Method and system for constructing a control model after aortic dissection surgery based on big data

Abstract

The present invention discloses a method and system for constructing a control model after aortic dissection surgery based on big data, and relates to the technical field of model construction, including: obtaining a control case database; establishing a basic analysis model, and performing a preliminary construction to obtain a preliminary analysis model; using a random analysis method to perform a primary construction and a secondary construction on the preliminary analysis model, and obtain a case construction model; analyzing the case construction model; the present invention is used to solve the problem that the existing analysis method for aortic dissection surgery has a relatively single prediction direction, and is unable to comprehensively screen abnormal items that may exist in patients after aortic dissection surgery through existing control cases, resulting in a large amount of time spent on a comprehensive examination of the patient or the examination is not comprehensive enough, causing delays in the patient's condition.

Description

Technical Field

[0001] The present invention relates to the technical field of model construction, and in particular to a method and system for constructing a control model after aortic dissection surgery based on big data. Background Art

[0002] Aortic dissection surgery is a surgical method for treating aortic dissection, which repairs damaged blood vessels by placing a stent on the aorta or performing open-chest surgery. Aortic dissection surgery is currently one of the main methods for treating aortic dissection. Surgery can repair damaged blood vessels, prevent hematoma expansion and reduce the occurrence of complications. Although aortic dissection surgery is a relatively safe and effective treatment method, there are still certain risks. Common risks include bleeding, infection, thrombosis, etc. Therefore, after aortic dissection surgery, postoperative treatment of aortic dissection is more important.

[0003] Existing analysis methods for aortic dissection surgery are usually improvements in risk prediction after aortic dissection surgery. For example, by performing vector analysis on preoperative three-dimensional angiography images, corresponding feature information is obtained to predict the patient's postoperative risk based on the feature information. Although this improved method can predict the risk probability of postoperative complications, the prediction direction and prediction angle are relatively single, and only the risk probability can be predicted. It is impossible to comprehensively screen the abnormal items that may exist in patients after aortic dissection surgery through existing control cases, resulting in a lot of time spent on comprehensive examinations of patients or incomplete examinations, causing delays in patients' condition. For example, in a patent application with publication number CN110742633A, a method, device and method for predicting risk after type B aortic dissection surgery are disclosed. Electronic equipment, this solution is to obtain multiple preoperative three-dimensional angiography images, pre-process the three-dimensional angiography images of each case to obtain a data set to be processed, and obtain risk factor variables, and construct a risk prediction model based on the risk factor variables and preset clinical lesion types; other analysis methods for postoperative aortic dissection are usually improvements in nursing devices. When applied to patients, this improvement method is usually used to solve problems in cardiopulmonary rehabilitation. There is still a relatively single prediction direction, and it is impossible to comprehensively screen the abnormal items that may exist in patients after aortic dissection surgery through existing control cases, resulting in a lot of time spent on a comprehensive examination of the patient or the examination is not comprehensive enough, causing delays in the patient's condition. In view of this, it is necessary to improve the existing analysis methods for postoperative aortic dissection. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent. By proposing a method and system for constructing a control model for postoperative aortic dissection based on big data, the present invention is used to solve the problem that the existing analysis methods for postoperative aortic dissection have a relatively single prediction direction and are unable to conduct a comprehensive screening of abnormal items that may exist in patients after postoperative aortic dissection through existing control cases, resulting in a lot of time being spent on a comprehensive examination of the patient or the examination is not comprehensive enough, causing delays in the patient's condition.

[0005] To achieve the above objectives, in a first aspect, the present application provides a method for constructing a control case analysis model for aortic dissection surgery based on big data, comprising the following steps:

[0006] Based on big data, multiple postoperative cases of aortic dissection were obtained and stored in a control case database; a basic analysis model was established, and the basic analysis model was preliminarily constructed based on the postoperative cases of aortic dissection to obtain a preliminary analysis model;

[0007] Based on the control case database, the preliminary analysis model was constructed once and twice using the random analysis method, and the case construction model was obtained;

[0008] The model is constructed based on the case to analyze the postoperative cases of aortic dissection to be analyzed, and the data to be analyzed is obtained based on the analysis results.

[0009] Furthermore, multiple aortic dissection postoperative cases were obtained based on big data and stored in the control case database; the basic analysis model was established including:

[0010] Based on big data, multiple aortic dissection postoperative cases were obtained and recorded as dissection cases YA 1 To mezzanine case YA n ; All interlayer cases YA are stored in the control case database;

[0011] A basic analysis model is established, wherein the input end of the basic analysis model is two neurons for inputting control cases and cases to be analyzed; the output end is multiple neurons for outputting difference items obtained by analysis; the function within the basic analysis model is used to perform difference analysis on the control cases and cases to be analyzed received by the two neurons at the input end, and obtain difference items, wherein the control cases are cases in which the values ​​of various indicators are within the normal range, and the difference items include heart failure difference items and conventional difference items.

[0012] Furthermore, the basic analysis model was initially constructed based on the postoperative case of aortic dissection, and the preliminary analysis model included:

[0013] The difference analysis includes: obtaining multiple project names in the postoperative cases of aortic dissection, recording them as dissection case names; extracting the dissection case names in the cases to be analyzed based on text extraction, and recording the content contained in each dissection case name as case analysis items;

[0014] For any case analysis item: record the interlayer case name of the case analysis item as the name to be compared, obtain the name to be compared in the comparison case based on text extraction, and record the content contained in the name to be compared in the comparison case as the case comparison item.

