Coronary heart disease atherosclerosis data grading model construction method and system and medium

Abstract

The invention discloses a coronary heart disease atherosclerosis data grading model construction method and system and a medium, and belongs to the technical field of data analys.The coronary heart disease atherosclerosis data grading model construction method comprises the steps that diagnosis records of a patient are obtained, and comprehensive characteristic values of the diagnosis records are calculated; classifying and counting the diagnosis records of all patients, calculating a characteristic threshold value of each type of diagnosis records, and constructing a hardening grade model; calculating the characteristic variation of the patient, and matching a target patient; acquiring a treatment record of a target patient, generating suggestion information, and perfecting the sclerosis level model; according to the real-time diagnosis record of the patient and the sclerosis level model, outputting the sclerosis level and suggestion information of the patient; according to the invention, more accurate and comprehensive evaluation can be carried out according to the diagnosis record of the patient, a proper treatment scheme can be recommended for the early patient, medical intervention measures are adjusted in time, and the diagnosis and treatment effect and the health management level of the patient are comprehensively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data analysis, and specifically relates to a method, system and medium for constructing a data classification model for coronary heart disease atherosclerosis. Background Art

[0002] Atherosclerosis refers to the formation of plaques on the inner wall of the artery due to the deposition of lipids (such as cholesterol), calcium salts, inflammatory cells and fibrous tissue, which causes the blood vessels to harden and narrow. In severe cases, it can cause blood flow obstruction, thereby affecting the blood supply to organs and tissues. This pathological process is a common form of arteriosclerosis and is also the main cause of cardiovascular diseases (such as coronary heart disease);

[0003] The existing grading model for atherosclerosis does not perform any invasive operation. It collects plasma samples from patients and measures the levels of biomarkers (such as cholesterol sulfate, azelaic acid, tryptophan, etc.) in the plasma to distinguish the severity of atherosclerosis and perform graded diagnosis.

[0004] However, there are errors in grading patients only by testing biomarkers. In addition, there is a lack of comprehensive analysis of multi-dimensional atherosclerosis data of patients, which cannot accurately reflect the development process of atherosclerosis. In particular, for early or mild patients, it is impossible to provide sufficiently accurate grading information and treatment recommendations.

[0005] Therefore, people are in urgent need of a coronary heart disease atherosclerosis data classification model construction system to solve the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide a method for constructing a data classification model for coronary heart disease atherosclerosis to solve the problems raised in the above background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] A method for constructing a data classification model for coronary heart disease atherosclerosis, the method comprising the following steps:

[0009] S1. Obtain the patient's diagnosis record from the database, perform feature extraction on the diagnosis record, obtain the blood sample feature value and image feature value of the diagnosis record, and calculate the comprehensive feature value of the diagnosis record;

[0010] S2. Classify and count the diagnostic records of all patients, obtain the characteristic intervals of different types of diagnostic records, calculate the characteristic thresholds of each type of diagnostic record, and construct a sclerosis grade model;

[0011] S3. Filter out all the diagnostic records of a certain patient, calculate the characteristic change of the patient, and obtain the sclerosis grade of the patient through the sclerosis grade model to match the target patient; obtain the treatment record of the target patient, generate suggestion information, and improve the sclerosis grade model;

[0012] S4. Output the patient's sclerosis level and recommended information based on the patient's real-time diagnostic record and sclerosis level model.

[0013] According to the above technical solution, step S1 includes the following:

[0014] S1-1. Each time a patient undergoes a diagnosis, a diagnosis record is generated and stored in the database; the diagnosis record includes the patient number, diagnosis time, blood sample data and imaging data;

[0015] S1-2. Extract biomarkers from blood sample data of a certain diagnosis record, wherein the biomarkers include lipid metabolism markers, inflammatory response markers, endothelial function markers, oxidative stress markers, plaque stability markers and thrombosis markers; extract the content of each biomarker from the blood sample data of the diagnosis record, using the formula: B1 = ∑ i 1 (α1 i ×A1 i ), calculate the blood sample characteristic value of the diagnostic record; where B1 represents the blood sample characteristic value of the diagnostic record, α1 i represents the content weight of the i-th biomarker, A1 i represents the content of the ith biomarker;

