after obtaining the vital sign time series data of each vital sign, the starting time of each vital sign time series data can be unified, so that each time of each vital sign time series data corresponds.

[0063] The nursing intervention performed by the nursing staff refers to the manual nursing intervention performed by the nursing staff when the anesthetized patient suddenly has an abnormal risk of vital signs during the operation, for example, the anesthetized patient has respiratory depression due to anesthetic drugs, and low blood oxygen may be caused due to low blood oxygen saturation. The nursing staff needs to ensure that the blood oxygen of the patient returns to a normal level through a mask, a catheter oxygen supply or other means.

[0064] The nursing intervention time of each vital sign of each historical patient is the time point when the nursing staff performs the nursing intervention, which is defined as the historical nursing intervention time. It should be understood that the historical nursing intervention time of each vital sign of each historical patient is the time when the nursing staff performs the nursing intervention according to the specific situation of each vital sign of each historical patient, and therefore the historical nursing intervention time of each vital sign of each historical patient is independent of each other. The historical nursing intervention time may be different or the same for different historical patients and different vital signs. Moreover, the historical nursing intervention time corresponding to the vital sign data of each vital sign of each historical patient is also obtained.

[0065] In an exemplary embodiment, the vital sign time series data of each vital sign of each historical patient can also be recorded by the nursing staff during nursing and stored in the server, which can be directly called from the server in subsequent data processing.

[0066] Then, the vital sign unstable period of the vital sign time series data is obtained. In an exemplary embodiment, as shown in Figure 2 , a specific acquisition process of the vital sign unstable period is given:

[0067] Step 1-1: Obtain a stable state reference range centered on the preset vital sign stable state data of the first vital sign.

[0068] For ease of illustration, the first historical patient is set to be any one of the historical patients, and the first vital sign is set to be any one of the vital signs.

[0069] Since not all vital sign data during the operation is in an abnormal change state, but the abnormal change occurs from a stable state, and then the nursing staff performs the nursing intervention and restores to the normal stable state, but the comparison of the data during the stable period has a low reference degree for the correlation analysis, it is necessary to obtain the vital sign unstable period of the vital sign time series data, that is, the time period in which the vital sign data is abnormally changed, so as to obtain the influence correlation evaluation between any two vital signs according to the vital sign unstable period.

[0070] For the first vital sign, preset the vital sign stable state data of the first vital sign, that is, the vital sign stable state data corresponding to the patient when the first vital sign is not abnormal. Taking heart rate as an example, the preset vital sign stable state data of heart rate can be set to 80 times per minute.

[0071] It should be understood that the preset vital sign stable state data of each vital sign can reflect the stable state of the vital sign data, which can be directly obtained by relevant research, or can be obtained in the following manner: detecting the vital sign data of the first vital sign of a large number of healthy people in history, thereby obtaining the vital sign stable state data of the first vital sign, and then calculating the average value of the vital sign stable state data of the first vital sign of all historical healthy people as the preset vital sign stable state data of the first vital sign. The preset vital sign stable state data of other vital signs is also obtained in this way.

[0072] In order to ensure the accuracy of data analysis, a stable state reference range is determined with the preset vital sign stable state data of the first vital sign as the center. If the vital sign data of the first vital sign is within the stable state reference range, it is determined that the vital sign data is normal. The size of the stable state reference range is set by the specific vital sign. Taking heart rate as an example, the stable state reference range of heart rate can be set to 60-100 times per minute.

[0073] By using the above process, the preset vital sign stable state data of each vital sign and the stable state reference range centered on the preset vital sign stable state data are obtained.

[0074] Step 1-2: Determine the abnormal vital sign data that is not within the stable state reference range from the vital sign time series data of the first vital sign of the first historical patient, and regard the time period formed by the abnormal vital sign data of the first historical patient as the vital sign unstable time period of the first vital sign of the first historical patient.

[0075] When the vital sign data deviates from the corresponding stable state reference range, it indicates that the vital sign data is abnormal. Therefore, each vital sign data in the vital sign time series data of the first vital sign of the first historical patient is compared with the stable state reference range of the first vital sign to determine whether each vital sign data is within the stable state reference range, thereby obtaining the vital sign data that is not within the stable state reference range in the vital sign time series data of the first vital sign of the first historical patient. These vital sign data are abnormal vital sign data.