[0015] Furthermore, the difference analysis also includes:

[0016] Based on text extraction, the left ventricular ejection fraction is recorded as the main reference point, and the natriuretic peptide is recorded as the auxiliary reference point; when there is a main reference point in the case analysis item, the value corresponding to the main reference point is recorded as the ejection fraction. When the ejection fraction is greater than or equal to 50%, the value corresponding to the auxiliary reference point is obtained. When BNP>35 and NT-proBNP>125, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 50% and greater than or equal to 40%, the medical history in the case analysis item is extracted, and when there is heart disease in the medical history and BNP>35 and NT-proBNP>125 in the values ​​corresponding to the auxiliary reference point, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 40%, the case analysis item is recorded as a heart failure screening item;

[0017] When a case analysis item is not recorded as a heart failure screening item, the case analysis item is recorded as a regular screening item.

[0018] Furthermore, the difference analysis also includes:

[0019] When the case analysis item is recorded as a heart failure screening item, multiple heart failure indicators are obtained based on big data, and the heart failure indicators in the case analysis item and the case control item are obtained respectively; for any heart failure indicator: multiple determination intervals corresponding to the heart failure indicator are obtained based on the determination criteria of the heart failure indicator, when the determination interval of the heart failure indicator in the case analysis item is the same as the determination interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a same-area indicator; when the determination interval of the heart failure indicator in the case analysis item is different from the determination interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a different-area indicator;

[0020] When all heart failure indicators are same-region indicators, the case analysis item is recorded as the same-control item; when any heart failure indicator is a different-region indicator, the case analysis item is recorded as the heart failure difference item.

[0021] Furthermore, the difference analysis also includes:

[0022] When the case analysis item is recorded as a regular item to be screened, the name of each data in the case analysis item is obtained based on text extraction and recorded as the name to be compared; for any name to be compared, the normal interval of the data corresponding to the name to be compared is obtained based on big data and recorded as the normal data interval. When the data corresponding to the name to be compared is within the normal data interval, the name to be compared is recorded as a normal name; when the data corresponding to the name to be compared is outside the normal data interval, the name to be compared is recorded as an abnormal name;

[0023] When all the names to be compared in the case analysis item are normal names, the case analysis item is recorded as a same-control item; when any name to be compared is an abnormal name, the case analysis item is recorded as a regular difference item.

[0024] Furthermore, the basic analysis model was initially constructed based on the postoperative case of aortic dissection, and the preliminary analysis model also includes:

[0025] All dissection cases YA in the control case database are processed as cases to be analyzed in the basic analysis model in turn, and the heart failure difference items and conventional difference items corresponding to each dissection case YA are obtained; all heart failure difference items and conventional difference items are recorded using the basic analysis model, and the basic analysis model at this time is recorded as the preliminary analysis model.

[0026] Furthermore, the stochastic analysis method includes:

[0027] Randomly obtain control case database The mezzanine cases YA are recorded as first-construction cases, and the mezzanine cases YA that are not recorded as first-construction cases are recorded as second-construction cases;

[0028] A single construction is as follows: for any single construction case α, obtain all single construction cases in which the names of the heart failure difference items in the single construction cases are different from the names of the heart failure difference items in the single construction case α, and record them as single comparison cases;

[0029] The name of the heart failure difference item of a comparison case is recorded as a comparison item, and the name of all data in the comparison item is recorded as a comparison name YB 1 To the first comparison name YB c ; For a comparison name YB 1 To the first comparison name YB c Any one of the comparison names YB v , based on big data to obtain a comparison name YB v In all corresponding data intervals, the name YB is compared once v The corresponding data interval is recorded as a comparison interval YQ v; When a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison name YB is in the comparison item in the construction case α v Recorded as the normal comparison name; when a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison item in a construction case α is outside, the comparison name YB v Recorded as the name of the difference comparison;

[0030] When any one of the primary comparison names corresponding to a construction case α is recorded as a difference comparison name, the primary comparison item of the construction case α is recorded as a difference item;

[0031] The secondary construction is as follows: for any secondary construction case β, obtain the secondary construction cases whose names of the conventional difference items in all secondary construction cases are different from the names of the conventional difference items in the secondary construction case β, and record them as secondary comparison cases;

[0032] The name of the regular difference item of the secondary comparison case is recorded as the secondary comparison item, and the name of all data in the secondary comparison item is recorded as the secondary comparison name EB 1 To the secondary alignment name EB p ; For the secondary alignment name EB 1 To the secondary alignment name EB p Any secondary alignment name EB u , based on big data to obtain the secondary comparison name EB u In all corresponding data intervals, the secondary comparison name is YB u The corresponding data interval is recorded as the secondary comparison interval YQ u ; When the secondary alignment name EB in the secondary alignment item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQ u When the secondary alignment name EB in the secondary alignment item in the secondary construction case β is u Recorded as the normal comparison name; when the secondary comparison name EB in the secondary comparison item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQ u When the secondary comparison name EB in the secondary comparison item in the secondary construction case β is u Recorded as the name of the difference comparison;

[0033] When any secondary alignment name corresponding to the secondary construction case β is recorded as a difference alignment name, the secondary alignment item of the secondary construction case β is recorded as a difference item;

[0034] The preliminary analysis model after the first construction and the second construction is recorded as the case construction model.