[0016] Lipid metabolism markers include but are not limited to oxidized low-density lipoprotein and low-density lipoprotein cholesterol. Lipid metabolism markers accumulate in the arterial wall and are prone to form plaques. Inflammatory response markers include but are not limited to interleukins and tumor necrosis factors, which have an important impact on the development of atherosclerosis. Endothelial function markers include but are not limited to endothelin and vascular endothelial cell adhesion molecules, which reflect the damage of vascular endothelial function and further cause atherosclerosis. Oxidative stress markers include but are not limited to malondialdehyde and superoxide dismutase, which reflect that the level of oxidative stress plays an important role in the occurrence of atherosclerosis. Plaque stability markers include but are not limited to matrix metalloproteinases and lipoprotein phospholipases, which participate in the degradation of vascular matrix and affect the formation and stability of plaques. Thrombosis markers include but are not limited to fibrinogen and dimers, which affect thrombosis and dissolution. Thrombosis is an important complication of atherosclerosis. When calculating the characteristic values ​​of blood samples in diagnostic records, each marker selects any one of the substances it includes as a reference for calculation;

[0017] S1-3, preprocessing the image data of a certain diagnostic record, wherein the preprocessing includes denoising, normalization and contrast enhancement;

[0018] The edge detection algorithm is used to extract the inner wall contour and the outer wall contour of the artery in the image data of the diagnosis record, and the pixel points of the inner wall contour of the artery are counted to form an inner wall pixel point set; the pixel points of the inner wall contour of the artery are counted to form an outer wall pixel point set; any pixel point in the image data of the diagnosis record is selected, and the shortest distance between the pixel point and the inner wall pixel point set is calculated, which is recorded as D1; ​​the shortest distance between the pixel point and the outer wall pixel point set is calculated, which is recorded as D2; the formula: D=|D1-D2| is used to calculate the vascular wall thickness of the diagnosis record; wherein D represents the vascular wall thickness of the diagnosis record;

[0019] A threshold segmentation algorithm is used to extract the plaque area of ​​the arterial blood vessels in the image data of the diagnostic record, and the total number of pixels in the plaque area of ​​the arterial blood vessels in the image data of the diagnostic record is counted, which is recorded as N; the plaque area of ​​the diagnostic record is calculated using the formula: S=N×V; wherein S represents the plaque area of ​​the diagnostic record, and V represents the actual physical area corresponding to each pixel in the image data of the diagnostic record;

[0020] Edge detection algorithms and threshold segmentation algorithms are widely used in the analysis of medical images. The actual physical area corresponding to each pixel in the image data is determined by the spatial resolution. Spatial resolution refers to the size of each pixel in the actual physical space, which is set by the image acquisition device.

[0021] The image characteristic value of the diagnosis record is calculated using the formula: B2 = α2 × D + α3 × S; wherein B2 represents the image characteristic value of the diagnosis record, α2 represents the weight of the vascular wall thickness, and α3 represents the weight of the plaque area;

[0022] Blood vessel wall thickness is one of the important indicators for assessing atherosclerosis. Thickened blood vessel walls are often an early manifestation of atherosclerosis. Atherosclerosis is usually accompanied by the deposition of lipids, calcium and fibrous tissue on the inner wall of the blood vessel to form plaques. The area of ​​the plaque is related to the severity of atherosclerosis. Larger plaques usually mean more serious lesions, which may lead to vascular stenosis or even complete occlusion. Using blood vessel wall thickness and plaque area together with blood sample characteristics as detection conditions for atherosclerosis makes the grading results more accurate and comprehensive.

[0023] S1-4. Calculate the comprehensive characteristic value of the diagnostic record using the formula: C=β×B1+(1-β)×B2; wherein C represents the comprehensive characteristic value of the diagnostic record, and β represents the adjustment parameter.

[0024] According to the above technical solution, step S2 includes the following:

[0025] S2-1. By expert annotation of the patient's diagnostic records, the diagnostic records are divided into different types, including mild sclerosis records, moderate sclerosis records and severe sclerosis records;

[0026] Expert annotation refers to the process of labeling, classifying or annotating data by experts in the field according to their own experience standards and requirements; it is widely used in machine learning and data analysis to further improve the training accuracy of the model;

[0027] S2-2, counting the minimum value of the comprehensive characteristic value corresponding to all diagnostic records of a certain type, as the minimum value of the characteristic interval corresponding to the diagnostic records of this type; counting the maximum value of the comprehensive characteristic value corresponding to all diagnostic records of a certain type, as the maximum value of the characteristic interval corresponding to the diagnostic records of this type;

