Data collection and analysis system and method for intracranial postoperative infected patient

Abstract

The invention relates to the technical field of intelligent medical treatment, and particularly discloses a data collection and analysis system and method for an intracranial postoperative infected patient, and the method comprises the steps: obtaining the medical record information of the patient, and carrying out the initial classification of the data in the medical record information according to a preset type; performing secondary classification on the primarily classified data based on preset infection factors; comparing each type of data after the secondary classification, and calculating the relevancy between the types; displaying each type of data based on the calculated relevancy; the display mode comprises a pie chart, a histogram and a multi-dimensional relation table. According to the method, the problems that data of an intracranial postoperative infected patient is mottle and complex and accurate positioning is difficult are solved through an efficient data processing means, powerful information is provided for clinical causal relationships, and clear causal relationships are established among various factors; the potential risk that a certain factor or some factors develop into diseases can be directly exposed.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical technology, and specifically to a data collection and analysis system and method for patients with intracranial postoperative infections. Background Art

[0002] In recent years, the number of patients undergoing neurosurgical craniotomy has been increasing, and the skills of surgical staff have been improving day by day. However, the continuous increase in the intracranial infection rate after surgery still cannot be avoided. For the risk factors that may cause intracranial infection after craniotomy, many clinical medical workers have conducted rigorous statistical analyses based on the cases in their own hospitals. Due to the different operating room environments, the qualities of medical staff, and the hospitalization environments, different research conclusions are also different.

[0003] Many doctors have studied a large number of craniotomy case materials and used methods such as retrospective analysis, single-factor variable analysis, and x2 test to statistically analyze the relevant factors that may cause intracranial infection: indwelling drainage tubes after surgery, cerebrospinal fluid leakage after surgery, operation time, operation site, preoperative underlying diseases (such as diabetes, hypertension, pulmonary infection, etc.), preoperative disturbance of consciousness, postoperative use of hormones, operation type, as well as gender and age. And based on the analysis results, guidance on prevention and treatment strategies is provided from aspects such as nursing, nutritional support, and application of antibiotics. This process is not only cumbersome, but also because it is manually counted, it not only takes a long time, but also the accuracy cannot be guaranteed, so the progress is slow.

[0004] Therefore, there is an urgent need for a method and system that can efficiently process data, clean a large amount of useless data, and analyze various medical data to formulate a series of relatively reasonable and general prevention and treatment strategies for intracranial infection after craniotomy. Summary of the Invention

[0005] The purpose of the present invention is to provide a data collection and analysis system and method for patients with intracranial postoperative infections to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A data collection and analysis method for patients with intracranial postoperative infections, the method comprising:

[0008] Obtaining the medical record information of the patient and initially classifying the data in the medical record information according to a preset type;

[0009] Performing secondary classification on the initially classified data based on preset infection factors; the infection factors include patient's own factors, surgery-related factors, postoperative factors, and postoperative comprehensive factors;

[0010] Compare each category of data after secondary classification, and calculate the correlation between types;

[0011] Based on the calculated correlation, display the data of each type; The display methods include pie charts, bar charts, and multi-dimensional relationship tables.

[0012] As a further solution of the present invention: The step of obtaining the medical record information of the patient and initially classifying the data in the medical record information according to a preset type includes:

[0013] Receive the information sharing permission granted by the patient, and obtain the medical record information of the patient; The medical record information includes, but is not limited to, medical data obtained by means of CT, color Doppler ultrasound, urine test, and blood test;

[0014] Input the medical record information into a preset evaluation model to determine whether there is an infection;

[0015] Based on whether there is an infection, conduct a preliminary classification of the patient;

[0016] Among them, the evaluation model is a regression model based on a preset threshold. For any parameter, calculate the difference between it and the preset threshold of the parameter, and use the difference as the independent variable of the evaluation model. The output of the evaluation model is an infection score, which is used to characterize the degree of infection of the user. When the infection score is greater than the preset score threshold, it is determined that there is an infection.