[0035] Furthermore, the model is constructed based on the case to be analyzed for the aortic dissection postoperative case, and the data to be analyzed are obtained based on the analysis results, including:

[0036] When performing a comparative analysis on postoperative cases of aortic dissection, the postoperative cases of aortic dissection to be analyzed are recorded as cases to be placed, and the cases to be placed are input into the neurons corresponding to the cases to be analyzed at the input end of the case construction model, and the data in the difference items of the output of the case construction model are obtained as data to be analyzed, and the data to be analyzed are reported to the staff.

[0037] In the second aspect, the present application also provides a system for constructing a model for postoperative control case analysis of aortic dissection based on big data, including a preliminary model building module, a model construction module, and a model application module;

[0038] The preliminary model building module is used to obtain multiple postoperative cases of aortic dissection based on big data and store them in the control case database; establish a basic analysis model, and preliminarily construct the basic analysis model based on the postoperative cases of aortic dissection to obtain a preliminary analysis model;

[0039] The model building module is used to build the preliminary analysis model once and twice based on the control case database using the random analysis method, and obtain the case building model;

[0040] The model application module is used to build a model based on the case to analyze the postoperative cases of aortic dissection to be analyzed, and obtain the data to be analyzed based on the analysis results.

[0041] Beneficial effects of the present invention: the present application first obtains a plurality of aortic dissection postoperative cases based on big data and stores them in a control case database; establishes a basic analysis model, and preliminarily constructs the basic analysis model based on the aortic dissection postoperative cases to obtain a preliminary analysis model. The advantage of this is that by constructing the basic analysis model based on a plurality of aortic dissection postoperative cases, it can be ensured that the obtained preliminary analysis model can effectively analyze the case data of aortic dissection postoperative aspects, and prevents the use of a general analysis model, which leads to the inability to effectively screen the characteristic parameters that need to be paid attention to in aortic dissection postoperative aspects, resulting in the problem that the preliminary analysis model is not comprehensive enough in the analysis of the case in actual application;

[0042] The present application also uses a random analysis method to perform a primary and secondary construction of a preliminary analysis model based on a control case database, and obtains a case construction model; finally, the postoperative aortic dissection case to be analyzed is analyzed based on the case construction model, and the data to be analyzed is obtained based on the analysis results. The advantage of this is that, by performing a primary and secondary construction of the preliminary analysis model, it can ensure that the obtained case construction model can be combined with all the analyzed postoperative aortic dissection cases, so that the obtained difference items can be used to effectively analyze the case to be analyzed, which helps to reduce the time spent by staff on comprehensive examinations of patients, while ensuring that the staff's examinations are more comprehensive. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a functional block diagram of the system of the present invention;

[0044] Figure 2 is a flow chart of the steps of the method of the present invention;

[0045] Figure 3 It is a structural schematic diagram of the basic analysis model of the present invention;

[0046] Figure 4 It is a schematic structural diagram of the electronic device of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Example 1, first aspect, please refer to Figure 1 As shown, the present application provides a system for constructing a model for postoperative control case analysis of aortic dissection based on big data, including a preliminary model building module, a model construction module and a model application module;

[0049] The preliminary model building module is used to obtain multiple postoperative cases of aortic dissection based on big data and store them in the control case database; establish a basic analysis model, and preliminarily construct the basic analysis model based on the postoperative cases of aortic dissection to obtain a preliminary analysis model; the preliminary model building module includes a preliminary model acquisition unit, and the preliminary model acquisition unit is configured with a preliminary model acquisition strategy, and the preliminary model acquisition strategy includes:

[0050] Based on big data, multiple aortic dissection postoperative cases were obtained and recorded as dissection cases YA 1 To mezzanine case YA n; All dissection cases YA are stored in the control case database; in the specific implementation process, the aortic dissection postoperative cases obtained may include successful cases after aortic dissection and failed cases after aortic dissection. At the same time, when obtaining cases after aortic dissection, priority may be given to obtaining cases in which the equipment used in the postoperative process of aortic dissection is relatively similar to the equipment actually used, which is helpful to reduce data differences during data acquisition and improve data analysis efficiency;

[0051] A basic analysis model is established, wherein the input end of the basic analysis model is two neurons for inputting control cases and cases to be analyzed; the output end is multiple neurons for outputting difference items obtained by analysis; the function in the basic analysis model is used to perform difference analysis on the control cases and cases to be analyzed received by the two neurons at the input end to obtain difference items, wherein the control cases are cases in which the values ​​of various indicators are within the normal range, and the difference items include heart failure difference items and conventional difference items; in the specific implementation process, the structural diagram of the basic analysis model can be found in Figure 3 As shown;

[0052] The difference analysis includes: obtaining multiple project names in the postoperative cases of aortic dissection, recording them as dissection case names; extracting the dissection case names in the cases to be analyzed based on text extraction, and recording the content contained in each dissection case name as case analysis items;

[0053] In the specific implementation process, the project names in the aortic dissection postoperative cases may include items related to the physical examination of the patient before the operation, such as blood tests, cardiac ultrasound examinations, and coronary angiography, etc.; they may also include items for postoperative examinations of the patient, such as prothrombin time, electrocardiograms, and color Doppler ultrasound examinations, etc.; in actual implementation, the multiple project names obtained may be screened to ensure that the names of the dissection cases obtained are all items that can be used as references during application;

[0054] For any case analysis item: record the interlayer case name of the case analysis item as the name to be compared, obtain the name to be compared in the comparison case based on text extraction, and record the content contained in the name to be compared in the comparison case as the case comparison item;

