Electronic medical record assisted cardiothoracic surgery icu prognosis support system and method

Abstract

The present application relates to the technical field of medical physiological data processing, in particular to a kind of cardiac surgery ICU prognosis support system and method assisted by electronic medical record.The method determines category representative point after initial classification based on physiological data, i.e., the number of categories to be classified is determined using the relatively rough results obtained by initial classification.Furthermore, the initial postoperative abnormality difference between patient sample points and category representative points is corrected based on the initial classification results, and the final classification results of each sample point are determined using the corrected postoperative abnormality difference.The present application can make the corrected postoperative abnormality difference obtained finally more accurately represent the distance between sample points and category representative points, and then determine accurate classification results, which is convenient for medical staff to provide postoperative prognosis support for patients.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical physiological data processing, in particular to a cardiac surgery ICU prognosis support system and method assisted by electronic medical records. BACKGROUND

[0002] In the management of postoperative patients in cardiac surgery, accurate medical record classification is crucial for optimizing resource allocation, reducing the burden on medical staff, and improving patient recovery quality. In the prior art, patients can be classified according to the postoperative physiological data recorded in the electronic medical record in the ICU, and patients with more serious postoperative conditions can be classified into a category for targeted intervention. However, in the prior art, only the abnormality of physiological data is used for classification, without considering the different surgical methods, different treatment methods, and different personal conditions of patients in actual situations, resulting in inaccurate classification results and inability to assist medical staff in effective postoperative treatment. SUMMARY

[0003] In order to solve the technical problem of inaccurate classification of postoperative patients in the prior art, which leads to the inability to assist medical staff in effective postoperative treatment, the purpose of the present application is to provide a cardiac surgery ICU prognosis support system and method assisted by electronic medical records, and the technical solution adopted is as follows:

[0004] The present application proposes a cardiac surgery ICU prognosis support method assisted by electronic medical records, which comprises:

[0005] Obtaining postoperative multi-dimensional physiological data of patients based on an electronic medical record database, and obtaining postoperative abnormality of each patient according to the physiological data;

[0006] Classifying all patients as sample points according to the postoperative abnormality to obtain class representative points in the initial classification result, and initial postoperative abnormality difference between each sample point and the class representative points;

[0007] For each sample point, adjusting the postoperative abnormality according to the distribution uniqueness of the same surgical method in the initial classification result and the treatment time abnormality to obtain the degree of medical record abnormality;

[0008] Obtaining a distance adjustment factor between each sample point and each class representative point according to the difference in medical record abnormality between each sample point and each class representative point; reducing the initial postoperative abnormality difference between each class representative point and the sample points of the same surgical method, and increasing the initial postoperative abnormality difference between each class representative point and the sample points of different surgical methods according to the distance adjustment factor, to obtain the corrected postoperative abnormality difference between each class representative point and each sample point;

[0009] According to the modified postoperative abnormality difference, the class representative point to which each sample point belongs is re-determined, a final classification result is obtained, and prognosis support is performed according to the final classification result.

[0010] Further, the physiological data includes first physiological data of multiple dimensions in a preset short time period after surgery and second physiological data of multiple dimensions in a preset long time period after surgery; the dimensions include blood pressure, heart rate, and respiratory rate.

[0011] Further, the method for obtaining the postoperative abnormality includes:

[0012] With physiological data in a standard short time after surgery as a reference, a Z-score absolute value of average first physiological data of each dimension in a preset short time period is obtained, the Z-score absolute values of the respective dimensions are weighted and summed according to preset dimension weights to obtain a short-time postoperative score; the dimension weight of blood pressure is the largest, and the dimension weight of respiratory rate is the smallest;

[0013] Data differences between average second physiological data of each dimension and standard average physiological data in a long time period after surgery are obtained, and the data differences of each dimension are normalized and averaged to obtain a long-time postoperative score;

[0014] Correlation coefficients of the second physiological data between different dimensions are obtained, and the correlation coefficients are averaged to obtain a comprehensive correlation factor;

[0015] According to the short-time postoperative score, the long-time postoperative score, and the comprehensive correlation factor, a postoperative abnormality of a patient is obtained.

[0016] Further, the method for obtaining the medical record abnormality degree includes:

[0017] For each sample point, a sample point proportion of sample points of the same surgical method in an initial classification result to which the sample point belongs is obtained, a treatment time difference between a treatment time of the sample point and a preset standard treatment time is obtained, and an abnormality adjustment factor is obtained according to the sample point proportion and the treatment time difference;

[0018] The abnormality adjustment factor is used as a weight to adjust the postoperative abnormality of the sample point, and the medical record abnormality degree is obtained.