[0076] Each abnormal vital sign data corresponds to a time point, so the time points of the abnormal vital sign data in the vital sign time series data of the first vital sign of the first historical patient are obtained, and a time period formed according to the time points is taken as a vital sign unstable time period of the first vital sign of the first historical patient.

[0077] Since the nursing intervention is a manual intervention after the occurrence of an abnormality, in general, the vital sign unstable time period includes a stage in which the patient's vital sign data is abnormal before the nursing intervention, a stage in which the patient's vital sign returns to a normal level after the nursing intervention, and the time point of the nursing intervention itself.

[0078] In the above manner, the vital sign unstable time period of each vital sign of each historical patient is obtained.

[0079] It should be understood that for any historical patient, among the vital signs of the historical patient, there can be a vital sign that does not have a vital sign unstable time period, and then when subsequent influence-related evaluations of the historical patient are obtained based on the vital sign unstable time period, only the vital signs that have a vital sign unstable time period are considered, and only the influence-related evaluations of the vital signs that have a vital sign unstable time period on other vital signs of the historical patient are obtained. Since the nursing intervention is performed by the nursing staff when the vital sign is abnormal, that is, when the vital sign is unstable, the vital sign unstable time period can have the same meaning as the nursing intervention, and the vital sign that does not have a vital sign unstable time period represents a vital sign that does not have a nursing intervention.

[0080] Step 2: Obtain the influence-related evaluation between any two vital signs according to the correlation of each vital sign of each historical patient with other vital signs with respect to the same vital sign unstable time period.

[0081] After obtaining the vital sign unstable time period of each vital sign of each historical patient, the influence-related evaluation between any two vital signs is obtained according to the correlation of each vital sign of each historical patient with other vital signs with respect to the same vital sign unstable time period.

[0082] In an exemplary embodiment, as shown in Figure 3 A specific process for obtaining the influence-related evaluation is as follows:

[0083] Step 2-1: Obtain the time series data of the first vital sign unstable time period in the vital sign time series data of the first vital sign of the first historical patient to obtain the first vital sign unstable time series data, and obtain the time series data of the first vital sign unstable time period in the vital sign time series data of the second vital sign of the first historical patient to obtain the second vital sign unstable time series data.

[0084] For the convenience of description, it is assumed that the second vital sign is also any one of the vital signs, and the first vital sign and the second vital sign are two different vital signs, and the first vital sign unstable period is the vital sign unstable period of the first vital sign of the first historical patient.

[0085] Since the first vital sign unstable period is a time period, the time sequence data of the first vital sign unstable period in the vital sign time sequence data of the first vital sign of the first historical patient is obtained, which is defined as the first vital sign unstable time sequence data. And the time sequence data of the first vital sign unstable period in the vital sign time sequence data of the second vital sign of the first historical patient is obtained, which is defined as the second vital sign unstable time sequence data.

[0086] Therefore, the first vital sign unstable time sequence data and the second vital sign unstable time sequence data are the vital sign data of the first vital sign unstable period corresponding to the first vital sign and the second vital sign, respectively. In this way, the vital sign data of the first vital sign unstable period corresponding to the first vital sign is obtained for each vital sign.

[0087] Step 2-2: Obtain the correlation of the first vital sign unstable time sequence data and the second vital sign unstable time sequence data as the correlation of the first vital sign to the second vital sign.

[0088] The correlation of the first vital sign unstable time sequence data and the second vital sign unstable time sequence data is obtained as the correlation of the first vital sign to the second vital sign. In an exemplary embodiment, the Pearson correlation coefficient of the first vital sign unstable time sequence data and the second vital sign unstable time sequence data is calculated and normalized as the correlation of the first vital sign to the second vital sign.

[0089] Taking the correlation of the i-th vital sign to the j-th vital sign of the k-th historical patient as an example, the following calculation formula is used to calculate:

[0090] ;

[0091] Wherein, represents the correlation of the i-th vital sign to the j-th vital sign of the k-th historical patient, Pearson represents the Pearson correlation coefficient, represents the vital sign unstable time sequence data of the vital sign time sequence data of the i-th vital sign of the k-th historical patient in the vital sign unstable period of the i-th vital sign, The vital sign time series data representing the jth vital sign of the kth historical patient in the vital sign unstable time period of the ith vital sign is vital sign unstable time series data, and norm represents a normalization function. Since the numerical range of the Pearson correlation coefficient is -1 to 1, the normalization of the Pearson correlation coefficient here can be: add the Pearson correlation coefficient to the value 1, and then divide by the value 2, thereby completing the normalization.