[0055] Based on text extraction, the left ventricular ejection fraction is recorded as the main reference point, and the natriuretic peptide is recorded as the auxiliary reference point; when there is a main reference point in the case analysis item, the value corresponding to the main reference point is recorded as the ejection fraction. When the ejection fraction is greater than or equal to 50%, the value corresponding to the auxiliary reference point is obtained. When BNP>35 and NT-proBNP>125, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 50% and greater than or equal to 40%, the medical history in the case analysis item is extracted, and when there is heart disease in the medical history and BNP>35 and NT-proBNP>125 in the values ​​corresponding to the auxiliary reference point, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 40%, the case analysis item is recorded as a heart failure screening item;

[0056] In the specific implementation process, in this embodiment, heart failure is taken as the focus of the analysis, and by analyzing the left ventricular ejection analysis and natriuretic peptide, it is determined whether there are differences in the analysis of left ventricular ejection analysis and natriuretic peptide in the cases of patients after aortic dissection surgery, and then the subsequent heart failure screening items are analyzed, so that the difference items obtained are the items with large differences between the cases after aortic dissection surgery, that is, the items that should be analyzed emphatically; in actual analysis, other aspects after aortic dissection surgery can be selected as the focus of the analysis, such as prognostic analysis and preoperative analysis, and the analysis items corresponding to the focus of the analysis are obtained to meet the actual analysis needs;

[0057] When the case analysis item was not recorded as a heart failure screening item, the case analysis item was recorded as a regular screening item;

[0058] When the case analysis item is recorded as a heart failure screening item, multiple heart failure indicators are obtained based on big data, and the heart failure indicators in the case analysis item and the case control item are obtained respectively; for any heart failure indicator: multiple determination intervals corresponding to the heart failure indicator are obtained based on the determination criteria of the heart failure indicator, when the determination interval of the heart failure indicator in the case analysis item is the same as the determination interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a same-area indicator; when the determination interval of the heart failure indicator in the case analysis item is different from the determination interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a different-area indicator;

[0059] In the specific implementation process, the heart failure index may include data such as brain natriuretic peptide, troponin, B-type brain natriuretic peptide precursor and ultrasonic electrocardiogram. For example, in the actual analysis, the heart failure index analyzed is brain natriuretic peptide, and the normal value of brain natriuretic peptide obtained through data acquisition is: 17.2pmol / L to 24.4pmol / L for men, then the multiple judgment intervals of brain natriuretic peptide can be set as: less than 17.2pmol / L, greater than 24.4pmol / L and 17.2pmol / L to 24.4pmol / L; and at this time, the judgment interval of the heart failure index in the case analysis item is obtained through analysis to be 17.2pmol / L to 24.4pmol / L, and the judgment interval of the heart failure index in the case control item is 17.2pmol / L to 24.4pmol / L, then the heart failure index in the case analysis item can be marked as the same area index;

[0060] When all HF indicators were in the same region, the case analysis item was recorded as the same control item; when any HF indicator was a different region indicator, the case analysis item was recorded as the HF difference item;

[0061] When the case analysis item is recorded as a regular item to be screened, the name of each data in the case analysis item is obtained based on text extraction and recorded as the name to be compared; for any name to be compared, the normal interval of the data corresponding to the name to be compared is obtained based on big data and recorded as the normal data interval. When the data corresponding to the name to be compared is within the normal data interval, the name to be compared is recorded as a normal name; when the data corresponding to the name to be compared is outside the normal data interval, the name to be compared is recorded as an abnormal name;

[0062] In the specific implementation process, for example, in the actual analysis, the name to be compared is brain natriuretic peptide, and the normal data range of brain natriuretic peptide is 17.2pmol / L to 24.4pmol / L, and the data corresponding to the name to be compared is 25pmol / L, then the name to be compared should be recorded as an abnormal name, indicating that the data corresponding to the name to be compared is unconventional data and should be marked as abnormal;

[0063] When all the names to be compared in the case analysis item are normal names, the case analysis item is recorded as a same-control item; when any name to be compared is an abnormal name, the case analysis item is recorded as a regular difference item.

[0064] In the specific implementation process, when the names to be compared in all case analysis items are normal names, it means that the probability of data in the case analysis items being differentiated is small; when there are abnormal names in the names to be compared in all case analysis items, it means that the data in the case analysis items will be differentiated, so they need to be marked as regular difference items for distinction;

[0065] All dissection cases YA in the control case database are processed as cases to be analyzed in the basic analysis model in turn, and the heart failure difference items and conventional difference items corresponding to each dissection case YA are obtained; all heart failure difference items and conventional difference items are recorded using the basic analysis model, and the basic analysis model at this time is recorded as the preliminary analysis model.