[0019] Further, the method for obtaining the abnormality adjustment factor includes:

[0020] The sample point proportion is negatively correlated and normalized to obtain a surgical method category weight, and the treatment time difference is multiplied by the surgical method category weight after normalization to obtain the abnormality adjustment factor.

[0021] Further, the method for obtaining the distance adjustment factor includes:

[0022] For each sample point, the distance adjustment factor between the sample point and the category representative point is obtained by taking the difference in medical record abnormality degree between the sample point and the category representative point as the numerator and taking the medical record abnormality degree of the category representative point as the denominator.

[0023] Further, the method for obtaining the medical record abnormality degree further comprises:

[0024] For each sample point, the data difference between the average second physiological data in the blood pressure dimension and the standard blood pressure data is taken as the initial blood pressure abnormality; the longest continuous abnormal time period in the blood pressure data sequence composed of the second physiological data in the blood pressure dimension is obtained; the blood pressure abnormality factor is obtained according to the abnormality degree of the second physiological data in the longest continuous abnormal time period and the length of the longest continuous abnormal time period; and the final blood pressure abnormality is obtained according to the blood pressure abnormality factor and the initial blood pressure abnormality.

[0025] The abnormality adjustment factor is multiplied by the final blood pressure abnormality to obtain an adjustment weight, and the postoperative abnormality of the sample point is adjusted according to the adjustment weight to obtain the medical record abnormality degree.

[0026] Further, the method for obtaining the blood pressure abnormality factor comprises:

[0027] In the blood pressure dimension, for each second physiological data in the longest continuous abnormal time period, the difference between the second physiological data and the average of all second physiological data corresponding to the sample point is taken as the abnormality degree; and the product of the cumulative value of the abnormality degrees in the longest continuous abnormal time period and the length of the longest continuous abnormal time period is taken as the blood pressure abnormality factor.

[0028] Further, the method for obtaining the corrected postoperative abnormality difference comprises:

[0029] For each category representative point, if the sample point and the category representative point belong to the same surgical method, the normalized distance adjustment factor is taken as a correction weight, the correction weight is multiplied by the initial postoperative abnormality difference to obtain the corrected postoperative abnormality difference; and if the sample point and the category representative point belong to different surgical methods, the normalized distance adjustment factor is subtracted by a positive integer 1 to obtain a correction weight, and the correction weight is multiplied by the initial postoperative abnormality difference to obtain the corrected postoperative abnormality difference.

[0030] The application further provides a cardiac surgery ICU prognosis support system assisted by an electronic medical record, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of any one of the cardiac surgery ICU prognosis support methods assisted by the electronic medical record.

[0031] The application has the following advantages:

[0032] After initial classification based on physiological data, the application determines a category representative point, that is, determines the number of categories to be classified by using the relatively rough result obtained by initial classification. Further, the initial postoperative abnormality difference between the patient sample point and the category representative point is corrected based on the initial classification result, and the final classification result of each sample point is determined by using the corrected postoperative abnormality difference. In the correction process, considering the difference in surgical methods, the distribution uniqueness of the same surgical method of each sample point in the initial classification result is analyzed, that is, the greater the distribution uniqueness, the more the patient corresponding to the sample point needs to be focused on; further, considering the treatment time abnormality, the more abnormal the treatment time, the worse the recovery of the patient corresponding to the sample point, and the more the patient needs to be focused on, and thus the medical record abnormality degree is obtained. Taking the category representative point as the basis, the distance adjustment factor can be obtained by analyzing the difference in the medical record abnormality degree, and the initial postoperative abnormality difference between the sample point and the category representative point is corrected based on the distance adjustment factor and the consistency of the surgical method, so that the corrected postoperative abnormality difference obtained finally can more accurately represent the distance between the sample point and the category representative point, and thus an accurate classification result is determined, facilitating the prognosis support of the postoperative patient by medical staff. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained from these drawings without creative labor.

[0034] Figure 1 A cardiac surgery ICU prognosis support method assisted by an electronic medical record provided by one embodiment of the present application is shown in a flowchart. DETAILED DESCRIPTION

[0035] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined inventive objectives, the following describes in detail the specific implementation, structure, features and effects of a cardiac surgery ICU prognosis support system and method assisted by electronic medical records according to the present application, in combination with the accompanying 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.

[0036] 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.

[0037] The specific scheme of a cardiac surgery ICU prognosis support system and method assisted by electronic medical records provided by the present application is described in detail below in combination with the accompanying drawings.

[0038] Please refer to Figure 1 which shows a cardiac surgery ICU prognosis support method flowchart provided by an embodiment of the present application, the method comprises:

[0039] Step S1: obtaining postoperative multidimensional physiological data of patients based on an electronic medical record database, and obtaining postoperative abnormality of each patient according to the physiological data.