[0092] It should be understood that according to the above correlation calculation method, for the ith vital sign and the jth vital sign of the kth historical patient, the correlation of the ith vital sign of the kth historical patient to the jth vital sign is obtained based on the vital sign unstable time period of the ith vital sign, and the correlation of the jth vital sign of the kth historical patient to the ith vital sign is obtained based on the vital sign unstable time period of the jth vital sign, although both are related to the ith vital sign and the jth vital sign, but they are different parameters, and cannot be confused.

[0093] It should be understood that if the ith vital sign of the kth historical patient includes multiple vital sign unstable time periods, the correlation of the ith vital sign of the kth historical patient to the jth vital sign is obtained for each vital sign unstable time period in the above manner, and then the average value is calculated to obtain the correlation of the ith vital sign of the kth historical patient to the jth vital sign.

[0094] Step 2-3: Fusion of the correlation of the first vital sign to the second vital sign of all historical patients to obtain the influence correlation evaluation of the first vital sign to the second vital sign.

[0095] According to the correlation of the first vital sign to the second vital sign of all historical patients, the influence correlation evaluation of the first vital sign to the second vital sign is obtained. Taking the ith vital sign to the jth vital sign as an example, if all historical patients, in the vital sign unstable time period of the ith vital sign, the jth vital sign all appears the same change trend, then the ith vital sign to the jth vital sign indeed shows strong correlation.

[0096] In an exemplary embodiment, as shown in Figure 4 The calculation process of the influence correlation evaluation includes:

[0097] Step 2-3-1: Calculate the average value of the correlation of the first vital sign to the second vital sign of all historical patients;

[0098] Step 2-3-2: Obtain the variance of the correlation of the first vital sign to the second vital sign of all historical patients;

[0099] Step 2-3-3: According to the correlation average and the correlation variance, an influence correlation evaluation is obtained, which is proportional to the correlation average and inversely proportional to the correlation variance.

[0100] Taking the i-th vital sign and the j-th vital sign as an example, the average of the correlation between the i-th vital sign and the j-th vital sign of all historical patients is calculated. The correlation average reflects the overall level of the correlation between the i-th vital sign and the j-th vital sign of all historical patients. The higher the overall level is, the higher the influence of the i-th vital sign on the j-th vital sign is, and the higher the influence correlation evaluation is.

[0101] The variance of the correlation between the i-th vital sign and the j-th vital sign of all historical patients is calculated. The variance represents the fluctuation of the correlation between the i-th vital sign and the j-th vital sign of all historical patients. The greater the variance is, the greater the fluctuation is, and the lower the influence of the i-th vital sign on the j-th vital sign is, and the lower the influence correlation evaluation is.

[0102] It should be understood that when calculating the correlation average and the correlation variance, the historical patients participating are the historical patients with vital sign unstable periods of the i-th vital sign.

[0103] In an exemplary embodiment, a specific calculation formula of the influence correlation evaluation of the i-th vital sign on the j-th vital sign is given as follows:

[0104] ;

[0105] wherein, represents the influence correlation evaluation of the i-th vital sign on the j-th vital sign, represents the number of historical patients with vital sign unstable periods of the i-th vital sign among all historical patients, represents the variance of the correlation between the i-th vital sign and the j-th vital sign of the historical patients, and the calculation object of the correlation variance is also the historical patients with vital sign unstable periods of the i-th vital sign among all historical patients. represents the negative correlation normalization of The negative correlation normalization method here can be: , represents the processing object, and exp represents the exponential function with the natural constant e as the base.

[0106] In the above manner, the influence correlation evaluations of each vital sign on other vital signs are obtained.

[0107] Step 3: According to the influence correlation evaluation, the historical nursing intervention time corresponding to the vital sign data and the vital sign data change amount of each historical patient, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained.

[0108] Generally, when the historical patient has an abnormal vital sign, the process of the nursing staff identifying, monitoring and responding to the abnormal situation is related to the professional knowledge, response ability and sensitivity of the nursing staff. Therefore, in actual operation, the nursing intervention time for the historical patient will be different due to the subjective ability of the nursing staff, and it is not necessarily the most ideal intervention time. When the nursing staff fails to identify the abnormal risk in time or respond in time, the nursing intervention time may be delayed. Whether the nursing intervention time is timely is analyzed through the vital sign change trend of the historical patient.

[0109] According to the influence correlation evaluation, the historical nursing intervention time corresponding to the vital sign data and the vital sign data change amount of each historical patient, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained.