[0017] As a further solution of the present invention: The step of secondary-classifying the data after initial classification based on preset infection factors includes:

[0018] Obtain the medical record information of the patients whose determination result is infection;

[0019] Compare the medical record information with a preset classification standard to determine the secondary classification result of the medical record information; The secondary classification is used to divide the patients into four categories of infection factors;

[0020] Among them, when obtaining the medical record information, perform repetitive recognition on the medical record information based on historical data. When the medical record information is duplicate data, eliminate the medical record information; When obtaining the medical record information, traverse each parameter in the medical record information. If there is a missing position, use the average value of the corresponding parameter as the replacement value and insert it into the medical record information.

[0021] As a further solution of the present invention: The step of comparing each category of data after secondary classification and calculating the correlation between types includes:

[0022] For each category of patients after secondary classification, read the patient information of each patient;

[0023] Compare the patient information of different categories of patients and calculate the correlation of each category of patients;

[0024] Among them, the calculation process of the relevance is as follows: for two classes to be compared, patients are sequentially selected from the first class of patients, and the similarity between the selected patients and each patient in the other class of patients is calculated. When the similarity reaches the preset threshold, they are regarded as relevant patients; the ratio of the number of relevant patients to the larger value of the total number of patients in the two classes of patients is calculated as the relevance.

[0025] As a further solution of the present invention: the multi-dimensional relationship table in the display party is as follows:

[0026] S = [s 111 , s 112 , ……, s ijk ;

[0027] s ijk = [q i , p j , r k ;

[0028] Q = [q 1 , q 2 , q 3 , ……, q i ;

[0029] P = [p 1 , p 2 , p 3 , ……, p j ;

[0030] R = [r 1 , r 2 , r 3 , ……, r k ;

[0031] Among them, the elements of the array Q are q i , representing the number of patients with the i-th factor of the Q-type factor, the elements of the array P are p j , representing the number of patients with the j-th factor of the P-type factor, the elements of the array R are r k , representing the number of patients with the k-th factor of the R-type factor; the elements of the array S are s ijk , representing being simultaneously related to q i , p j and r k .

[0032] The technical solution of the present invention also provides a data collection and analysis system for patients with intracranial postoperative infections, and the system includes:

[0033] A preliminary classification module, configured to obtain the medical record information of patients and perform preliminary classification on the data in the medical record information according to the preset types;

[0034] The secondary classification module is used to perform secondary classification on the data after primary classification based on preset infection factors; the infection factors include patient's own factors, surgery-related factors, postoperative factors, and postoperative comprehensive factors;

[0035] The relevance calculation module is used to compare each type of data after secondary classification and calculate the relevance between types;

[0036] The visual display module is used to display various types of data based on the calculated relevance; the display methods include pie charts, bar charts, and multi-dimensional relationship tables.

[0037] As a further solution of the present invention: the primary classification module includes:

[0038] The first information acquisition unit is used to receive the information sharing permission granted by the patient and acquire the patient's medical record information; the medical record information includes, but is not limited to, medical data obtained by means of CT, color Doppler ultrasound, urine test, and blood test;

[0039] The information evaluation unit is used to input the medical record information into a preset evaluation model to determine whether there is an infection;

[0040] The first execution unit is used to perform a preliminary classification on the patient according to whether there is an infection;

[0041] Among them, the evaluation model is a regression model based on a preset threshold. For any parameter, calculate the difference between it and the preset threshold of the parameter, and use the difference as the independent variable of the evaluation model. The output of the evaluation model is an infection score, which is used to characterize the infection degree of the user. When the infection score is greater than the preset score threshold, it is determined that there is an infection.

[0042] As a further solution of the present invention: the secondary classification module includes:

[0043] The second information acquisition unit is used to acquire the medical record information of the patient whose determination result is infection;

[0044] The second execution unit is used to compare the medical record information with the preset classification criteria to determine the secondary classification result of the medical record information; the secondary classification is used to divide the patient into four categories of infection factors;

[0045] Among them, when acquiring the medical record information, perform repetitive recognition on the medical record information based on historical data. When the medical record information is duplicate data, eliminate the medical record information; when acquiring the medical record information, traverse each parameter in the medical record information. If there is a missing position, use the average value of the corresponding parameter as a substitute value and insert it into the medical record information.