[0028] S2-3, taking the minimum value of the characteristic interval corresponding to the diagnostic record of type mild hardening record as the first threshold; counting all diagnostic records whose comprehensive characteristic values ​​are between the maximum value of the characteristic interval corresponding to the diagnostic record of type mild hardening record and the minimum value of the characteristic interval corresponding to the diagnostic record of type moderate hardening record, and calculating the average value of the corresponding comprehensive characteristic values, as the second threshold; counting all diagnostic records whose comprehensive characteristic values ​​are between the maximum value of the characteristic interval corresponding to the diagnostic record of type moderate hardening record and the minimum value of the characteristic interval corresponding to the diagnostic record of type severe hardening record, and calculating the average value of the corresponding comprehensive characteristic values, as the third threshold;

[0029] The obtained first threshold, second threshold and third threshold are used as feature thresholds;

[0030] According to the comprehensive characteristic value and the type of diagnostic record, the characteristic threshold is calculated to provide a basis for model construction;

[0031] S2-4, construct a hardening grade model, specifically as follows: patients whose diagnostic records correspond to comprehensive characteristic values ​​less than the first threshold are at the first level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the first threshold and less than the second threshold are at the second level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the second threshold and less than the third threshold are at the third level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the third threshold are at the fourth level of hardening;

[0032] The first hardening, the second hardening, the third hardening and the fourth hardening are regarded as the hardening grades;

[0033] The first sclerosis grade indicates that the patient does not have atherosclerosis.

[0034] According to the above technical solution, step S3 includes the following:

[0035] S3-1. According to the patient number and diagnosis time in the diagnosis record, all diagnosis records of a patient are screened out and sorted from the earliest to the latest according to the diagnosis time; the average value of the comprehensive characteristic value corresponding to the latter diagnosis record minus the comprehensive characteristic value corresponding to the previous diagnosis record among all adjacent diagnosis records of the patient is calculated as the characteristic change amount of the patient;

[0036] The comprehensive characteristic value corresponding to each diagnostic record of the patient is input into the sclerosis grade model, and the sclerosis grade of the patient in each diagnostic record is output through the sclerosis grade model;

[0037] S3-2, if the characteristic change of a patient is less than zero, and the sclerosis level of the patient is the second level sclerosis, the patient is taken as the target patient;

[0038] S3-3. Obtain the target patient's medication dosage, medication course, diet control, exercise plan, and psychological assistance course from the database as the target patient's treatment record;

[0039] Take the absolute value of the corresponding feature changes of all target patients, screen out the target patient with the largest absolute value of the corresponding feature changes, and use the treatment record of this patient as the recommended information;

[0040] The suggested information is used as the suggested information of the second level hardening;

[0041] In the early stages, atherosclerosis usually manifests as lipid deposition and the formation of small plaques on the blood vessel walls. At this time, the blood vessel walls become thicker but not severely narrowed. At this time, atherosclerosis can be effectively controlled and may even be reversed through lifestyle changes and drug therapy. By screening target patients and generating recommendation information, treatment recommendations can be provided for the treatment of early or mild patients.

[0042] According to the above technical solution, step S4 includes the following:

[0043] According to the patient's real-time diagnosis record, the comprehensive characteristic value corresponding to the patient's real-time diagnosis record is calculated and input into the hardening grade model. The hardening grade model outputs the patient's hardening grade. If the patient is at the second hardening grade, the recommendation information is output at the same time.

[0044] A system for constructing a data classification model for coronary heart disease atherosclerosis, the system comprising an information collection module, an information processing module and an execution module;

[0045] The information acquisition module is used to acquire the information required by the system; the information processing module is used to store, analyze and transmit the information of each module; and the execution module is used to execute the information of the information processing module.

[0046] According to the above technical solution, the information acquisition module includes a diagnosis recording unit and a treatment recording unit;

[0047] The diagnosis record unit is used to collect the patient's diagnosis record; the treatment record unit is used to collect the patient's treatment record.

[0048] According to the above technical solution, the information processing module includes an information storage unit, an information analysis unit and an information transmission unit;

[0049] The information storage unit is used to store the information acquired by the information acquisition module; the information analysis unit is used to analyze the information acquired by the information acquisition module; and the information transmission unit is used for information transmission of each module in the system.

[0050] According to the above technical solution, the execution module includes a visualization unit and a push unit;

[0051] The visualization unit is used to form visualization data of the patient's sclerosis level and recommended information; the push unit is used to push the patient's sclerosis level and recommended information.

[0052] A medium for constructing a data classification model for coronary heart disease atherosclerosis includes a data storage medium for storing data.