[0110] In an exemplary embodiment, as shown in Figure 5 A specific acquisition process of the dangerous time is as follows:

[0111] Step 3-1: According to the influence correlation evaluation and the vital sign data change amount of each historical patient at the historical nursing intervention time corresponding to each vital sign, the vital sign change coefficient of each historical patient is obtained.

[0112] The change rate of different vital signs in the abnormal change stage may be different, which in turn will lead to different response windows for the nursing staff to identify the abnormality and take corresponding nursing measures when the abnormality occurs. For example, the change process of body temperature is relatively slow, and the response window for the nursing staff to find the abnormal trend and nursing intervention is longer, while the process of heart rate acceleration changes relatively faster, so the nursing staff needs to identify the abnormality and nursing intervention faster, otherwise it may threaten the life and health of the patient due to no timely intervention. Therefore, the vital sign change coefficient, i.e. the change rate, of different vital signs is analyzed.

[0113] For the same vital sign, the abnormality of different historical patients also has difference, taking the ith vital sign as an example, according to the change amount sequence of each historical patient, the time when the ith vital sign reaches the preset danger threshold without nursing intervention is obtained. The preset danger threshold is a preset initial danger threshold, which can also be understood as an initial warning threshold, and finally the target warning threshold is obtained according to the preset danger threshold. In this embodiment, the preset danger threshold refers to the critical value of the vital sign data exceeding the normal range, and the normal range here refers to the normal range reference value in medicine, for example, the normal range of human heart rate is 60-100 times per minute.

[0114] In analyzing the vital sign change coefficient of each vital sign, in addition to considering the change trend of the ith vital sign itself, the influence of other vital signs on the ith vital sign is also considered, so as to determine a more reasonable warning threshold. Therefore, according to the influence correlation evaluation and the vital sign data change amount corresponding to the historical nursing intervention moment of each vital sign of each historical patient, the vital sign change coefficient of each vital sign of each historical patient is obtained.

[0115] In an exemplary embodiment, as shown in Figure 6 A specific acquisition process of the vital sign change coefficient is as follows:

[0116] Step 3-1-1: According to the influence correlation evaluation of each reference vital sign of the first historical patient on the first vital sign and the vital sign data change amount corresponding to the historical nursing intervention moment of each reference vital sign of the first historical patient, the vital sign data change characteristics of each reference vital sign of the first historical patient are obtained.

[0117] For the convenience of description, it is assumed that each vital sign other than the first vital sign is the reference vital sign of the first vital sign.

[0118] The difference sequence of the vital sign time series data of each vital sign, i.e. the change amount sequence, is obtained, and each vital sign data change amount in the change amount sequence is the absolute value of the difference between the latter vital sign data and the former vital sign data in the adjacent two vital sign data of the vital sign time series data.

[0119] For example, it is assumed that the vital sign time series data of the ith vital sign of the kth historical patient is It is assumed that:

[0120] ;

[0121] Wherein, indicates the vital sign data at the 1st time in the vital sign time series data of the ith vital sign of the kth historical patient, the vital sign data of the 2nd time point in the vital sign time series data of the i-th vital sign of the k-th historical patient, the vital sign data of the x-th time point in the vital sign time series data of the i-th vital sign of the k-th historical patient, the vital sign data of the T-th time point in the vital sign time series data of the i-th vital sign of the k-th historical patient, the total number of time points in the vital sign time series data.

[0122] the change amount sequence of the vital sign time series data of the i-th vital sign of the k-th historical patient is set as:

[0123]

[0124] wherein,

[0125] In the above manner, the change amount sequence of each reference vital sign of the i-th vital sign of the k-th historical patient is obtained. Then, the vital sign data change amount corresponding to the historical nursing intervention time point of the vital sign is found from the change amount sequence. Taking the i-th vital sign of the k-th historical patient as an example, if the historical nursing intervention time point of the i-th vital sign of the k-th historical patient is assumed to be the x-th time point, then the vital sign data corresponding to the historical nursing intervention time point is , and the vital sign data change amount corresponding to the historical nursing intervention time point is .