[0040] The embodiment of the present application aims to analyze medical records of the category of cardiac surgery. For postoperative physiological data of cardiac surgery ICU patients, if the patient's surgery is successful and the patient recovers normally after the surgery, the patient's postoperative physiological data should be within a normal range. If it is not within the range, it means that the patient is more abnormal and has greater postoperative abnormality. The electronic medical record database can collect data through the sign monitoring instrument in the ICU and upload it to the medical data management platform. Therefore, the postoperative multidimensional physiological data of patients can be obtained based on the electronic medical record database, and the postoperative abnormality of each patient can be obtained according to the physiological data.

[0041] Preferably, in the embodiment of the present application, considering that the physiological data in a short time after surgery can effectively represent the state of surgery implementation, can reflect the initial reaction of the patient to the surgery and the preliminary recovery situation, the physiological data in a short time can assist medical staff to quickly assess the physiological state of the patient, and then timely adjust the treatment plan. Further considering that the physiological data in a long period of time after surgery can reflect the recovery degree of the patient. Therefore, the physiological data of the embodiment of the present application includes: a plurality of dimensions of first physiological data in a preset short period of time after surgery, and a plurality of dimensions of second physiological data in a preset long period of time after surgery. For patients in cardiac surgery, the main physiological data includes blood pressure, heart rate and respiratory rate. The monitoring of the physiological data in these three dimensions can effectively determine the health status of the patient. For example, a sudden increase or decrease in postoperative blood pressure may indicate bleeding, shock or other circulatory system problems; a postoperative tachycardia may indicate pain, anxiety or heart problems, while a postoperative bradycardia may indicate heart conduction system problems or drug side effects; a sudden increase in postoperative respiratory rate may indicate pain, anxiety or pulmonary complications, while a slow respiratory rate may indicate anesthesia or respiratory depression. Therefore, the dimensions of the physiological data in the embodiment of the present application include blood pressure, heart rate and respiratory rate.

[0042] In a specific implementation manner of the embodiment of the present application, the preset short period of time after surgery is set to 24 hours after surgery, and the preset long period of time after surgery is set to one week after surgery.

[0043] Further, on the basis of the physiological data including the first physiological data and the second physiological data, in an embodiment of the present application, the method for obtaining postoperative abnormality includes:

[0044] The first physiological data and the second physiological data are analyzed by different methods in the embodiment of the present application. Considering that the first physiological data represents the physiological data in a short period of time after surgery, the patient's condition in a short time after surgery needs to be observed. Therefore, the embodiment of the present application takes the standard physiological data in a short time after surgery as a reference to obtain the Z-score absolute value of the average first physiological data in each dimension in a preset short time. It should be noted that the Z-score absolute value is the result of Z-standardization of data, that is, the process of obtaining the Z-score absolute value in the embodiment of the present application can be regarded as adding the average first physiological data to the set of standard physiological data in a short time after surgery as a new data, and further using the Z-standardization algorithm to obtain the Z-score absolute value of the new data, so that the Z-score absolute value can represent the difference between the average first physiological data of the current patient in a short time after surgery and the physiological data of the normal patient, that is, the larger the Z-score absolute value, the more abnormal the first physiological data in the dimension, and the more the patient needs to be focused on, and the Z-score absolute value can also eliminate the dimension factor of each dimension, facilitating the subsequent operation of data.

[0045] In the patients in the category of cardiac surgery, blood pressure can best represent the overall state of the patient after the operation, followed by heart rate, and finally respiratory rate, so the embodiments of the present application set different weights for the three dimensions, wherein the dimension weight of blood pressure is the largest, and the dimension weight of respiratory rate is the smallest. Based on the dimension weight, the absolute values of the Z scores of each dimension are weighted and summed to obtain the short-term postoperative score of the patient. The larger the short-term postoperative score is, the more abnormal the postoperative state of the patient is, and the more attention is needed. In the embodiments of the present application, the dimension weight of blood pressure is set to 0.5, the dimension weight of heart rate is set to 0.3, and the dimension weight of respiratory rate is set to 0.2. It should be noted that those skilled in the art can set the weights according to the specific implementation scene, as long as the dimension weight of blood pressure is the largest and the dimension weight of respiratory rate is the smallest.

[0046] For the second physiological data in a long postoperative period, the reference is not as strong as the physiological data in a short period, and the data difference between the average second physiological data of each dimension and the standard average physiological data in a long postoperative period can be directly obtained, and then the data difference of each dimension is normalized. The purpose of normalization is to eliminate the influence of dimension, and the average of the normalized data differences of all dimensions can obtain the long-term postoperative score. The larger the long-term postoperative score is, the more abnormal the physiological data of the patient in the long postoperative period is.