[0066] The model building module is used to build the preliminary analysis model once and twice based on the control case database using the random analysis method, and obtain the case construction model; the model building module includes a model building unit, and the model building unit is configured with a model building strategy, and the model framework strategy includes:

[0067] Randomly obtain control case database The mezzanine cases YA are recorded as first-construction cases, and the mezzanine cases YA that are not recorded as first-construction cases are recorded as second-construction cases;

[0068] A single construction is as follows: for any single construction case α, obtain all single construction cases in which the names of the heart failure difference items in the single construction cases are different from the names of the heart failure difference items in the single construction case α, and record them as single comparison cases;

[0069] In the specific implementation process, for example, when processing data, the heart failure difference item of the first construction case α is the brain natriuretic peptide analysis item, and the heart failure difference item of the first construction case γ is the troponin analysis item, then the first construction case γ can be recorded as a comparison case; the purpose of obtaining a comparison case is to be able to more comprehensively screen the items that need to be differentially analyzed in the postoperative aspects of aortic dissection, so that the obtained difference items can more comprehensively reflect the items that may have hidden dangers in patients after aortic dissection surgery;

[0070] The name of the heart failure difference item of a comparison case is recorded as a comparison item, and the name of all data in the comparison item is recorded as a comparison name YB 1 To the first comparison name YB c ; For a comparison name YB 1 To the first comparison name YB c Any one of the comparison names YB v , based on big data to obtain a comparison name YB v In all corresponding data intervals, the name YB is compared once v The corresponding data interval is recorded as a comparison interval YQ v ; When a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison name YB is in the comparison item in the construction case αv Recorded as the normal comparison name; when a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison item in a construction case α is outside, the comparison name YB v Recorded as the name of the difference comparison;

[0071] In the specific implementation process, for example, when processing data, a comparison item is obtained and the name YB is compared in sequence. v It is a 6-minute walking test, and all data intervals of the 6-minute walking test include: less than 150M, less than or equal to 450M, greater than or equal to 150M, and greater than 450M; one comparison name YB v The corresponding data is 100M, that is, a comparison interval YQ v is less than 150M; the data corresponding to the 6-minute walking test in a construction case α is 200M, which is in the first comparison interval YQ v In addition, the 6-minute walk test in a comparison item in a constructed case α can be recorded as the name of the difference comparison; in subsequent analysis, the patient's 6-minute walk test can be analyzed emphatically;

[0072] When any one of the primary comparison names corresponding to the primary construction case α is recorded as a difference comparison name, the primary comparison item of the primary construction case α is recorded as a difference item.

[0073] The secondary construction is as follows: for any secondary construction case β, obtain the secondary construction cases whose names of the conventional difference items in all secondary construction cases are different from the names of the conventional difference items in the secondary construction case β, and record them as secondary comparison cases;

[0074] The name of the regular difference item of the secondary comparison case is recorded as the secondary comparison item, and the name of all data in the secondary comparison item is recorded as the secondary comparison name EB 1 To the secondary alignment name EB p ; For the secondary alignment name EB 1 To the secondary alignment name EB p Any secondary alignment name EB u , based on big data to obtain the secondary comparison name EB u In all corresponding data intervals, the secondary comparison name is YB u The corresponding data interval is recorded as the secondary comparison interval YQ u ; When the secondary alignment name EB in the secondary alignment item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQ u When the secondary alignment name EB in the secondary alignment item in the secondary construction case β isu Recorded as the normal comparison name; when the secondary comparison name EB in the secondary comparison item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQ u When the secondary comparison name EB in the secondary comparison item in the secondary construction case β is u Recorded as the name of the difference comparison;

[0075] When any secondary alignment name corresponding to the secondary construction case β is recorded as a difference alignment name, the secondary alignment item of the secondary construction case β is recorded as a difference item;

[0076] The preliminary analysis model after the first construction and the second construction is recorded as the case construction model.

[0077] The model application module is used to analyze the aortic dissection postoperative cases to be analyzed based on the case construction model, and obtain the data to be analyzed based on the analysis results; the model application module includes a case analysis unit, and the case analysis unit is used to record the aortic dissection postoperative cases to be analyzed as the cases to be placed when performing a comparative analysis on the aortic dissection postoperative cases, input the cases to be placed into the neurons corresponding to the cases to be analyzed at the input end of the case construction model, obtain the data in the difference items of the output of the case construction model as the data to be analyzed, and report the data to be analyzed to the staff;

[0078] In the specific implementation process, by building a model based on cases to obtain the data to be analyzed, it can be ensured that the data to be analyzed are the patient's post-aortic dissection data and the data that are different from those of other post-aortic dissection cases, which helps to reduce the data screening time of the staff while conducting a comprehensive analysis of the patient's data.

[0079] Example 2, second aspect, please refer to Figure 2 As shown, the present application also provides a method for constructing a control case analysis model for aortic dissection surgery based on big data, comprising the following steps:

[0080] Step S1, obtaining multiple post-operative cases of aortic dissection based on big data and storing them in a control case database; establishing a basic analysis model, and preliminarily constructing the basic analysis model based on the post-operative cases of aortic dissection to obtain a preliminary analysis model;

[0081] Step S1 includes: Step S101, obtaining multiple aortic dissection postoperative cases based on big data, and recording them as dissection cases YA 1 To mezzanine case YA n ; Store all interlayer cases YA in the control case database;

[0082] Step S102, establishing a basic analysis model, wherein the input end of the basic analysis model is two neurons for inputting control cases and cases to be analyzed; the output end is multiple neurons for outputting difference items obtained by analysis; the function in the basic analysis model is used to perform difference analysis on the control cases and cases to be analyzed received by the two neurons at the input end to obtain difference items, wherein the control cases are all cases in which the values ​​of various indicators are within the normal range, and the difference items include heart failure difference items and conventional difference items;

[0083] Step S103, the difference analysis includes: step S1031, obtaining multiple project names in the aortic dissection postoperative cases, recording them as dissection case names; extracting the dissection case names in the cases to be analyzed based on text extraction, and recording the content contained in each dissection case name as a case analysis item;