[0053] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0054] The present invention realizes accurate disease classification by monitoring and analyzing data such as historical diagnosis records and real-time diagnosis records of patients with coronary heart disease and atherosclerosis, and optimizes early treatment plans in real time by analyzing changes in patients' characteristics; the present invention can not only perform more accurate and comprehensive evaluations based on patients' diagnosis records, but also recommend appropriate treatment plans for early patients, adjust medical intervention measures in a timely manner, and comprehensively improve diagnosis and treatment effects and patient health management levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0056] Figure 1 It is a schematic diagram of the process of constructing a data classification model of coronary heart disease atherosclerosis according to the present invention;

[0057] Figure 2 It is a structural schematic diagram of the coronary heart disease atherosclerosis data classification model construction system of the present invention. DETAILED DESCRIPTION

[0058] 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 technical visitors in this field without creative work are within the scope of protection of the present invention.

[0059] See also Figure 1 , the present invention provides a technical solution:

[0060] A method for constructing a data classification model for coronary heart disease atherosclerosis, the method comprising the following steps:

[0061] S1. Obtain the patient's diagnosis record from the database, perform feature extraction on the diagnosis record, obtain the blood sample feature value and image feature value of the diagnosis record, and calculate the comprehensive feature value of the diagnosis record;

[0062] According to the above technical solution, step S1 includes the following:

[0063] S1-1. Each time a patient undergoes a diagnosis, a diagnosis record is generated and stored in the database; the diagnosis record includes the patient number, diagnosis time, blood sample data and imaging data;

[0064] S1-2. Extract biomarkers from blood sample data of a certain diagnosis record, wherein the biomarkers include lipid metabolism markers, inflammatory response markers, endothelial function markers, oxidative stress markers, plaque stability markers and thrombosis markers; extract the content of each biomarker from the blood sample data of the diagnosis record, using the formula: B1 = ∑ i 1 (α1 i ×A1 i ), calculate the blood sample characteristic value of the diagnostic record; where B1 represents the blood sample characteristic value of the diagnostic record, α1 i represents the content weight of the i-th biomarker, A1 i represents the content of the ith biomarker;

[0065] Lipid metabolism markers include but are not limited to oxidized low-density lipoprotein and low-density lipoprotein cholesterol. Lipid metabolism markers accumulate in the arterial wall and are prone to form plaques. Inflammatory response markers include but are not limited to interleukins and tumor necrosis factors, which have an important impact on the development of atherosclerosis. Endothelial function markers include but are not limited to endothelin and vascular endothelial cell adhesion molecules, which reflect the damage of vascular endothelial function and further cause atherosclerosis. Oxidative stress markers include but are not limited to malondialdehyde and superoxide dismutase, which reflect that the level of oxidative stress plays an important role in the occurrence of atherosclerosis. Plaque stability markers include but are not limited to matrix metalloproteinases and lipoprotein phospholipases, which participate in the degradation of vascular matrix and affect the formation and stability of plaques. Thrombosis markers include but are not limited to fibrinogen and dimers, which affect thrombosis and dissolution. Thrombosis is an important complication of atherosclerosis. When calculating the characteristic values ​​of blood samples in diagnostic records, each marker selects any one of the substances it includes as a reference for calculation;

[0066] S1-3, preprocessing the image data of a certain diagnostic record, wherein the preprocessing includes denoising, normalization and contrast enhancement;

[0067] The edge detection algorithm is used to extract the inner wall contour and the outer wall contour of the artery in the image data of the diagnosis record, and the pixel points of the inner wall contour of the artery are counted to form an inner wall pixel point set; the pixel points of the inner wall contour of the artery are counted to form an outer wall pixel point set; any pixel point in the image data of the diagnosis record is selected, and the shortest distance between the pixel point and the inner wall pixel point set is calculated, which is recorded as D1; ​​the shortest distance between the pixel point and the outer wall pixel point set is calculated, which is recorded as D2; the formula: D=|D1-D2| is used to calculate the vascular wall thickness of the diagnosis record; wherein D represents the vascular wall thickness of the diagnosis record;

[0068] A threshold segmentation algorithm is used to extract the plaque area of ​​the arterial blood vessels in the image data of the diagnostic record, and the total number of pixels in the plaque area of ​​the arterial blood vessels in the image data of the diagnostic record is counted, which is recorded as N; the plaque area of ​​the diagnostic record is calculated using the formula: S=N×V; wherein S represents the plaque area of ​​the diagnostic record, and V represents the actual physical area corresponding to each pixel in the image data of the diagnostic record;

[0069] Edge detection algorithms and threshold segmentation algorithms are widely used in the analysis of medical images. The actual physical area corresponding to each pixel in the image data is determined by the spatial resolution. Spatial resolution refers to the size of each pixel in the actual physical space, which is set by the image acquisition device.