[0047] It should be noted that the standard physiological data in a short postoperative period and the standard average physiological data in a long postoperative period can be obtained by statistical analysis of the electronic medical record database, which will not be repeated here. The normalization method used in the embodiments of the present application can use range standardization, hyperbolic tangent function mapping, linear normalization function mapping and other basic mathematical algorithms, which are well known to those skilled in the art, and will not be repeated here.

[0048] Further considering that in the normal state of the patient in a long postoperative period, the increase of blood pressure will lead to the increase of heart rate, and the increase of heart rate will also increase the respiratory rate, that is, the physiological data in the three dimensions show obvious correlation, so the correlation coefficient of the second physiological data between different dimensions is obtained, and the correlation coefficient is averaged to obtain a comprehensive correlation factor. That is, the larger the comprehensive correlation factor is, the more normal the postoperative recovery state of the patient is. In the embodiments of the present application, the correlation coefficient is selected as the Pearson correlation coefficient, that is, the second data of each dimension in a long postoperative period is regarded as a sequence, three data sequences are obtained, and the Pearson correlation coefficients between the three sequences are calculated.

[0049] The greater the short-term postoperative score and the long-term postoperative score, the more abnormal the postoperative state of the patient, and the more attention is needed. The greater the comprehensive correlation factor, the more normal the postoperative state of the patient. Therefore, the postoperative abnormality of the patient is obtained according to the short-term postoperative score, the long-term postoperative score and the comprehensive correlation factor. That is, the short-term postoperative score and the long-term postoperative score are positively correlated with the postoperative abnormality, and the comprehensive correlation factor is negatively correlated with the postoperative abnormality.

[0050] As an example, in a specific implementation manner of the embodiment of the present application, the postoperative abnormality is obtained by multiplying the short-term postoperative score, the long-term postoperative score and the negative correlation factor after mapping and normalization. The method of mapping and normalization of the negative correlation in the embodiment of the present application can be selected from the basic mathematical algorithms well known to those skilled in the art, for example, the opposite number of data is taken as the power of the exponential function with the natural constant as the base.

[0051] Step S2: All patients are regarded as sample points for initial classification according to the postoperative abnormality, and the class representative points in the initial classification result and the initial postoperative abnormality difference between each sample point and the class representative point are obtained.

[0052] After obtaining the postoperative abnormality of each patient, all patients are regarded as sample points for initial classification according to the postoperative abnormality. It should be noted that the initial classification result of the initial classification is a relatively rough classification result, which only considers the abnormality represented by the physiological data, but does not consider the specific conditions such as the type of surgery performed by different patients and the treatment time. Therefore, the initial classification result obtained by the embodiment of the present application is only used to determine the number of categories to be finally classified, that is, to screen the class representative points in the initial classification result. The class representative points can be regarded as representative points in a category, representing the postoperative abnormality information of a category. That is, there are several categories in the final classification result represented by several class representative points. There is an initial postoperative abnormality difference between each sample point and the class representative point, that is, the absolute value of the difference between the postoperative abnormality of the sample point and the postoperative abnormality of the class representative point is the initial postoperative abnormality difference.

[0053] It should be noted that in the embodiment of the present application, an abnormality threshold is set, patients with a postoperative abnormality greater than or equal to 0.7 are classified as one category, patients with a postoperative abnormality greater than or equal to 0.4 and less than 0.7 are classified as one category, and patients with a postoperative abnormality less than 0.4 are classified as one category. Thus, three patient categories are obtained. The sample point with the maximum postoperative abnormality in each patient category is selected as the class representative point.

[0054] Step S3: For each sample point, the postoperative abnormality is adjusted according to the distribution uniqueness of the same type of surgery in the initial classification result to which the sample point belongs and the treatment time abnormality, and the medical record abnormality degree is obtained.

[0055] For patients with the same surgical method, if the patients belong to the same category in the initial classification result and the severity of the cases is small, it can be determined that the patients belong to the same category, and the sample distance between the patients is reduced on the basis of the initial classification result, so that the patients are more likely to be clustered into one category. For patients with different surgical methods but similar severity of the disease, it indicates that there are some key differences between the patients, and the patients are no longer suitable to be classified into one category, and the sample distance between the patients needs to be increased. Therefore, in order to obtain an accurate classification result, it is necessary to first evaluate the abnormality degree of the medical record of each patient.

[0056] The abnormality degree of the medical record represents the abnormality of the patient's disease and postoperative state, and therefore can be obtained on the basis of the postoperative abnormality obtained in the above step. Embodiments of the present application consider that, for a patient, the more the same type of surgical method of the patient is distributed in the category in the initial classification result, the more common the surgical method is, the less unique the distribution is, and the more experienced the medical staff is in handling such cases, which indicates that the abnormality degree of the medical record of the patient is smaller; the longer the treatment time of the patient is, the greater the abnormality of the treatment time of the patient is, which indicates that the patient's disease is more serious or abnormal, and the abnormality degree of the medical record is greater. Therefore, embodiments of the present application adjust the postoperative abnormality according to the distribution uniqueness of the same type of surgical method in the classification result and the abnormality of the treatment time, to obtain the abnormality degree of the medical record.