[0126] Then, according to the influence-related evaluation of each reference vital sign on the i-th vital sign of the k-th historical patient, and the vital sign data change amount corresponding to the historical nursing intervention time point of each reference vital sign of the i-th vital sign of the k-th historical patient, the vital sign data change characteristics of each reference vital sign of the i-th vital sign of the k-th historical patient are obtained. In an exemplary embodiment, for any one reference vital sign, the product of the influence-related evaluation of the reference vital sign on the i-th vital sign of the k-th historical patient and the vital sign data change amount corresponding to the historical nursing intervention time point of the reference vital sign of the i-th vital sign of the k-th historical patient is taken as the vital sign data change characteristic of the reference vital sign of the i-th vital sign of the k-th historical patient.

[0127] ​​​​​Step 3-1-2: Integrate the vital sign data change characteristics of the first historical patient's various reference vital signs, and combine them with the changes in vital sign data corresponding to the historical nursing intervention time of the first historical patient's first vital sign to obtain the vital sign change coefficient of the first historical patient's first vital sign.

[0128] The vital sign data change characteristics of the first historical patient are integrated, for example, by calculating the average value. Then, combined with the change in vital sign data corresponding to the historical nursing intervention time of the first historical patient, the vital sign change coefficient of the first historical patient is obtained.

[0129] Let the j-th vital sign be used as an example for any reference vital sign of the i-th vital sign. In an exemplary embodiment, the formula for calculating the coefficient of change of vital signs is given below:

[0130] ;

[0131] in, The coefficient of change of the i-th vital sign represents the vital sign change of the k-th historical patient. This represents the change in vital sign data corresponding to the historical nursing intervention time for the i-th vital sign of the k-th historical patient (the previous example used the x-th time as the historical nursing intervention time; therefore, the change in vital sign data here is...). (to be represented) This represents the relevant evaluation of the impact of the j-th vital sign of the k-th historical patient on the i-th vital sign. The value represents the change in vital sign data corresponding to the historical nursing intervention time for the j-th vital sign of the k-th historical patient, and n represents the number of reference vital signs for the k-th historical patient.

[0132] Using the above process, the vital sign change coefficients of each historical patient were obtained.

[0133] Step 3-2: Based on the vital sign change coefficients of each historical patient, the vital sign data corresponding to the historical nursing intervention time of each historical patient, and the preset danger thresholds of each vital sign, obtain the danger time of each vital sign of each historical patient.

[0134] Using the vital sign change coefficients of each historical patient as the rate of change of vital sign data if no nursing intervention was performed after the intervention time, and combining the vital sign data corresponding to the historical nursing intervention time for each historical patient, as well as the preset risk thresholds for each vital sign, the risk time for each vital sign of each historical patient is obtained. In an exemplary embodiment, such as...Figure 7 As shown, a specific calculation process of the dangerous time is as follows:

[0135] Step 3-2-1: Calculate the vital sign data difference between the preset dangerous threshold of the first vital sign and the vital sign data corresponding to the historical nursing intervention time of the first vital sign of the first historical patient.

[0136] Step 3-2-2: Calculate the ratio of the vital sign data difference and the vital sign change coefficient of the first vital sign of the first historical patient as the dangerous time of the first vital sign of the first historical patient.

[0137] In an exemplary embodiment, a specific calculation formula of the dangerous time is as follows:

[0138] ;

[0139] Wherein, represents the dangerous time of the i-th vital sign of the k-th historical patient, represents the preset dangerous threshold of the i-th vital sign, represents the vital sign data corresponding to the historical nursing intervention time of the i-th vital sign of the k-th historical patient.

[0140] In the above manner, the dangerous time of each vital sign of each historical patient is obtained.

[0141] Step 4: Obtain the target early warning threshold based on the dangerous time of each vital sign of each historical patient and the historical nursing intervention time of each vital sign of each historical patient.

[0142] After obtaining the dangerous time of each vital sign of each historical patient, the target early warning threshold is obtained in combination with the historical nursing intervention time of each vital sign of each historical patient. In an exemplary embodiment, as shown, a specific acquisition process of the target early warning threshold is as follows: Figure 8

[0143] Step 4-1: Obtain the time difference between the intervention effective time and the historical nursing intervention time of the first vital sign of the first historical patient, the intervention effective time being the time of the first extreme point after the historical nursing intervention time.