[0046] As a further solution of the present invention: the relevance calculation module includes:

[0047] A patient information reading unit for reading the patient information of each patient in each category after secondary classification;

[0048] A calculation unit for comparing the patient information of different categories of patients and calculating the relevance of each category of patients;

[0049] Among them, the calculation process of the relevance is as follows: for two categories to be compared, select patients in the first category of patients in turn, calculate the similarity between the selected patients and each patient in the other category of patients, and when the similarity reaches the preset threshold, regard them as relevant patients; calculate the ratio of the number of relevant patients to the larger value of the total number of patients in the two categories of patients as the relevance.

[0050] As a further solution of the present invention: The multi-dimensional relationship table in the display party is as follows:

[0051] S = [s 111 ,s 112 ,……,s ijk ;

[0052] s ijk = [q i ,p j ,r k ;

[0053] Q = [q 1 ,q 2 ,q 3 ,……,q i ;

[0054] P = [p 1 ,p 2 ,p 3 ,……,p j ;

[0055] R = [r 1 ,r 2 ,r 3 ,……,r k ;

[0056] Among them, the elements of the array Q are q i , representing the number of patients with the i-th factor of the Q-type factor, the elements of the array P are p j , representing the number of patients with the j-th factor of the P-type factor, the elements of the array R are r k , representing the number of patients with the k-th factor of the R-type factor; the elements of the array S are s ijk , representing those who are simultaneously related to q i , p j and r k .

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention designs a system and method for data collection and analysis for patients with intracranial postoperative infections. Through efficient data processing means, it solves the problem that the data of patients with intracranial postoperative infections is mottled and complex and difficult to accurately locate. The present invention provides strong information for clinical causal relationships, establishes clear causal relationships among various factors, and through the present invention, the potential risks of a certain factor or certain factors developing into diseases can be directly exposed. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0059] Figure 1 It is a flowchart of a data collection and analysis method for patients with intracranial postoperative infections.

[0060] Figure 2 It is an architecture diagram of a data collection and analysis system for patients with intracranial postoperative infections. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the following further details the present invention with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0062] Figure 1 It is a flowchart of a data collection and analysis method for patients with intracranial postoperative infections. In an embodiment of the present invention, a data collection and analysis method for patients with intracranial postoperative infections includes:

[0063] Step S100: Obtain the medical record information of the patient, and perform primary classification on the data in the medical record information according to a preset type;

[0064] Step S200: Perform secondary classification on the data after primary classification based on preset infection factors; the infection factors include patient's own factors, surgery-related factors, postoperative factors, and postoperative comprehensive factors;

[0065] Step S300: Compare each type of data after secondary classification, and calculate the correlation degree between types;

[0066] Step S400: Display each type of data based on the calculated correlation degree; the display methods include pie charts, bar charts, and multi-dimensional relationship tables.

[0067] In an example of the technical solution of the present invention, the intelligent device applying this method is divided into three modules, including a data collection module, a data processing module, and a data visualization module. The data collection module is used to obtain the medical data of the patient from the medical means experienced by the patient, including but not limited to CT, color Doppler ultrasound, urine test, blood test and other means; the data processing module includes a data cleaning sub-module and a data analysis module, which are used to clean the data collected by the data collection module and perform modeling analysis on the useful data. The data visualization module is used to display the data analysis results generated by the data analysis sub-module in the form of a pie chart, a bar chart, etc.

[0068] As a preferred embodiment of the technical solution of the present invention, the steps of obtaining the medical record information of the patient and performing preliminary classification on the data in the medical record information according to a preset type include:

[0069] Receive the information sharing permission granted by the patient and obtain the medical record information of the patient; the medical record information includes but is not limited to medical data obtained by means of CT, color Doppler ultrasound, urine test, and blood test;

[0070] Input the medical record information into a preset evaluation model to determine whether there is an infection;

[0071] Perform preliminary classification on the patient according to whether there is an infection.

[0072] At the beginning, the patient is registered as an intracranial surgery patient, and the registration process is as follows: The medical staff enters the hospital's patient medical record information database, adds the patient's medical record information as an intracranial surgery patient, and pushes the patient's medical record information to the data collection module. The premise of this process is that the patient must first grant the information sharing permission. If the patient does not grant the permission, the execution entity of this method cannot obtain the patient's medical record information.