[0070] The image characteristic value of the diagnosis record is calculated using the formula: B2 = α2 × D + α3 × S; wherein B2 represents the image characteristic value of the diagnosis record, α2 represents the weight of the vascular wall thickness, and α3 represents the weight of the plaque area;

[0071] Blood vessel wall thickness is one of the important indicators for assessing atherosclerosis. Thickened blood vessel walls are often an early manifestation of atherosclerosis. Atherosclerosis is usually accompanied by the deposition of lipids, calcium and fibrous tissue on the inner wall of the blood vessel to form plaques. The area of ​​the plaque is related to the severity of atherosclerosis. Larger plaques usually mean more serious lesions, which may lead to vascular stenosis or even complete occlusion. Using blood vessel wall thickness and plaque area together with blood sample characteristics as detection conditions for atherosclerosis makes the grading results more accurate and comprehensive.

[0072] S1-4. Calculate the comprehensive characteristic value of the diagnostic record using the formula: C=β×B1+(1-β)×B2; wherein C represents the comprehensive characteristic value of the diagnostic record, and β represents the adjustment parameter.

[0073] S2. Classify and count the diagnostic records of all patients, obtain the characteristic intervals of different types of diagnostic records, calculate the characteristic thresholds of each type of diagnostic record, and construct a sclerosis grade model;

[0074] According to the above technical solution, step S2 includes the following:

[0075] S2-1. By expert annotation of the patient's diagnostic records, the diagnostic records are divided into different types, including mild sclerosis records, moderate sclerosis records and severe sclerosis records;

[0076] Expert annotation refers to the process of labeling, classifying or annotating data by experts in the field according to their own experience standards and requirements; it is widely used in machine learning and data analysis to further improve the training accuracy of the model;

[0077] S2-2, counting the minimum value of the comprehensive characteristic value corresponding to all diagnostic records of a certain type, as the minimum value of the characteristic interval corresponding to the diagnostic records of this type; counting the maximum value of the comprehensive characteristic value corresponding to all diagnostic records of a certain type, as the maximum value of the characteristic interval corresponding to the diagnostic records of this type;

[0078] S2-3, taking the minimum value of the characteristic interval corresponding to the diagnostic record of type mild hardening record as the first threshold; counting all diagnostic records whose comprehensive characteristic values ​​are between the maximum value of the characteristic interval corresponding to the diagnostic record of type mild hardening record and the minimum value of the characteristic interval corresponding to the diagnostic record of type moderate hardening record, and calculating the average value of the corresponding comprehensive characteristic values, as the second threshold; counting all diagnostic records whose comprehensive characteristic values ​​are between the maximum value of the characteristic interval corresponding to the diagnostic record of type moderate hardening record and the minimum value of the characteristic interval corresponding to the diagnostic record of type severe hardening record, and calculating the average value of the corresponding comprehensive characteristic values, as the third threshold;

[0079] The obtained first threshold, second threshold and third threshold are used as feature thresholds;

[0080] For example:

[0081] The corresponding characteristic interval of the diagnostic records of mild hardening records is [30, 60], the corresponding characteristic interval of the diagnostic records of moderate hardening records is [40, 70], and the corresponding characteristic interval of the diagnostic records of severe hardening records is [80, 85]. The minimum value of the corresponding characteristic interval of the diagnostic records of the type of mild hardening records is used as the first threshold, and the first threshold is 30. The statistical comprehensive characteristic value of all diagnostic records between the maximum value of the corresponding characteristic interval of the diagnostic records of the type of mild hardening records and the minimum value of the corresponding characteristic interval of the diagnostic records of the type of moderate hardening records is calculated. There are 5 records, and the corresponding comprehensive characteristic values ​​are 40, 40, 50, 60, 60 respectively, and the average value of the corresponding comprehensive characteristic values ​​is 50, which is used as the second threshold value, and the second threshold value is 30; there are 3 diagnostic records whose comprehensive characteristic values ​​are between the maximum value of the characteristic interval corresponding to the diagnostic record of the type of moderate hardening record and the minimum value of the characteristic interval corresponding to the diagnostic record of the type of severe hardening record, and the corresponding comprehensive characteristic values ​​are 70, 75, 80 respectively, and the average value of the corresponding comprehensive characteristic values ​​is 75, which is used as the third threshold value, and the third threshold value is 75;

[0082] According to the comprehensive characteristic value and the type of diagnostic record, the characteristic threshold is calculated to provide a basis for model construction;

[0083] S2-4, construct a hardening grade model, specifically as follows: patients whose diagnostic records correspond to comprehensive characteristic values ​​less than the first threshold are at the first level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the first threshold and less than the second threshold are at the second level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the second threshold and less than the third threshold are at the third level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the third threshold are at the fourth level of hardening;

[0084] The first sclerosis, the second sclerosis, the third sclerosis and the fourth sclerosis are regarded as sclerosis grades; wherein the first sclerosis grade indicates that the patient does not have atherosclerosis.