[0144] ​The nursing staff can have a certain lag after the nursing intervention until the patient's vital signs change, for example, the patient's body temperature continues to decrease, and the decrease stops after several seconds of nursing intervention. This stage is the lag of the intervention. If the nursing staff intervenes late, the patient's vital sign data can still be out of the normal range due to the lag of the intervention effect, threatening the patient's life and health. Therefore, the intervention effect time of the first historical patient's first vital sign needs to be obtained. The intervention effect time is the time when the patient's vital signs improve after the nursing intervention. In an exemplary embodiment, the intervention effect time is the time when the first extreme point appears after the historical nursing intervention time of the first historical patient's first vital sign. Specifically, the vital sign time series data of the ith vital sign of the kth historical patient is obtained The data curve is obtained by curve fitting. In the data curve, the time when the first extreme point (which can be a maximum value or a minimum value, determined by the actual situation of the vital sign) appears is found after the historical nursing intervention time of the ith vital sign of the kth historical patient, and the time is the intervention effect time.

[0145] Then, the time difference between the intervention effect time of the ith vital sign of the kth historical patient and the historical nursing intervention time of the ith vital sign of the kth historical patient, i.e., the time interval, is obtained.

[0146] Step 4-2: According to the size relationship between the time difference and the dangerous time of the first vital sign of the first historical patient, the lag feature of the first vital sign of the first historical patient is obtained.

[0147] The lag feature of the ith vital sign of the kth historical patient is determined according to the size relationship between the time difference obtained in step 4-1 and the dangerous time of the ith vital sign of the kth historical patient. When the time difference corresponding to the ith vital sign of the kth historical patient is greater than the dangerous time of the ith vital sign of the kth historical patient, it means that there is a greater lag and a certain danger. In an exemplary embodiment, the ratio of the time difference corresponding to the ith vital sign of the kth historical patient to the dangerous time of the ith vital sign of the kth historical patient is taken as the lag feature of the ith vital sign of the kth historical patient.

[0148] Step 4-3: Fuse the lag features of the first vital signs of all historical patients to obtain the comprehensive lag of the first vital sign.

[0149] The comprehensive lag of the first vital sign is obtained by combining the lag features of the first vital signs of all historical patients. In this embodiment, the average value of the lag features of the first vital signs of the historical patients is taken as the comprehensive lag of the first vital sign.

[0150] In an exemplary embodiment, the specific calculation formula of the comprehensive hysteresis is given as follows:

[0151] ;

[0152] wherein, represents the comprehensive hysteresis of the i-th vital sign, represents the intervention effective time of the i-th vital sign of the k-th historical patient, represents the historical nursing intervention time of the i-th vital sign of the k-th historical patient, represents the time difference between the intervention effective time and the historical nursing intervention time, i.e., the time interval between the two times, represents the risk time of the i-th vital sign of the k-th historical patient, represents a normalization function. The normalization function here can be specifically set according to actual conditions, for example: a linear normalization method such as a maximum-minimum normalization method can be used, or the following commonly used method can be used: .

[0153] Step 4-4: According to the comprehensive hysteresis of the first vital sign, the preset risk threshold of the first vital sign, and the historical nursing intervention time of the first vital sign of each historical patient, the target warning threshold of the first vital sign is obtained.

[0154] The greater the comprehensive hysteresis, the more the warning threshold needs to be advanced to timely issue a warning and remind the nursing staff to perform nursing intervention. Therefore, according to the comprehensive hysteresis of the first vital sign, the preset risk threshold of the first vital sign, and the historical nursing intervention time of the first vital sign of each historical patient, the target warning threshold of the first vital sign is obtained.

[0155] In an exemplary embodiment, the vital sign data average value of the vital sign data corresponding to the historical nursing intervention time of the first vital sign of all historical patients is obtained.

[0156] For the first vital sign, a comprehensive hysteresis threshold is preset, the preset comprehensive hysteresis threshold is between 0-1, the higher the preset comprehensive hysteresis threshold is set, the higher the adjustment threshold of the warning threshold is, and the specific value of the preset comprehensive hysteresis threshold is set by the actual situation of the first vital sign, such as 0.3. It should be understood that the comprehensive hysteresis thresholds of different vital signs are independent of each other and can be equal or not equal.

[0157] When the comprehensive hysteresis of the i-th vital sign is greater than or equal to the comprehensive hysteresis threshold of the i-th vital sign, the warning threshold of the i-th vital sign needs to be adjusted.

[0158] The target early warning threshold is calculated by using the following calculation formula:

[0159]

[0160] Among them, The target early warning threshold of the i th vital sign is represented by The preset dangerous threshold of the i th vital sign is represented by

[0161] The comprehensive hysteresis of the i th vital sign is represented by The preset comprehensive hysteresis threshold of the i th vital sign is represented by The vital sign data average value of the i th vital sign of the historical patient in the vital sign unstable period is represented by The calculation method of is as follows: the vital sign data corresponding to the historical nursing intervention time of the i th vital sign of each historical patient in the vital sign unstable period is obtained, and then the average value of the vital sign data is calculated.