[0073] After obtaining the medical record information, input the medical record information into a preset evaluation model to determine whether there is an infection, and then complete the preliminary classification. The preliminary classification is of two types, one is already infected, and the other is not infected.

[0074] Among them, the evaluation model is a regression model based on a preset threshold. For any parameter, calculate the difference between it and the preset threshold of the parameter, and use the difference as the independent variable of the evaluation model. The output of the evaluation model is the infection score, which is used to characterize the infection degree of the user. When the infection score is greater than the preset score threshold, it is determined that there is an infection.

[0075] The operation process of the evaluation model is as follows: for any parameter in the medical record information, it is compared with the preset standard data, and the difference is calculated. The more it meets the standard data, the smaller the difference. The difference reflects whether the parameter is standard. Based on this principle, taking the difference as the independent variable, the influence of this parameter on the infection determination process can be calculated, which is represented by the infection score. Different parameters may correspond to different authorities, which are also preset by the designers.

[0076] As a preferred embodiment of the technical solution of the present invention, the step of performing secondary classification on the data after primary classification based on preset infection factors includes:

[0077] Obtain the medical record information of the patients whose determination result is infection;

[0078] Compare the medical record information with the preset classification criteria to determine the secondary classification result of the medical record information; the secondary classification is used to divide the patients into four categories of infection factors.

[0079] In the technical solution of the present invention, for the infected patients, the data is divided into four major categories according to the common postoperative infection factors in clinical practice: including the first category: patient's own factors, that is, whether the age is greater than 70 years old, whether suffering from diabetes or poor blood glucose control, whether the immunity is low, whether the GCS is less than 9, and whether the primary injury is severe. The second category: surgery-related factors, that is, whether the wound is contaminated, whether it is a sub-tentorial surgery, whether the operation time is greater than four hours, whether ≥2 craniotomies are performed, and whether there is a large amount of blood loss or implants during the operation. The third category: postoperative factors, that is, whether the placement time of the ventricular or lumbar cistern drainage tube is greater than 5 days, whether cerebrospinal fluid specimens are frequently collected during the indwelling of the drainage tube, whether cerebrospinal fluid leakage occurs at the drainage tube orifice, whether bleeding occurs in the puncture tract, and whether bilateral EDV is performed simultaneously and other factors related to the drainage tube, whether a wound occurs after the operation, and whether subcutaneous fluid accumulation occurs at the surgical incision. The fourth category: other factors, that is, whether chemotherapy and immunosuppressive therapy have been received recently, whether a ventilator has been used for a long time after the operation, whether total parenteral nutrition has been received for a long time after the operation and combined with severe hypoproteinemia, whether a large dose of glucocorticoid has been used for a long time after the operation, whether neurosurgical operations have been performed in the intensive care unit, and whether there is improper wound care.

[0080] Further, when obtaining the medical record information, repetitive identification of the medical record information is performed based on historical data. When the medical record information is duplicate data, the medical record information is excluded; when obtaining the medical record information, each parameter in the medical record information is traversed. If there is a missing position, the average value of the corresponding parameter is used as the replacement value and inserted into the medical record information.

[0081] The above content is the data cleaning process, including: removing duplicate records and irrelevant data, filling in missing values (such as through interpolation or using the mean to replace), data standardization (ensuring that data formats from different sources are consistent for easy comparison and analysis), and outlier detection and handling (to ensure data accuracy).

[0082] As a preferred embodiment of the technical solution of the present invention, the step of comparing each type of data after secondary classification and calculating the correlation degree between types includes:

[0083] For each type of patient after secondary classification, read the patient information of each patient;

[0084] Compare the patient information of different types of patients and calculate the correlation degree of each type of patient.

[0085] Statistical analysis is performed on various types of data, and the number of patients in relevant categories is classified, that is, the postoperative infection situation of patients meeting each category is statistically analyzed, and the correlation between the four major categories is analyzed based on the consistency of the number of patients in each category and the patient identity. For example, if the number of patients in category one and category two is similar and their identities are similar, that is, a large number of patients in category one are also patients in category two, and only a small number are patients in category three, it can be considered that the correlation between factor one and factor two is large, and the correlation with factor three is small.