[0084] Step S1032, for any case analysis item: record the interlayer case name of the case analysis item as the name to be compared, obtain the name to be compared in the comparison case based on text extraction, and record the content contained in the name to be compared in the comparison case as the case comparison item;

[0085] Step S1033, based on text extraction, the left ventricular ejection fraction is recorded as the main reference point, and the natriuretic peptide is recorded as the auxiliary reference point; when there is a main reference point in the case analysis item, the value corresponding to the main reference point is recorded as the ejection fraction, and when the ejection fraction is greater than or equal to 50%, the value corresponding to the auxiliary reference point is obtained, and when BNP>35 and NT-proBNP>125, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 50% and greater than or equal to 40%, the medical history in the case analysis item is subjected to text extraction, and when there is heart disease in the medical history and BNP>35 and NT-proBNP>125 in the values ​​corresponding to the auxiliary reference point, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 40%, the case analysis item is recorded as a heart failure screening item;

[0086] Step S1034, when the case analysis item is not recorded as a heart failure screening item, the case analysis item is recorded as a regular screening item;

[0087] Step S1035, when the case analysis item is recorded as a heart failure screening item, multiple heart failure indicators are obtained based on big data, and the heart failure indicators in the case analysis item and the case control item are obtained respectively; for any heart failure indicator: based on the judgment standard of the heart failure indicator, multiple judgment intervals corresponding to the heart failure indicator are obtained, when the judgment interval of the heart failure indicator in the case analysis item is the same as the judgment interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a same-area indicator; when the judgment interval of the heart failure indicator in the case analysis item is different from the judgment interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a different-area indicator;

[0088] Step S1036, when all heart failure indicators are same-region indicators, the case analysis item is recorded as a same-control item; when any heart failure indicator is a different-region indicator, the case analysis item is recorded as a heart failure difference item;

[0089] Step S1037, when the case analysis item is recorded as a regular item to be screened, the name of each data in the case analysis item is obtained based on text extraction and recorded as a name to be compared; for any name to be compared, the normal interval of the data corresponding to the name to be compared is obtained based on big data and recorded as a normal data interval. When the data corresponding to the name to be compared is within the normal data interval, the name to be compared is recorded as a normal name; when the data corresponding to the name to be compared is outside the normal data interval, the name to be compared is recorded as an abnormal name;

[0090] Step S1038, when all the names to be compared in the case analysis item are normal names, the case analysis item is recorded as a same-control item; when any name to be compared is an abnormal name, the case analysis item is recorded as a regular difference item.

[0091] Step S104, use all the dissection cases YA in the control case database as the cases to be analyzed of the basic analysis model in turn, and obtain the heart failure difference items and conventional difference items corresponding to each dissection case YA; use the basic analysis model to record all the heart failure difference items and conventional difference items, and record the basic analysis model at this time as the preliminary analysis model.

[0092] Step S2, based on the control case database, the preliminary analysis model is constructed once and twice using the random analysis method, and the case construction model is obtained; the random analysis method includes: step S201, randomly obtaining the control case database The mezzanine cases YA are recorded as first-construction cases, and the mezzanine cases YA that are not recorded as first-construction cases are recorded as second-construction cases;

[0093] Step S202, a primary construction is as follows: Step S2021, for any primary construction case α, obtaining the primary construction cases in which the names of the heart failure difference items in the primary construction cases are different from the names of the heart failure difference items in the primary construction case α, and recording them as primary comparison cases;

[0094] Step S2022, record the name of the heart failure difference item of the primary comparison case as the primary comparison item, and record the name of all data in the primary comparison item as the primary comparison name YB 1 To the first comparison name YB c ; For a comparison name YB 1 To the first comparison name YB c Any one of the comparison names YBv , based on big data to obtain a comparison name YB v In all corresponding data intervals, the name YB is compared once v The corresponding data interval is recorded as a comparison interval YQ v ; When a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison name YB is in the comparison item in the construction case α v Recorded as the normal comparison name; when a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison item in a construction case α is outside, the comparison name YB v Recorded as the name of the difference comparison;

[0095] Step S2023, when any one of the primary comparison names corresponding to the primary construction case α is recorded as a difference comparison name, the primary comparison item of the primary construction case α is recorded as a difference item;

[0096] Step S203, secondary construction is as follows: Step S2031, for any secondary construction case β, obtain secondary construction cases in which the names of the conventional difference items in all secondary construction cases are different from the names of the conventional difference items of the secondary construction case β, and record them as secondary comparison cases;

[0097] Step S2032: record the name of the regular difference item of the secondary comparison case as the secondary comparison item, and record the name of all data in the secondary comparison item as the secondary comparison name EB 1 To the secondary alignment name EB p ; For the secondary alignment name EB 1 To the secondary alignment name EB p Any secondary alignment name EB u , based on big data to obtain the secondary comparison name EB u In all corresponding data intervals, the secondary comparison name is YB u The corresponding data interval is recorded as the secondary comparison interval YQ u ; When the secondary alignment name EB in the secondary alignment item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQ u When the secondary alignment name EB in the secondary alignment item in the secondary construction case β is u Recorded as the normal comparison name; when the secondary comparison name EB in the secondary comparison item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQu When the secondary comparison name EB in the secondary comparison item in the secondary construction case β is u Recorded as the name of the difference comparison;

[0098] Step S2033, when any secondary comparison name corresponding to the secondary construction case β is recorded as a difference comparison name, the secondary comparison item of the secondary construction case β is recorded as a difference item;

[0099] Step S2034: the preliminary analysis model after the first construction and the second construction is recorded as a case construction model.