[0057] Preferably, in an embodiment of the present application, the method for obtaining the abnormality degree of the medical record comprises:

[0058] For each sample point, the proportion of the same type of surgical method of the sample point in the category in the initial classification result is obtained, which can reflect the distribution characteristics of the same type of surgical method of a certain sample point in the category in the initial classification result. The treatment time difference between the treatment time of the sample point and the preset standard treatment time is obtained. In embodiments of the present application, it is considered that the longer the treatment time is, the more serious the disease is, and the more the patient's state needs to be monitored, so the difference between the treatment time of the sample point and the preset standard treatment time is taken as the treatment time difference. The smaller the sample point proportion is, the greater the treatment time difference is, which indicates that the surgical type of the sample point is more unique and the disease needs more attention, so the abnormality adjustment factor can be obtained according to the sample point proportion and the treatment time difference.

[0059] The abnormality adjustment factor is taken as a weight to adjust the postoperative abnormality of the sample point, to obtain the abnormality degree of the medical record. For each sample point, including the category representative point, the corresponding abnormality degree of the medical record can be obtained. The greater the abnormality adjustment factor corresponding to the sample point is, the more abnormal the disease of the patient corresponding to the sample point is, and the greater the abnormality degree of the medical record is.

[0060] Further, in embodiments of the present application, the method for obtaining the abnormality adjustment factor comprises:

[0061] The sample point proportion is negatively correlated and normalized to obtain the surgical method category weight; the treatment time difference is normalized and multiplied by the surgical method category weight to obtain the abnormal adjustment factor. The obtained abnormal adjustment factor is data with a value range of 0 to 1. Because each sample point can obtain an abnormal adjustment factor, it can be considered that the postoperative abnormality of each sample point is reduced in different degrees according to the abnormal adjustment factor, and the abnormality degree of the medical record is obtained. The more abnormal the disease is, the smaller the degree of reduction is.

[0062] As an example, the medical record abnormality degree is expressed by the formula: ; wherein is the medical record abnormality degree of the i th sample point, is the number of sample points of the same surgical method in the category of the initial classification result to which the i th sample point belongs, is the number of all sample points in the category of the initial classification result to which the i th sample point belongs, exp( ) is an exponential function with a natural constant as the base, is the treatment time of the i th sample point, is the standard treatment time of the patient corresponding to the i th sample point, norm( ) is a normalization function, is the postoperative abnormality of the i th sample point.

[0063] In the medical record abnormality degree formula, the sample point proportion is negatively correlated and normalized by using an exponential function with a natural constant as the base. The postoperative abnormality is weighted by multiplication.

[0064] Preferably, in another embodiment of the present application, further considering that the blood pressure data in the physiological data of the cardiac surgery patient has strong reference significance, in order to further improve the reference of the medical record abnormality degree, the method for obtaining the medical record abnormality degree further comprises:

[0065] For each sample point, the data difference between the average second physiological data in the blood pressure dimension and the standard blood pressure data is taken as the initial blood pressure abnormality. That is, the greater the data difference, the more abnormal the second physiological data in the blood pressure dimension is.

[0066] The longest continuous abnormal time period in the blood pressure data sequence composed of the second physiological data in the blood pressure dimension is obtained. The longest continuous abnormal time period can be used as a reference for analyzing the abnormal state of the blood pressure data, that is, it is considered that the blood pressure data of the patient appears abnormal for a long time period within the longest continuous abnormal time period. According to the abnormality degree of the second physiological data in the longest continuous abnormal time period and the length of the longest continuous abnormal time period, a blood pressure abnormality factor is obtained. That is, the greater the abnormality degree of the blood pressure data in this time period and the longer the abnormality lasts, the more abnormal the blood pressure state of the patient is, and the greater the blood pressure abnormality factor is.

[0067] The final blood pressure abnormality is obtained according to the blood pressure abnormality factor and the initial blood pressure abnormality. That is, the greater the blood pressure abnormality factor, the greater the blood pressure abnormality, and the greater the final blood pressure abnormality. In one specific implementation of the embodiment of the present application, the product of the blood pressure abnormality factor and the blood pressure abnormality is taken as the final blood pressure abnormality, which is expressed by the formula as follows: ; wherein is the final blood pressure abnormality of the i th sample point, is the initial blood pressure abnormality of the i th sample point, is the blood pressure abnormality factor of the i th sample point.