[0162] By using the above process, the target early warning threshold of each vital sign is obtained. Compared with the traditional manual setting method, the accuracy and reliability of the obtained target early warning threshold are greatly improved.

[0163] In subsequent actual application, for any vital sign, according to the size relationship between the vital sign data of the vital sign of the current patient and the corresponding target early warning threshold, if the actual vital sign data of the current patient reaches the target early warning threshold, a warning is sent to the nursing staff, so that the nursing staff can take relevant nursing intervention measures.

[0164] It should be understood that in the following, after the target early warning threshold is obtained, the nursing intervention time of the nursing staff will also change, so the target early warning threshold can be updated again by using the new nursing intervention time according to the above method. When updating, the preset dangerous threshold is replaced by the latest target early warning threshold to recalculate the comprehensive hysteresis, and whether to obtain a new target early warning threshold is judged according to the comprehensive hysteresis, and so on, so as to realize the continuous updating of the early warning threshold.

[0165] It should be noted that the above sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0166] ​The various embodiments described in this specification are presented by way of example, and each embodiment is not inherently more important than any other embodiment.

Claims

1. An intelligent processing method of anesthesia patient care data based on an operating room, characterized in that, The method comprises the following steps: acquiring vital sign time series data of each vital sign of a plurality of historical patients, and acquiring a vital sign unstable period of the vital sign time series data; obtaining an influence correlation evaluation between any two vital signs according to the correlation of each vital sign of each historical patient with other vital signs for the same vital sign unstable period; obtaining a dangerous time of each vital sign of each historical patient according to the influence correlation evaluation, vital sign data corresponding to a historical nursing intervention time of each vital sign of each historical patient, a vital sign data change amount, and a preset dangerous threshold of each vital sign; acquiring a target early warning threshold based on the dangerous time of each vital sign of each historical patient and the historical nursing intervention time of each vital sign of each historical patient; the vital signs include heart rate, blood pressure, blood oxygen saturation, respiratory rate and body temperature; the obtaining of the influence correlation evaluation between any two vital signs comprises: obtaining time series data of a first vital sign unstable period in the vital sign time series data of a first vital sign of a first historical patient to obtain first vital sign unstable time series data, and obtaining time series data of the first vital sign unstable period in the vital sign time series data of a second vital sign of the first historical patient to obtain second vital sign unstable time series data; the first historical patient is any one of the historical patients, the first vital sign and the second vital sign are any two different vital signs, and the first vital sign unstable period is a vital sign unstable period of the first vital sign of the first historical patient; obtaining the correlation of the first vital sign unstable time series data and the second vital sign unstable time series data as the correlation of the first vital sign to the second vital sign; fusing the correlation of the first vital sign to the second vital sign of all the historical patients to obtain the influence correlation evaluation of the first vital sign to the second vital sign.

2. A method for intelligent processing of anesthesia patient care data based on operating room as claimed in claim 1, wherein, the acquiring of the vital sign unstable period of the vital sign time series data comprises: acquiring a stable state reference range centered on preset vital sign stable state data of a first vital sign, the first vital sign being any vital sign; determining abnormal vital sign data not in the stable state reference range from the vital sign time series data of the first vital sign of the first historical patient, and regarding a period formed by the abnormal vital sign data in the vital sign time series data of the first vital sign of the first historical patient as a vital sign unstable period of the first vital sign of the first historical patient; the first historical patient is any one of the historical patients.

3. A method for intelligent processing of anesthesia patient care data based on operating room as claimed in claim 1, wherein, the fusing of the correlation of the first vital sign to the second vital sign of all the historical patients to obtain the influence correlation evaluation of the first vital sign to the second vital sign comprises: calculating a correlation average value of the first vital sign to the second vital sign of all the historical patients; acquiring a correlation variance of the first vital sign to the second vital sign of all the historical patients; According to the correlation average and the correlation variance, the influence correlation evaluation is obtained, the influence correlation evaluation is proportional to the correlation average and is inversely proportional to the correlation variance.