[0100] Step S3, constructing a model based on the case to analyze the aortic dissection surgery case to be analyzed, and obtaining the data to be analyzed based on the analysis result.

[0101] Step S3 includes: when performing a comparative analysis on postoperative cases of aortic dissection, the postoperative cases of aortic dissection to be analyzed are recorded as cases to be placed, the cases to be placed are input into the neurons corresponding to the cases to be analyzed at the input end of the case construction model, the data in the difference items of the output of the case construction model are obtained as data to be analyzed, and the data to be analyzed are reported to the staff.

[0102] Example 3, please refer to Figure 4 As shown, Figure 4 The structural schematic diagram of an electronic device is illustrated, and the electronic device may include: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the method for constructing a control case analysis model for aortic dissection surgery based on big data are executed to achieve the following functions: first, multiple aortic dissection surgery cases are obtained based on big data, and stored in a control case database; a basic analysis model is established, and the basic analysis model is preliminarily constructed based on the aortic dissection surgery cases to obtain a preliminary analysis model; then, the preliminary analysis model is constructed once and twice based on the control case database using a random analysis method, and a case construction model is obtained; finally, the aortic dissection surgery cases to be analyzed are analyzed based on the case construction model, and the data to be analyzed are obtained based on the analysis results.

[0103] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0104] Example 4. The present application also provides a computer-readable storage medium. The present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the method for constructing a control case analysis model for aortic dissection surgery based on big data are executed to achieve the following functions: first, multiple aortic dissection surgery cases are obtained based on big data, and stored in a control case database; a basic analysis model is established, and the basic analysis model is preliminarily constructed based on the aortic dissection surgery cases to obtain a preliminary analysis model; then, the preliminary analysis model is constructed once and twice based on the control case database using a random analysis method, and a case construction model is obtained; finally, the aortic dissection surgery cases to be analyzed are analyzed based on the case construction model, and the data to be analyzed are obtained based on the analysis results.

[0105] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0106] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing a control case analysis model for aortic dissection surgery based on big data, characterized in that: The steps include: Based on big data, multiple postoperative cases of aortic dissection were obtained and stored in a control case database; a basic analysis model was established, and the basic analysis model was preliminarily constructed based on the postoperative cases of aortic dissection to obtain a preliminary analysis model; Based on the control case database, the preliminary analysis model was constructed once and twice using the random analysis method, and the case construction model was obtained; The model is constructed based on the case to be analyzed for the aortic dissection surgery case, and the data to be analyzed is obtained based on the analysis results; Based on big data, multiple aortic dissection surgery cases were obtained and stored in the control case database; Building a basic analysis model includes: Based on big data, multiple aortic dissection postoperative cases were obtained and recorded as dissection case YA1 to dissection case YA n ; All interlayer cases YA are stored in the control case database; A basic analysis model is established, wherein the input end of the basic analysis model is two neurons for inputting control cases and cases to be analyzed; the output end is multiple neurons for outputting difference items obtained by analysis; the function in the basic analysis model is used to perform difference analysis on the control cases and cases to be analyzed received by the two neurons at the input end to obtain difference items, wherein the control cases are all cases in which the values ​​of various indicators are within the normal range, and the difference items include heart failure difference items and conventional difference items; The basic analysis model was initially constructed based on the postoperative case of aortic dissection, and the preliminary analysis model also includes: All the dissection cases YA in the control case database are processed as the cases to be analyzed of the basic analysis model in turn, and the heart failure difference items and the conventional difference items corresponding to each dissection case YA are obtained; all the heart failure difference items and the conventional difference items are recorded using the basic analysis model, and the basic analysis model at this time is recorded as the preliminary analysis model; Stochastic analysis methods include: Randomly obtain control case database The mezzanine cases YA are recorded as first-construction cases, and the mezzanine cases YA that are not recorded as first-construction cases are recorded as second-construction cases; A single construction is as follows: for any single construction case α, obtain all single construction cases in which the names of the heart failure difference items in the single construction cases are different from the names of the heart failure difference items in the single construction case α, and record them as single comparison cases; The name of the heart failure difference item of the primary comparison case is recorded as the primary comparison item, and the names of all the data in the primary comparison item are recorded as primary comparison name YB1 to primary comparison name YB c ; For a comparison name YB1 to a comparison name YB c Any one of the comparison names YB v , based on big data to obtain a comparison name YB v In all corresponding data intervals, the name YB is compared once v The corresponding data interval is recorded as a comparison interval YQ v ; When a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison name YB is in the comparison item in the construction case α v Recorded as the normal comparison name; when a comparison name YB in a comparison item in a construction case α v The corresponding data is in the first comparison interval YQ v When the comparison item in a construction case α is outside, the comparison name YB v Recorded as the name of the difference comparison; When any one of the primary comparison names corresponding to a construction case α is recorded as a difference comparison name, the primary comparison item of the construction case α is recorded as a difference item; The secondary construction is as follows: for any secondary construction case β, obtain the secondary construction cases whose names of the conventional difference items in all secondary construction cases are different from the names of the conventional difference items in the secondary construction case β, and record them as secondary comparison cases; The name of the regular difference item of the secondary comparison case is recorded as the secondary comparison item, and the names of all data in the secondary comparison item are recorded as secondary comparison name EB1 to secondary comparison name EB p ; For secondary alignment name EB1 to secondary alignment name EB p Any secondary alignment name EB u , based on big data to obtain the secondary comparison name EB u In all corresponding data intervals, the secondary comparison name is YB u The interval where the corresponding data is located is recorded as the secondary comparison interval EQ u ; When the secondary alignment name EB in the secondary alignment item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQ u When the secondary alignment name EB in the secondary alignment item in the secondary construction case β is u Recorded as the normal comparison name; when the secondary comparison name EB in the secondary comparison item in the secondary construction case β u The corresponding data is in the secondary comparison interval EQ u When the secondary comparison name EB in the secondary comparison item in the secondary construction case β is u Recorded as the name of the difference comparison; When any secondary alignment name corresponding to the secondary construction case β is recorded as a difference alignment name, the secondary alignment item of the secondary construction case β is recorded as a difference item; The preliminary analysis model after the first construction and the second construction is recorded as the case construction model.