[0086] Among them, the calculation process of the correlation degree is as follows: for two classes to be compared, patients are sequentially selected from the first class of patients, and the similarity between the selected patients and each patient in the other class is calculated. When the similarity reaches the preset threshold, they are regarded as relevant patients; calculate the ratio of the number of relevant patients to the larger value of the total number of patients in the two classes of patients as the correlation degree.

[0087] For two types of data, first take one type as the benchmark and analyze the data in it sequentially. For each data, query the closest data in the other type. The comparison process can use existing string comparison techniques and numerical comparison techniques (medical record information can be regarded as a set of strings and numerical values. The simplest way is to apply AI for recognition. Existing AI models have extremely strong comprehension ability, and it is very convenient to judge whether two medical record information are similar); query the similarity of the closest data. If the similarity is high enough, regard the two patients as relevant patients and calculate the ratio of the number of relevant patients to the maximum value of the total number of patients in the two classes of patients as the correlation degree.

[0088] It is worth mentioning that the comparison result between patients is similarity, and the comparison result between two types of patients is correlation degree. These are two different parameters.

[0089] As a preferred embodiment of the technical solution of the present invention, the display processes of the pie chart and the bar chart are conventional and will not be elaborated here. For the multi-dimensional relational table in the display party, the following is defined:

[0090] S = [s 111 , s 112 , ……, s ijk ;

[0091] s ijk = [q i , p j , r k ;

[0092] Q = [q 1 , q 2 , q 3 , ……, q i ;

[0093] P = [p 1 , p 2 , p 3 , ……, p j ;

[0094] R = [r 1 , r 2 , r 3 , ……, r k ;

[0095] Among them, the elements of the array Q are q i , representing the number of patients with the i-th factor of the Q-type factor, the elements of the array P are p j , representing the number of patients with the j-th factor of the P-type factor, and the elements of the array R are r k , representing the number of patients with the k-th factor of the R-type factor; the elements of the array S are s ijk , representing those that are simultaneously related to q i , p j and r k .

[0096] As a preferred embodiment of the technical solution of the present invention, the present invention also discloses a data collection and analysis system for patients after intracranial surgery. The system includes:

[0097] A primary classification module, configured to obtain the medical record information of the patient and perform primary classification on the data in the medical record information according to a preset type;

[0098] A secondary classification module, configured to perform secondary classification on the data after primary classification based on preset infection factors; the infection factors include patient's own factors, surgery-related factors, postoperative factors, and postoperative comprehensive factors;

[0099] A relevance calculation module for comparing each type of data after secondary classification and calculating the relevance between types;

[0100] A visualization display module for displaying each type of data based on the calculated relevance; The display methods include pie charts, bar charts, and multi-dimensional relationship tables.

[0101] Furthermore, the primary classification module includes:

[0102] A first information acquisition unit for receiving the information sharing permission granted by the patient and acquiring the patient's medical record information; The medical record information includes, but is not limited to, medical data obtained by means of CT, color Doppler ultrasound, urine test, and blood test;

[0103] An information evaluation unit for inputting the medical record information into a preset evaluation model to determine whether there is an infection;

[0104] A first execution unit for preliminarily classifying the patient according to whether there is an infection;

[0105] Among them, the evaluation model is a regression model based on a preset threshold. For any parameter, calculate the difference between it and the preset threshold of the parameter, and use the difference as the independent variable of the evaluation model. The output of the evaluation model is an infection score, which is used to characterize the user's infection degree. When the infection score is greater than the preset score threshold, it is determined that there is an infection.

[0106] Specifically, the secondary classification module includes:

[0107] A second information acquisition unit for acquiring the medical record information of patients whose determination result is infection;

[0108] A second execution unit for comparing the medical record information with a preset classification standard to determine the secondary classification result of the medical record information; The secondary classification is used to divide the patients into four categories of infection factors;

[0109] Among them, when acquiring the medical record information, repetitive recognition of the medical record information is performed based on historical data. When the medical record information is duplicate data, the medical record information is excluded; When acquiring the medical record information, each parameter in the medical record information is traversed. If there is a missing position, the average value of the corresponding parameter is used as a substitute value and inserted into the medical record information.