[0085] S3. Filter out all the diagnostic records of a certain patient, calculate the characteristic change of the patient, and obtain the sclerosis grade of the patient through the sclerosis grade model to match the target patient; obtain the treatment record of the target patient, generate suggestion information, and improve the sclerosis grade model;

[0086] According to the above technical solution, step S3 includes the following:

[0087] S3-1. According to the patient number and diagnosis time in the diagnosis record, all diagnosis records of a patient are screened out and sorted from the earliest to the latest according to the diagnosis time; the average value of the comprehensive characteristic value corresponding to the latter diagnosis record minus the comprehensive characteristic value corresponding to the previous diagnosis record among all adjacent diagnosis records of the patient is calculated as the characteristic change amount of the patient;

[0088] Inputting the comprehensive characteristic value corresponding to each diagnosis record of the patient into the sclerosis grade model, and outputting the sclerosis grade of the patient in each diagnosis record through the sclerosis grade model;

[0089] For example:

[0090] The corresponding comprehensive characteristic values ​​of a patient's diagnosis record are 40, 38, and 35; the corresponding comprehensive characteristic values ​​of the adjacent diagnosis records of the patient are 40 and 38, wherein the corresponding comprehensive characteristic value of the latter diagnosis record in the adjacent diagnosis records is 38, and the corresponding comprehensive characteristic value of the previous diagnosis record is 40; the corresponding comprehensive characteristic values ​​of the adjacent diagnosis records of the patient are 38 and 35, wherein the corresponding comprehensive characteristic value of the latter diagnosis record in the adjacent diagnosis records is 35, and the corresponding comprehensive characteristic value of the previous diagnosis record is 35; the average value of the corresponding comprehensive characteristic value of the latter diagnosis record minus the corresponding comprehensive characteristic value of the previous diagnosis record in all adjacent diagnosis records of the patient is calculated as [(38-40)+(35-38)] / 2=-2.5;

[0091] S3-2, if the characteristic change of a patient is less than zero, and the sclerosis level of the patient is the second level sclerosis, the patient is taken as the target patient;

[0092] S3-3. Obtain the target patient's medication dosage, medication course, diet control, exercise plan, and psychological assistance course from the database as the target patient's treatment record;

[0093] Take the absolute value of the corresponding feature changes of all target patients, screen out the target patient with the largest absolute value of the corresponding feature changes, and use the treatment record of this patient as the recommended information;

[0094] The suggested information is used as the suggested information of the second level hardening;

[0095] In the early stages, atherosclerosis usually manifests as lipid deposition and the formation of small plaques on the blood vessel walls. At this time, the blood vessel walls become thicker but not severely narrowed. At this time, atherosclerosis can be effectively controlled and may even be reversed through lifestyle changes and drug therapy. By screening target patients and generating recommendation information, treatment recommendations can be provided for the treatment of early or mild patients.

[0096] S4, outputting the patient's sclerosis level and recommended information according to the patient's real-time diagnostic record and sclerosis level model;

[0097] According to the above technical solution, step S4 includes the following:

[0098] According to the patient's real-time diagnosis record, the comprehensive characteristic value corresponding to the patient's real-time diagnosis record is calculated and input into the hardening grade model. The hardening grade model outputs the patient's hardening grade. If the patient is at the second hardening grade, the recommendation information is output at the same time.

[0099] See also Figure 2 , a coronary heart disease atherosclerosis data classification model construction system, the system includes an information collection module, an information processing module and an execution module;

[0100] The information acquisition module is used to acquire the information required by the system; the information processing module is used to store, analyze and transmit the information of each module; and the execution module is used to execute the information of the information processing module.

[0101] According to the above technical solution, the information acquisition module includes a diagnosis recording unit and a treatment recording unit;

[0102] The diagnosis record unit is used to collect the patient's diagnosis record; the treatment record unit is used to collect the patient's treatment record.

[0103] According to the above technical solution, the information processing module includes an information storage unit, an information analysis unit and an information transmission unit;

[0104] The information storage unit is used to store the information acquired by the information acquisition module; the information analysis unit is used to analyze the information acquired by the information acquisition module; and the information transmission unit is used for information transmission of each module in the system.