2. The method for constructing a control case analysis model for aortic dissection surgery based on big data according to claim 1, characterized in that: The basic analysis model was initially constructed based on the postoperative case of aortic dissection, and the preliminary analysis model included: The difference analysis includes: obtaining multiple project names in the postoperative cases of aortic dissection, recording them as dissection case names; extracting the dissection case names in the cases to be analyzed based on text extraction, and recording the content contained in each dissection case name as case analysis items; For any case analysis item: record the interlayer case name of the case analysis item as the name to be compared, obtain the name to be compared in the comparison case based on text extraction, and record the content contained in the name to be compared in the comparison case as the case comparison item.

3. The method for constructing a control case analysis model for aortic dissection surgery based on big data according to claim 2, characterized in that: The differential analysis also includes: Based on text extraction, the left ventricular ejection fraction is recorded as the main reference point, and the natriuretic peptide is recorded as the auxiliary reference point; when there is a main reference point in the case analysis item, the value corresponding to the main reference point is recorded as the ejection fraction. When the ejection fraction is greater than or equal to 50%, the value corresponding to the auxiliary reference point is obtained. When BNP>35 and NT-proBNP>125, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 50% and greater than or equal to 40%, the medical history in the case analysis item is extracted, and when there is heart disease in the medical history and BNP>35 and NT-proBNP>125 in the values ​​corresponding to the auxiliary reference point, the case analysis item is recorded as a heart failure screening item; when the ejection fraction is less than 40%, the case analysis item is recorded as a heart failure screening item; When a case analysis item is not recorded as a heart failure screening item, the case analysis item is recorded as a regular screening item.

4. The method for constructing a control case analysis model for aortic dissection surgery based on big data according to claim 3, characterized in that: The differential analysis also includes: When the case analysis item is recorded as a heart failure screening item, multiple heart failure indicators are obtained based on big data, and the heart failure indicators in the case analysis item and the case control item are obtained respectively; for any heart failure indicator: multiple determination intervals corresponding to the heart failure indicator are obtained based on the determination criteria of the heart failure indicator, when the determination interval of the heart failure indicator in the case analysis item is the same as the determination interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a same-area indicator; when the determination interval of the heart failure indicator in the case analysis item is different from the determination interval of the heart failure indicator in the case control item, the heart failure indicator in the case analysis item is marked as a different-area indicator; When all heart failure indicators are same-region indicators, the case analysis item is recorded as the same-control item; when any heart failure indicator is a different-region indicator, the case analysis item is recorded as the heart failure difference item.

5. The method for constructing a control case analysis model for aortic dissection surgery based on big data according to claim 4, characterized in that: The differential analysis also includes: When the case analysis item is recorded as a regular item to be screened, the name of each data in the case analysis item is obtained based on text extraction and recorded as the name to be compared; for any name to be compared, the normal interval of the data corresponding to the name to be compared is obtained based on big data and recorded as the normal data interval. When the data corresponding to the name to be compared is within the normal data interval, the name to be compared is recorded as a normal name; when the data corresponding to the name to be compared is outside the normal data interval, the name to be compared is recorded as an abnormal name; When all the names to be compared in the case analysis item are normal names, the case analysis item is recorded as a same-control item; when any name to be compared is an abnormal name, the case analysis item is recorded as a regular difference item.

6. The method for constructing a control case analysis model for aortic dissection surgery based on big data according to claim 5, characterized in that: The model is constructed based on the case to be analyzed for the aortic dissection surgery case, and the data to be analyzed are obtained based on the analysis results, including: When performing a comparative analysis on postoperative cases of aortic dissection, the postoperative cases of aortic dissection to be analyzed are recorded as cases to be placed, and the cases to be placed are input into the neurons corresponding to the cases to be analyzed at the input end of the case construction model, and the data in the difference items of the output of the case construction model are obtained as data to be analyzed, and the data to be analyzed are reported to the staff.

7. A system for constructing a control case analysis model for aortic dissection surgery based on big data, used to implement the method for constructing a control case analysis model for aortic dissection surgery based on big data as described in any one of claims 1 to 6, characterized in that: It includes preliminary model building module, model construction module and model application module; The preliminary model building module is used to obtain multiple postoperative cases of aortic dissection based on big data and store them in the control case database; Establish a basic analysis model, and preliminarily construct the basic analysis model based on aortic dissection postoperative case to obtain a preliminary analysis model; The model building module is used to build the preliminary analysis model once and twice based on the control case database using the random analysis method, and obtain the case building model; The model application module is used to build a model based on the case to analyze the postoperative cases of aortic dissection to be analyzed, and obtain the data to be analyzed based on the analysis results.

Images