[0110] Even further, the relevance calculation module includes:

[0111] A patient information reading unit for reading the patient information of each patient after secondary classification;

[0112] A calculation unit for comparing the patient information of different types of patients and calculating the relevance of each type of patient;

[0113] Among them, the calculation process of the relevance is as follows: for two classes to be compared, select patients in the first class of patients in sequence, calculate the similarity between the selected patients and each patient in the other class of patients, and when the similarity reaches the preset threshold, regard them as relevant patients; calculate the ratio of the number of relevant patients to the larger value of the total number of patients in the two classes of patients as the relevance.

[0114] In addition, the multi-dimensional relationship table in the display party is as follows:

[0115] S = [s 111 , s 112 , ……, s ijk ;

[0116] s ijk = [q i , p j , r k ;

[0117] Q = [q 1 , q 2 , q 3 , ……, q i ;

[0118] P = [p 1 , p 2 , p 3 , ……, p j ;

[0119] R = [r 1 , r 2 , r 3 , ……, r k ;

[0120] Among them, the elements of the array Q are q i , representing the number of patients with the i-th factor of the Q-type factor, the elements of the array P are p j , representing the number of patients with the j-th factor of the P-type factor, the elements of the array R are r k , representing the number of patients with the k-th factor of the R-type factor; the elements of the array S are s ijk , representing those who are simultaneously related to q i , p j and r k .

[0121] Please refer to Figure 2, in an example of the technical solution of the present invention, the system is divided into three modules, including a data collection module, a data processing module, and a data visualization module. The data collection module is used to obtain the medical data of the patient from the medical means experienced by the patient, including but not limited to CT, color Doppler ultrasound, urine test, blood test, etc.; the data processing module includes a data cleaning sub-module and a data analysis module, which are used to clean the data collected by the data collection module and perform modeling analysis on the useful data. The data visualization module is used to display the data analysis results generated by the data analysis sub-module in the form of a pie chart, a bar chart, etc.

[0122] The working process of the data collection module is as follows: Step 201: The patient registers as an intracranial surgery patient, and the registration process is as follows: The medical staff enters the hospital's patient medical record information database, adds the patient's medical record information as an intracranial surgery patient, and pushes the patient's medical record information to the data collection module.

[0123] After receiving the patient's medical record information, the data collection module queries various data of the patient through the patient's information and makes a preliminary classification of whether there is postoperative infection.

[0124] Send the classified data to the data processing module.

[0125] Furthermore, the data processing module contains a data cleaning sub-module, and the working process of the data cleaning sub-module is as follows:

[0126] The data cleaning sub-module receives the data from the data collection module;

[0127] The data cleaning module classifies the patient's data and divides the data into four categories according to the common postoperative infection factors in clinical practice: including the first category: patient's own factors, that is, whether the age is greater than 70 years old, whether suffering from diabetes or poor blood sugar control, whether the immunity is low, whether the GCS is less than 9, and whether the primary injury is severe. The second category: surgery-related factors, that is, whether the wound is contaminated, whether it is a sub-tentorial surgery, whether the operation time is greater than four hours, whether ≥2 craniotomies are received, and whether there is massive blood loss or implants during the operation. The third category: postoperative factors, that is, whether the placement time of the ventricular or lumbar cistern drainage tube is greater than 5 days, whether cerebrospinal fluid specimens are frequently collected during the indwelling of the drainage tube, whether there is cerebrospinal fluid leakage at the drainage tube orifice, whether there is bleeding in the puncture tract and whether bilateral EDV is performed simultaneously, whether there is a wound after the operation, and whether there is subcutaneous fluid accumulation at the surgical incision. The fourth category: other factors, that is, whether chemotherapy and immunosuppressive therapy have been received recently, whether the ventilator has been used for a long time after the operation, whether total parenteral nutrition has been received for a long time after the operation and whether severe hypoproteinemia is combined, whether high-dose glucocorticoids have been used for a long time after the operation, whether neurosurgical operations have been performed in the intensive care unit, and whether there is improper wound care.

[0128] According to the data classification results in the above content, classify and fill the data from the data collection module into four major categories, discard other irrelevant data, and export the results of data classification and cleaning.