[0068] The abnormality adjustment factor is multiplied by the final blood pressure abnormality to obtain an adjustment weight, and the postoperative abnormality of the sample point is adjusted according to the adjustment weight to obtain the abnormality degree of the medical record. That is, on the basis of the above-mentioned embodiment, as an example, the abnormality degree of the medical record of the embodiment of the present application is expressed by the formula as follows: ; wherein the parameters of the formula are explained and the basic logic is clearly described in the above-mentioned embodiment description, and will not be repeated here.

[0069] In the embodiment of the present application, the method for obtaining the longest continuous abnormal time period comprises: in the second physiological data set under the blood pressure dimension in a long postoperative period, abnormal data points are obtained by using the LOF algorithm, wherein the abnormal data points that are continuous in time sequence form an abnormal time period, and the longest continuous abnormal time period is the longest continuous abnormal time period.

[0070] Further, in some implementations of the embodiment of the present application, the method for obtaining the blood pressure abnormality factor comprises:

[0071] Under the blood pressure dimension, for each second physiological data in the longest continuous abnormal time period of the sample point, the difference between the second physiological data and the average value of all second physiological data corresponding to the sample point is taken as the abnormality degree. That is, the average value of the blood pressure data of all second physiological data corresponding to the sample point is taken as a reference value, and the more the blood pressure data deviates from the reference value in the longest continuous abnormal time period, the more abnormal the blood pressure data is.

[0072] The cumulative value of the abnormality degree in the longest continuous abnormal time period is multiplied by the length of the longest continuous abnormal time period to obtain the blood pressure abnormality factor. As an example, it is expressed by the formula as follows: ; wherein is the blood pressure abnormality factor of the i th sample point, is the length of the longest continuous abnormal time period of the i th sample point, is the x th second physiological data in the longest continuous abnormal time period of the i th sample point under the blood pressure dimension, average value of all second physiological data corresponding to the i th sample point in the blood pressure dimension.

[0073] Step S4: obtaining a distance adjustment factor between each sample point and each category representative point according to the medical record abnormality degree difference between each sample point and each category representative point; and reducing the initial postoperative abnormality difference between each category representative point and the sample points of the same surgical method, and increasing the initial postoperative abnormality difference between each category representative point and the sample points of different surgical methods, to obtain a modified postoperative abnormality difference between each category representative point and each sample point.

[0074] For a category representative point and a sample point, if they are of the same surgical method, the smaller the medical record abnormality degree difference between them, the greater the similarity between them, and thus the distance between them needs to be shortened. If they are of different surgical methods but have a small medical record abnormality degree difference, it indicates that there is a key difference between the diseases and surgical methods, and thus the distance between them needs to be increased in order to achieve accurate classification. Therefore, the distance adjustment factor between each sample point and each category representative point is obtained according to the medical record abnormality degree difference between each sample point and each category representative point. Based on the distance adjustment factor, the initial postoperative abnormality difference between the category representative point and the sample point is adjusted to obtain the modified postoperative abnormality difference between each category representative point and each sample point.

[0075] Preferably, in the embodiment of the present application, the method for obtaining the distance adjustment factor comprises:

[0076] For each sample point, the medical record abnormality degree difference between the sample point and the category representative point is taken as the numerator, and the medical record abnormality degree of the category representative point is taken as the denominator to obtain the distance adjustment factor between the sample point and the category representative point. That is, in the embodiment of the present application, the distance adjustment factor is represented as the proportion of the case abnormality degree difference in the case abnormality degree of the category representative point. The greater the proportion, the lower the similarity between the diseases of the sample point and the category representative point, and the greater the distance adjustment factor.

[0077] It should be noted that the medical record abnormality degree difference is the absolute value of the difference between the two medical record abnormality degrees.

[0078] Preferably, in an embodiment of the present application, the method for obtaining the modified postoperative abnormality difference comprises:

[0079] For each category representative point, if the sample point and the category representative point belong to the same surgical method, the normalized distance adjustment factor is taken as a correction weight, and the correction weight is multiplied by the initial postoperative abnormality difference to obtain the modified postoperative abnormality difference. The formula is: ; wherein is the initial postoperative abnormality difference between the i-th sample point and the j-th category representative point, is the initial postoperative abnormality difference between the i-th sample point and the j-th category representative point, is the distance adjustment factor between the i-th sample point and the j-th category representative point after normalization. That is, the greater the distance adjustment factor, the greater the difference in disease between the sample point and the category representative point, and the less similar they are, so the initial postoperative abnormality difference needs to be adjusted to be larger, and the distance between the sample point and the category representative point is increased; on the contrary, the smaller the distance adjustment factor, the more similar the sample point and the category representative point are. Because the distance adjustment factor is normalized, the value range is between 0 and 1, so by multiplying it, the initial postoperative abnormality difference can be further reduced.