[0105] According to the above technical solution, the execution module includes a visualization unit and a push unit;

[0106] The visualization unit is used to form visualization data of the patient's sclerosis level and recommended information; the push unit is used to push the patient's sclerosis level and recommended information.

[0107] A medium for constructing a data classification model for coronary heart disease atherosclerosis includes a data storage medium for storing data.

[0108] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, for technical visitors in the field, it is still possible to modify the technical solutions recorded in the aforementioned embodiments, or to replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a data classification model for coronary heart disease atherosclerosis, characterized by: The method comprises the following steps: S1. Obtain the patient's diagnosis record from the database, perform feature extraction on the diagnosis record, obtain the blood sample feature value and image feature value of the diagnosis record, and calculate the comprehensive feature value of the diagnosis record; S2. Classify and count the diagnostic records of all patients, obtain the characteristic intervals of different types of diagnostic records, calculate the characteristic thresholds of each type of diagnostic record, and construct a sclerosis grade model; S3. Filter out all the diagnostic records of a certain patient, calculate the characteristic change of the patient, and obtain the sclerosis grade of the patient through the sclerosis grade model to match the target patient; obtain the treatment record of the target patient, generate suggestion information, and improve the sclerosis grade model; S4. Output the patient's sclerosis level and recommended information based on the patient's real-time diagnostic record and sclerosis level model.

2. The method for constructing a data classification model for coronary heart disease atherosclerosis according to claim 1, characterized in that: The step S1 comprises the following: S1-1. Each time a patient undergoes a diagnosis, a diagnosis record is generated and stored in the database; the diagnosis record includes the patient number, diagnosis time, blood sample data and imaging data; S1-2. Extract biomarkers from blood sample data of a certain diagnosis record, wherein the biomarkers include lipid metabolism markers, inflammatory response markers, endothelial function markers, oxidative stress markers, plaque stability markers and thrombosis markers; extract the content of each biomarker from the blood sample data of the diagnosis record, using the formula: B1 = ∑ i 1(α1 i ×A1 i ), calculate the blood sample characteristic value of the diagnostic record; where B1 represents the blood sample characteristic value of the diagnostic record, α1 i represents the content weight of the i-th biomarker, A1 i represents the content of the ith biomarker; S1-3, preprocessing the image data of a certain diagnostic record, wherein the preprocessing includes denoising, normalization and contrast enhancement; The edge detection algorithm is used to extract the inner wall contour and the outer wall contour of the artery in the image data of the diagnosis record, and the pixel points of the inner wall contour of the artery are counted to form an inner wall pixel point set; the pixel points of the inner wall contour of the artery are counted to form an outer wall pixel point set; any pixel point in the image data of the diagnosis record is selected, and the shortest distance between the pixel point and the inner wall pixel point set is calculated, which is recorded as D1; ​​the shortest distance between the pixel point and the outer wall pixel point set is calculated, which is recorded as D2; the formula: D=|D1-D2| is used to calculate the vascular wall thickness of the diagnosis record; wherein D represents the vascular wall thickness of the diagnosis record; A threshold segmentation algorithm is used to extract the plaque area of ​​the arterial blood vessels in the image data of the diagnostic record, and the total number of pixels in the plaque area of ​​the arterial blood vessels in the image data of the diagnostic record is counted, which is recorded as N; the plaque area of ​​the diagnostic record is calculated using the formula: S=N×V; wherein S represents the plaque area of ​​the diagnostic record, and V represents the actual physical area corresponding to each pixel in the image data of the diagnostic record; The image characteristic value of the diagnosis record is calculated using the formula: B2 = α2 × D + α3 × S; wherein B2 represents the image characteristic value of the diagnosis record, α2 represents the weight of the vascular wall thickness, and α3 represents the weight of the plaque area; S1-4. Calculate the comprehensive characteristic value of the diagnostic record using the formula: C=β×B1+(1-β)×B2; wherein C represents the comprehensive characteristic value of the diagnostic record, and β represents the adjustment parameter.