[0129] In addition, the data processing module also contains a data analysis sub-module, and the working process of the data analysis sub-module is as follows:

[0130] The data analysis sub-module receives the data classification and cleaning results exported by the data cleaning sub-module;

[0131] The data analysis module conducts statistics on various types of data, classifies the number of patients in relevant categories, that is, conducts statistics on the postoperative infection situation of patients meeting each category, and analyzes the correlation between the four major categories based on the number of patients in each category and the consistency of patient identities. For example, if the number of patients in category one and category two is similar and their identities are similar, that is, a large number of patients in category one are also patients in category two, and only a small number are patients in category three, it can be considered that the correlation between factor one and factor two is large, and the correlation with factor three is small.

[0132] The data visualization module obtains the data analysis structure from the data analysis sub-module and displays it in the form of a pie chart, a bar chart, and a multi-dimensional relationship table. The multi-dimensional relationship table is used to represent the correlation relationship of multiple types of factors. Taking the three-dimensional relationship as an example, the three-dimensional relationship is as follows:

[0133] S = [s 111 , s 112 , ……, s ijk ;

[0134] s ijk = [q i , p j , r k ;

[0135] Q = [q 1 , q 2 , q 3 , ……, q i ;

[0136] P = [p 1 , p 2 , p 3 , ……, p j ;

[0137] R = [r 1 , r 2 , r 3 , ……, r k ;

[0138] Among them, the elements of the array Q are q i, the number of patients with the i-th factor representing factor Q, and the elements of array P are p j , the number of patients with the j-th factor representing factor P, and the elements of array R are r k , the number of patients with the k-th factor representing factor R; the elements of array S are s ijk , representing simultaneously with q i 、p j and r k .

[0139] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A data collection and analysis method for patients with postoperative intracranial infection, characterized in that: The method comprises: Obtain the patient's medical record information and perform preliminary classification of the data in the medical record information according to the preset types; The data after the initial classification are secondary classified based on the preset infection factors; the infection factors include patient factors, surgery-related factors, postoperative factors and postoperative comprehensive factors; Compare each type of data after secondary classification and calculate the correlation between types; Various types of data are displayed based on the calculated correlation; the display methods include pie charts, bar charts and multidimensional relationship tables.

2. The data collection and analysis method for patients with intracranial postoperative infection according to claim 1 is characterized in that: The step of obtaining the patient's medical record information and initially classifying the data in the medical record information according to a preset type includes: Receive information sharing permissions granted by the patient and obtain the patient's medical records; the medical records include but are not limited to medical data obtained by CT, color ultrasound, urine test and blood test; Input medical record information into the preset assessment model to determine whether there is infection; Patients were initially classified according to whether they were infected or not; Among them, the evaluation model is a regression model based on a preset threshold. For any parameter, the difference between it and the preset threshold of the parameter is calculated, and the difference is used as the independent variable of the evaluation model. The output of the evaluation model is an infection score, which is used to characterize the user's infection level. When the infection score is greater than the preset score threshold, it is determined to be infected.

3. The data collection and analysis method for patients with intracranial postoperative infection according to claim 1 is characterized in that: The step of performing secondary classification on the data after the primary classification based on the preset infection factors includes: Obtain medical records of patients who are diagnosed with infection; Compare the medical record information with the preset classification criteria to determine the secondary classification results of the medical record information; the secondary classification is used to classify patients into four categories of infection factors; Among them, when obtaining medical record information, the medical record information is duplicated and identified based on historical data. When the medical record information is duplicate data, the medical record information is eliminated. When obtaining medical record information, the various parameters in the medical record information are traversed. If there is a missing position, the average value of the corresponding parameter is used as a replacement value to insert the medical record information.

4. The data collection and analysis method for patients with intracranial postoperative infection according to claim 1 is characterized in that: The step of comparing each type of data after secondary classification and calculating the correlation between types includes: For each type of patient after secondary classification, read the patient information of each patient; Compare the patient information of different categories of patients and calculate the relevance of each category of patients; The calculation process of the correlation is as follows: for the two classes that need to be compared, select patients in the first class in turn, calculate the similarity between the selected patients and each patient in the other class, and when the similarity reaches a preset threshold, regard them as related patients; calculate the ratio of the number of related patients to the larger value of the total number of patients in the two classes as the correlation.