[0080] If the sample point and the category representative point belong to different surgical methods, subtract the normalized distance adjustment factor from the positive integer 1 to obtain the correction weight, and multiply the correction weight by the initial postoperative abnormality difference to obtain the corrected postoperative abnormality difference. It is expressed by the formula: Because the sample point and the category representative point belong to different surgical methods, if the distance adjustment factor is small at this time, it means that the sample point and the category representative point are more similar in disease state, but they do not belong to the same surgical method, which means that there is a difference in the key information, so the initial postoperative abnormality difference will be larger, and after being multiplied by the initial postoperative abnormality difference, a larger corrected postoperative abnormality difference is obtained. It should be noted that because each sample point needs to be adjusted with each category representative point for postoperative abnormality difference, the correction weight obtained is between 0 and 1. However, because the correction weight is different, it is equivalent to reducing the different initial postoperative abnormality differences to different degrees. That is, the "greater", "smaller", "larger" and "smaller" described in the embodiments of the present application are comparative features in the results after correction, not directly increasing or decreasing the original data, but uniformly reducing to different degrees.

[0081] Step S5: Redetermine the category representative point to which each sample point belongs according to the corrected postoperative abnormality difference to obtain the final classification result, and provide prognosis support according to the final classification result.

[0082] For each sample point, the category representative point corresponding to the smallest corrected postoperative abnormality difference is selected as the category representative point to which it belongs, and the final classification result is obtained. In the description of the above embodiments of the present application, there are three category representative points, so after the corrected classification result, there are still three patient categories. According to the final classification result, effective prognosis support can be provided for medical staff.

[0083] According to the classification result, the doctor can identify different patient categories, customize a more personalized treatment plan for each category, or conduct more in-depth research on specific categories. By monitoring and evaluating the patient categories with higher abnormality, the doctor can more accurately assess the health status and postoperative recovery needs of each patient.

[0084] To sum up, in the embodiment of the application, after initial classification based on physiological data, the category representative point is determined, that is, the number of categories to be classified is determined by using the relatively rough result obtained by initial classification. Further, the initial postoperative abnormality difference between the patient sample point and the category representative point is corrected based on the initial classification result, and the final classification result of each sample point is determined by using the corrected postoperative abnormality difference. In the correction process, considering the difference in surgical method, the distribution uniqueness of the same surgical method of each sample point in the initial classification result is analyzed, that is, the greater the distribution uniqueness, the more the patient corresponding to the sample point needs to be focused on; further considering the treatment time abnormality, the more abnormal the treatment time, the worse the recovery of the patient corresponding to the sample point, the more the patient needs to be focused on, and thus the medical record abnormality degree is obtained. Based on the category representative point, the distance adjustment factor can be obtained by analyzing the medical record abnormality degree difference, and the initial postoperative abnormality difference between the sample point and the category representative point is corrected based on the distance adjustment factor and the consistency of the surgical method, which can make the final corrected postoperative abnormality difference more accurately represent the distance between the sample point and the category representative point, and thus determine an accurate classification result, facilitating the prognosis support of medical staff for postoperative patients.

[0085] Based on the same inventive concept, the application also proposes a cardiac surgery ICU prognosis support system assisted by electronic medical records, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of any one of the cardiac surgery ICU prognosis support methods assisted by electronic medical records when executing the computer program.

[0086] It should be noted that the above-mentioned sequence of the embodiments of the 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, multitasking and parallel processing are possible or can be advantageous.

[0087] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments.

Claims

1. A cardiac surgery ICU prognosis support method with electronic medical record assistance, characterized by, The method comprises: obtaining postoperative multidimensional physiological data of patients based on an electronic medical record database, and obtaining postoperative abnormality of each patient according to the physiological data; the physiological data comprises first physiological data of multiple dimensions within a preset short postoperative time period and second physiological data of multiple dimensions within a preset long postoperative time period; the dimensions comprise blood pressure, heart rate and respiratory rate; initially classifying all patients as sample points according to the postoperative abnormality, obtaining a class representative point in an initial classification result, and an initial postoperative abnormality difference between each sample point and the class representative point; for each sample point, adjusting the postoperative abnormality according to the distribution uniqueness of the same surgical method in the initial classification result and the treatment time abnormality, and obtaining a medical record abnormality degree; obtaining a distance adjustment factor between each sample point and each class representative point according to the medical record abnormality degree difference between each sample point and each class representative point; reducing the initial postoperative abnormality difference between each class representative point and the sample points of the same surgical method, and increasing the initial postoperative abnormality difference between each class representative point and the sample points of different surgical methods according to the distance adjustment factor, to obtain a modified postoperative abnormality difference between each class representative point and each sample point; redetermining the class representative point to which each sample point belongs according to the modified postoperative abnormality difference, obtaining a final classification result, and performing prognosis support according to the final classification result; The method for obtaining the postoperative abnormality comprises: taking physiological data within a standard short postoperative time period as a reference to obtain a Z-score absolute value of average first physiological data of each dimension within a preset short time period, weighting and summing the Z-score absolute values of various dimensions according to preset dimension weights to obtain a short-time postoperative score; wherein the dimension weight of blood pressure is the largest, and the dimension weight of respiratory rate is the smallest; obtaining a data difference between average second physiological data of each dimension and standard average physiological data within a long postoperative time period, normalizing and averaging the data difference of each dimension to obtain a long-time postoperative score; obtaining a correlation coefficient of the second physiological data between different dimensions, averaging the correlation coefficients to obtain a comprehensive correlation factor; obtaining the postoperative abnormality of the patient according to the short-time postoperative score, the long-time postoperative score and the comprehensive correlation factor.