4. A method for intelligent processing of anesthesia patient care data in an operating room as claimed in claim 1, wherein, According to the influence correlation evaluation, the historical nursing intervention time corresponding to the vital sign data and the vital sign data change amount of each historical patient, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained, including: According to the influence correlation evaluation, and the vital sign data change amount of each historical patient at the historical nursing intervention time, the vital sign change coefficient of each vital sign of each historical patient is obtained. According to the vital sign change coefficient of each vital sign of each historical patient, the vital sign data of each historical patient at the historical nursing intervention time, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained.

5. A method for intelligent processing of anesthesia patient care data in an operating room as claimed in claim 4, wherein, According to the influence correlation evaluation, and the vital sign data change amount of each historical patient at the historical nursing intervention time, the vital sign change coefficient of each vital sign of each historical patient is obtained, including: According to the influence correlation evaluation of each reference vital sign of the first historical patient on the first vital sign, and the vital sign data change amount of each reference vital sign of the first historical patient at the historical nursing intervention time, the vital sign data change feature of each reference vital sign of the first historical patient is obtained; the first historical patient is any one historical patient, the first vital sign is any one vital sign, and the reference vital sign is other vital signs except the first vital sign; The vital sign data change features of each reference vital sign of the first historical patient are fused, and the vital sign data change amount of the first vital sign of the first historical patient at the historical nursing intervention time is combined to obtain the vital sign change coefficient of the first vital sign of the first historical patient.

6. A method of intelligent processing of anesthesia patient care data based on operating room as claimed in claim 5, wherein, The calculation formula of the vital sign change coefficient is as follows: ； wherein, represents a vital sign change coefficient of the i-th vital sign of the k-th historical patient, represents a vital sign data change amount corresponding to the historical nursing intervention time of the i-th vital sign of the k-th historical patient, represents an influence-related evaluation of the j-th vital sign on the i-th vital sign of the k-th historical patient, represents a vital sign data change amount corresponding to the historical nursing intervention time of the j-th vital sign of the k-th historical patient, and n represents the number of reference vital signs of the k-th historical patient.

7. A method for intelligent processing of anesthesia patient care data based on operating room as claimed in claim 4 wherein, According to the vital sign change coefficient of each vital sign of each historical patient, the vital sign data of each historical patient at the historical nursing intervention time, and the preset dangerous threshold of each vital sign, the dangerous time of each vital sign of each historical patient is obtained, including: The preset dangerous threshold of the first vital sign is calculated, and the vital sign data difference of the vital sign data of the first historical patient at the historical nursing intervention time of the first vital sign is calculated. The ratio of the vital sign data difference and the vital sign change coefficient of the first vital sign of the first historical patient is calculated as the dangerous time of the first vital sign of the first historical patient.

8. A method for intelligent processing of anesthesia patient care data in an operating room as claimed in claim 1, wherein, The target early warning threshold is obtained based on the dangerous time of each vital sign of each historical patient and the historical nursing intervention time of each vital sign of each historical patient, including: obtaining a time difference between an intervention effective time of the first vital sign of the first historical patient and the historical nursing intervention time, the intervention effective time being a time of a first extreme point occurring after the historical nursing intervention time; obtaining a hysteresis characteristic of the first vital sign of the first historical patient according to a size relationship between the time difference and a dangerous time of the first vital sign of the first historical patient; fusing hysteresis characteristics of the first vital sign of all the historical patients to obtain a comprehensive hysteresis of the first vital sign; obtaining a target early warning threshold of the first vital sign according to the comprehensive hysteresis of the first vital sign, a preset dangerous threshold of the first vital sign, and the historical nursing intervention time of the first vital sign of each historical patient.

9. A method of intelligent processing of anesthesia patient care data based on operating room as claimed in claim 8, wherein, The obtaining of the target early warning threshold of the first vital sign according to the comprehensive hysteresis of the first vital sign, the preset dangerous threshold of the first vital sign, and the vital sign data corresponding to the historical nursing intervention time of the first vital sign of each historical patient comprises: obtaining a vital sign data average of the vital sign data corresponding to the historical nursing intervention time of the first vital sign of all the historical patients; when the comprehensive hysteresis of the first vital sign is greater than or equal to a comprehensive hysteresis threshold of the first vital sign, the target early warning threshold is calculated by using the following calculation formula: ； wherein, represents a target warning threshold of the i-th vital sign, represents a preset danger threshold of the i-th vital sign, represents a comprehensive hysteresis of the i-th vital sign, represents a preset comprehensive hysteresis threshold of the i-th vital sign, represents a vital sign data average value of the vital sign data corresponding to the historical nursing intervention moment of the i-th vital sign of all historical patients.

Images