5. The data collection and analysis method for patients with intracranial postoperative infection according to claim 1 is characterized in that: The multidimensional relationship table in the display side is as follows: S=[s 111 ,s 112 ,……,s ijk ]; s ijk =[q i ,p j ,r k ]; Q=[q1,q2,q3,……,q i ]; P=[p1,p2,p3,……,p j ]; R＝[r1，r2，r3，……，r k ]； Among them, the elements of array Q are q i , represents the number of patients with factor i of factor Q, and the element of array P is p j , represents the number of patients with factor j of factor P, and the element of array R is r k , represents the number of patients with k factors of R type factors; the element of array S is s ijk , indicating that at the same time with q i 、p j and r k .

6. A data collection and analysis system for patients with postoperative intracranial infection, characterized in that: The system comprises: The initial classification module is used to obtain the patient's medical record information and perform initial classification of the data in the medical record information according to the preset type; A secondary classification module is used to perform secondary classification on the data after the primary classification based on preset infection factors; the infection factors include patient factors, surgery-related factors, postoperative factors and postoperative comprehensive factors; The correlation calculation module is used to compare each type of data after secondary classification and calculate the correlation between types; The visualization display module is used to display various types of data based on the calculated correlation; the display methods include pie charts, bar charts and multidimensional relationship tables.

7. The data collection and analysis system for patients with intracranial postoperative infection according to claim 6 is characterized in that: The initial classification module comprises: The first information acquisition unit is used to receive the information sharing authority granted by the patient and acquire the patient's medical record information; the medical record information includes but is not limited to medical data acquired by means of CT, color ultrasound, urine test and blood test; An information evaluation unit, used to input medical record information into a preset evaluation model to determine whether there is infection; A first execution unit is used to preliminarily classify patients according to whether they are infected; Among them, the evaluation model is a regression model based on a preset threshold. For any parameter, the difference between it and the preset threshold of the parameter is calculated, and the difference is used as the independent variable of the evaluation model. The output of the evaluation model is an infection score, which is used to characterize the user's infection level. When the infection score is greater than the preset score threshold, it is determined to be infected.

8. The data collection and analysis system for patients with intracranial postoperative infection according to claim 6 is characterized in that: The secondary classification module comprises: A second information acquisition unit is used to acquire medical record information of patients whose results are determined to be infected; The second execution unit is used to compare the medical record information with the preset classification standard to determine the secondary classification result of the medical record information; the secondary classification is used to classify the patients into four categories of infection factors; Among them, when obtaining medical record information, the medical record information is duplicated and identified based on historical data. When the medical record information is duplicate data, the medical record information is eliminated. When obtaining medical record information, the various parameters in the medical record information are traversed. If there is a missing position, the average value of the corresponding parameter is used as a replacement value to insert the medical record information.

9. The data collection and analysis system for patients with intracranial postoperative infection according to claim 6, characterized in that: The correlation calculation module comprises: A patient information reading unit, used for reading the patient information of each patient after the secondary classification; a calculation unit, for comparing patient information of different categories of patients and calculating the relevance of each category of patients; The calculation process of the correlation is as follows: for the two classes that need to be compared, select patients in the first class in turn, calculate the similarity between the selected patients and each patient in the other class, and when the similarity reaches a preset threshold, regard them as related patients; calculate the ratio of the number of related patients to the larger value of the total number of patients in the two classes as the correlation.

10. The data collection and analysis system for patients with intracranial postoperative infection according to claim 6, characterized in that: The multidimensional relationship table in the display side is as follows: S=[s 111 ,s 112 ,……,s ijk ]; s ijk =[q i ,p j ,r k ]; Q=[q1,q2,q3,……,q i ]; P=[p1,p2,p3,……,p j ]; R＝[r1，r2，r3，……，r k ]； Among them, the elements of array Q are q i , represents the number of patients with factor i of factor Q, and the element of array P is p j , represents the number of patients with factor j of factor P, and the element of array R is r k , represents the number of patients with k factors of R type factors; the element of array S is s ijk , indicating that at the same time with q i 、p j and r k .

Images