3. The method for constructing a data classification model for coronary heart disease atherosclerosis according to claim 2, characterized in that: The step S2 comprises the following: S2-1. By expert annotation of the patient's diagnostic records, the diagnostic records are divided into different types, including mild sclerosis records, moderate sclerosis records and severe sclerosis records; S2-2, counting the minimum value of the comprehensive characteristic value corresponding to all diagnostic records of a certain type, as the minimum value of the characteristic interval corresponding to the diagnostic records of this type; counting the maximum value of the comprehensive characteristic value corresponding to all diagnostic records of a certain type, as the maximum value of the characteristic interval corresponding to the diagnostic records of this type; S2-3, taking the minimum value of the characteristic interval corresponding to the diagnostic record of type mild hardening record as the first threshold; counting all diagnostic records whose comprehensive characteristic values ​​are between the maximum value of the characteristic interval corresponding to the diagnostic record of type mild hardening record and the minimum value of the characteristic interval corresponding to the diagnostic record of type moderate hardening record, and calculating the average value of the corresponding comprehensive characteristic values, as the second threshold; counting all diagnostic records whose comprehensive characteristic values ​​are between the maximum value of the characteristic interval corresponding to the diagnostic record of type moderate hardening record and the minimum value of the characteristic interval corresponding to the diagnostic record of type severe hardening record, and calculating the average value of the corresponding comprehensive characteristic values, as the third threshold; The obtained first threshold, second threshold and third threshold are used as feature thresholds; S2-4, construct a hardening grade model, specifically as follows: patients whose diagnostic records correspond to comprehensive characteristic values ​​less than the first threshold are at the first level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the first threshold and less than the second threshold are at the second level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the second threshold and less than the third threshold are at the third level of hardening; patients whose diagnostic records correspond to comprehensive characteristic values ​​greater than or equal to the third threshold are at the fourth level of hardening; The first level hardening, the second level hardening, the third level hardening and the fourth level hardening are regarded as the hardening grades.

4. The method for constructing a data classification model for coronary heart disease atherosclerosis according to claim 3, characterized in that: The step S3 includes the following: S3-1. According to the patient number and diagnosis time in the diagnosis record, all diagnosis records of a patient are screened out and sorted from the earliest to the latest according to the diagnosis time; the average value of the comprehensive characteristic value corresponding to the latter diagnosis record minus the comprehensive characteristic value corresponding to the previous diagnosis record among all adjacent diagnosis records of the patient is calculated as the characteristic change amount of the patient; The comprehensive characteristic value corresponding to each diagnostic record of the patient is input into the sclerosis grade model, and the sclerosis grade of the patient in each diagnostic record is output through the sclerosis grade model; S3-2, if the characteristic change of a patient is less than zero, and the sclerosis level of the patient is the second level sclerosis, the patient is taken as the target patient; S3-3. Obtain the target patient's medication dosage, medication course, diet control, exercise plan, and psychological assistance course from the database as the target patient's treatment record; Take the absolute value of the corresponding feature changes of all target patients, screen out the target patient with the largest absolute value of the corresponding feature changes, and use the treatment record of this patient as the recommended information; The suggested information is used as the suggested information for the second level hardening.

5. The method for constructing a data classification model for coronary heart disease atherosclerosis according to claim 4, characterized in that: The step S4 comprises the following: According to the patient's real-time diagnosis record, the comprehensive characteristic value corresponding to the patient's real-time diagnosis record is calculated and input into the hardening grade model. The hardening grade model outputs the patient's hardening grade. If the patient is at the second hardening grade, the recommendation information is output at the same time.

6. A system for constructing a data classification model for coronary heart disease atherosclerosis, the system being used to implement the method for constructing a data classification model for coronary heart disease atherosclerosis according to any one of claims 1 to 5, characterized in that: The system includes an information collection module, an information processing module and an execution module; The information acquisition module is used to acquire the information required by the system; the information processing module is used to store, analyze and transmit the information of each module; and the execution module is used to execute the information of the information processing module.

7. The system for constructing a data classification model for coronary heart disease atherosclerosis according to claim 6, characterized in that: The information acquisition module includes a diagnosis record unit and a treatment record unit; The diagnosis record unit is used to collect the patient's diagnosis record; the treatment record unit is used to collect the patient's treatment record.

8. The system for constructing a data classification model for coronary heart disease atherosclerosis according to claim 6, characterized in that: The information processing module includes an information storage unit, an information analysis unit and an information transmission unit; The information storage unit is used to store the information acquired by the information acquisition module; the information analysis unit is used to analyze the information acquired by the information acquisition module; and the information transmission unit is used for information transmission of each module in the system.

9. The system for constructing a data classification model for coronary heart disease atherosclerosis according to claim 6, characterized in that: The execution module includes a visualization unit and a pushing unit; The visualization unit is used to form visualization data of the patient's sclerosis level and recommended information; the push unit is used to push the patient's sclerosis level and recommended information.

10. A medium for constructing a data classification model of coronary heart disease atherosclerosis, the system being used to implement the method for constructing a data classification model of coronary heart disease atherosclerosis according to any one of claims 1 to 5, characterized in that: The medium includes a data storage medium for storing data.

Images