2. The method of claim 1, wherein the method is a method of cardiac surgery ICU prognosis support using an electronic medical record. The method for obtaining the medical record abnormality degree comprises: for each sample point, obtaining a sample point proportion of the same surgical method in the initial classification result to which the sample point belongs; obtaining a treatment time difference between the treatment time of the sample point and a preset standard treatment time; and obtaining an abnormality adjustment factor according to the sample point proportion and the treatment time difference; using the abnormality adjustment factor as a weight to adjust the postoperative abnormality of the sample point, and obtaining the medical record abnormality degree.

3. The method of claim 2, wherein the method is a method of cardiac surgery ICU prognosis support using an electronic medical record as an aid, characterized by, The method for obtaining the abnormality adjustment factor comprises: mapping and normalizing the negative correlation of the sample point proportion to obtain a surgical method category weight; multiplying the normalized treatment time difference by the surgical method category weight to obtain the abnormality adjustment factor.

4. The method of claim 1, wherein the method is a method of cardiac surgery ICU prognosis support using an electronic medical record, characterized by, The method for obtaining the distance adjustment factor comprises: For each sample point, the distance adjustment factor between the sample point and the category representative point is obtained by taking the difference in medical record abnormality degree between the sample point and the category representative point as the numerator and taking the medical record abnormality degree of the category representative point as the denominator.

5. The method of claim 2, wherein the method is a method of cardiac surgery ICU prognosis support using an electronic medical record, characterized by, The method for obtaining the medical record abnormality degree further comprises: For each sample point, the data difference between the average second physiological data in the blood pressure dimension and the standard blood pressure data is taken as the initial blood pressure abnormality; the longest continuous abnormal time period in the blood pressure data sequence composed of the second physiological data in the blood pressure dimension is obtained; the blood pressure abnormality factor is obtained according to the abnormality degree of the second physiological data in the longest continuous abnormal time period and the length of the longest continuous abnormal time period; and the final blood pressure abnormality is obtained according to the blood pressure abnormality factor and the initial blood pressure abnormality. The abnormality adjustment factor is multiplied by the final blood pressure abnormality to obtain an adjustment weight, and the postoperative abnormality of the sample point is adjusted according to the adjustment weight to obtain the medical record abnormality degree.

6. The method of claim 5, wherein the method is a method of cardiac surgery ICU prognosis support using an electronic medical record, characterized by, The method for obtaining the blood pressure abnormality factor comprises: In the blood pressure dimension, for each second physiological data in the longest continuous abnormal time period, the difference between the second physiological data and the average of all second physiological data corresponding to the sample point is taken as the abnormality degree; and the product of the cumulative value of the abnormality degrees in the longest continuous abnormal time period and the length of the longest continuous abnormal time period is taken as the blood pressure abnormality factor.

7. The method of claim 1, wherein the method is a method of cardiac surgery ICU prognosis support using an electronic medical record. The method for obtaining the corrected postoperative abnormality difference comprises: For each category representative point, if the sample point and the category representative point belong to the same surgical method, the normalized distance adjustment factor is taken as a correction weight, the correction weight is multiplied by the initial postoperative abnormality difference to obtain the corrected postoperative abnormality difference; if the sample point and the category representative point belong to different surgical methods, a positive integer 1 is subtracted from the normalized distance adjustment factor to obtain a correction weight, and the correction weight is multiplied by the initial postoperative abnormality difference to obtain the corrected postoperative abnormality difference.

8. A cardiac surgery ICU prognosis support system with electronic medical record assistance, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized by, The processor executes the computer program to implement the steps of the cardiac surgery ICU prognosis support method assisted by electronic medical records according to any one of claims 1-7.

Images