A Cardiovascular and Cerebrovascular Disease Screening System and Method Based on Collaborative Data Acquisition

Abstract

This invention discloses a cardiovascular and cerebrovascular disease screening system and method based on collaborative data acquisition, belonging to the field of screening data technology. It addresses the problem of trigger logic separation by retrieving family history information from electronic health records, performing initial screening according to preset screening rules, marking target users and passing them into the screening assessment mechanism, while generating risk markers directly for non-target users. For target users, carotid artery plaque ultrasound imaging and hemodynamic monitoring are performed, plaque density parameters are calculated, and hemodynamic data are fused to form user status feature values. Users classified as abnormal undergo blood sample collection via a biometric calibration module to detect erythrocyte distribution width and brain natriuretic peptide (BNP) levels, calculate erythrocyte mutation rate, and combine it with BNP to generate a calibration ratio. The status feature values ​​are then corrected, forming corrected feature values ​​that are compared with a risk threshold to generate risk markers, improving the accuracy and sensitivity of risk screening and reducing false negative and false positive rates.

Description

Technical Field

[0001] This invention relates to the field of screening data technology, and more specifically, to a cardiovascular and cerebrovascular disease screening system and method based on collaborative data acquisition. Background Technology

[0002] With the continuous improvement of health informatization, electronic health records have gradually become an important data carrier reflecting individual health factors. In many chronic disease-related research and applications, how to use the basic data in electronic health records, combined with physiological parameters and biological indicators, to achieve characteristic modeling and status classification of individual health status has become one of the key directions in the current field of health data analysis.

[0003] The existing technology has the following shortcomings:

[0004] Currently, the triggering conditions and collaborative mechanisms for the collection of various types of health data are not yet perfect. In particular, there is a lack of structured linkage between family health information and current physiological parameters. The collaborative processing mechanism for hemodynamic features and structural parameters is not clear enough, and a stable sequence of feature parameters has not been formed for subsequent state modeling. This results in inconsistent attribute levels and separation of triggering logic in the health data collection stage. Therefore, a cardiovascular and cerebrovascular disease screening system and method based on collaborative collection is proposed.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a cardiovascular and cerebrovascular disease screening system and method based on collaborative acquisition. This system and method employ a multi-parameter feature extraction scheme that combines a family information-driven triggering mechanism, a structure-fluid collaborative modeling method, and a biomarker calibration strategy to address the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for screening cardiovascular and cerebrovascular diseases based on collaborative data acquisition, comprising the following steps:

[0008] Step S1: Retrieve the family history information of the users to be screened from the electronic health records, set family history screening rules, screen the users to be screened based on the family history information and mark the target users, and put the target users into the screening assessment mechanism.

[0009] Step S2: After entering the screening and assessment mechanism, collect the carotid plaque data and hemodynamic data of the target user, calculate the carotid plaque density using the carotid plaque data, generate user status characteristics by combining the carotid hemodynamic data, and classify the target user according to the user status characteristics.

[0010] Step S3: For target users whose classification results are abnormal, detect their blood samples to obtain red blood cell distribution width and brain natriuretic hormone content, calculate the red blood cell mutation rate based on the red blood cell distribution width, and generate a calibration ratio based on the brain natriuretic hormone content.

[0011] Step S4: Correct the user status characteristics of abnormal states using the calibration ratio, and mark the target users as risk based on the correction results.

[0012] In a preferred embodiment, in step S1, the family history information of the user to be screened is retrieved from the electronic health record. The family history information includes the type of direct blood relatives and abnormal characteristic events.

[0013] The family history screening rules include: setting different relationship weight coefficients for different types of direct kinship relationships, classifying different abnormal characteristic events into event levels and setting event level factors, and using the product of the relationship weight coefficient and the event level factor as the kinship characteristic score.

[0014] In a preferred embodiment, in step S1, the scores of all relatives who have a direct blood relationship with the user to be screened are accumulated, and the accumulated result is used as the comprehensive family history score.

[0015] The family history score is compared with a preset comprehensive score threshold to filter users to be screened.

[0016] If the family history score exceeds the preset comprehensive score threshold, the users to be screened will be marked as risk, and the risk-marked users to be screened will undergo key cardiac examinations.

[0017] Conversely, users to be screened are marked as target users, and these target users are included in the screening and evaluation mechanism.

[0018] In a preferred embodiment, in step S2, after the screening and assessment mechanism is introduced, the carotid artery of the target user is scanned by an ultrasound imaging device to obtain the number of plaques in the carotid artery and the area of ​​each plaque.

[0019] The number of plaques in the carotid artery and the area of ​​each plaque are used as plaque data for the carotid artery;

[0020] Blood flow velocity in the carotid artery plaque area of ​​the target user is monitored by Doppler ultrasound, and the maximum blood flow velocity is compared with a preset blood flow velocity threshold.

[0021] If the ratio of the maximum blood flow velocity to the preset blood flow velocity threshold is less than 1, then the ratio result is used as the hemodynamic data of the carotid artery; otherwise, the hemodynamic data of the carotid artery is set to 1.

[0022] In a preferred embodiment, in step S2, the carotid plaque density is calculated using carotid plaque data;

[0023] Carotid plaque density and carotid hemodynamic data are used to generate user status features through a weighted algorithm;

[0024] User status characteristics are compared with preset feature thresholds to classify target users:

[0025] If the user's status characteristics are greater than the preset characteristic threshold, the target user will be classified as an abnormal state.

[0026] Conversely, the target user is classified as normal.

[0027] In a preferred embodiment, in step S3, for target users whose classification results are abnormal, the red blood cell volume in blood samples is detected one by one, and the red blood cell volume dataset is merged to calculate the average red blood cell volume.

[0028] Calculate the standard deviation of red blood cell volume distribution based on mean corpuscular volume;

[0029] The ratio of the mean erythrocyte volume to the standard deviation of the erythrocyte volume distribution is used as the erythrocyte distribution width.

[0030] In a preferred embodiment, in step S3, the red blood cell mutation rate is calculated using a Gaussian distribution deviation function based on the red blood cell volume dataset and the mean and standard deviation of the red blood cell distribution width.

[0031] The brain natriuretic peptide (BNP) content in blood samples was measured and subjected to natural logarithmic transformation and normalization to obtain the BNP influence factor.

[0032] The calibration ratio was calculated using the hyperbolic tangent function based on the erythrocyte mutation rate and the influence factor of brain natriuretic peptide.

[0033] In a preferred embodiment, in step S4, the user status characteristics of the target user in the abnormal state are multiplied by 1 and the sum of the calibration ratio to obtain the corrected user status characteristics.

[0034] When the corrected user status feature value is greater than or equal to the risk threshold, the target user is marked for risk and subjected to key cardiac detection.

[0035] If the corrected user status characteristic value is lower than the risk threshold, the target user will not be marked as a risk.

[0036] A cardiovascular and cerebrovascular disease screening system based on collaborative data acquisition includes a family history screening module, a feature generation module, an indicator calibration module, and a risk assessment module.

[0037] The functions of each module are as follows:

[0038] The family history screening module is used to retrieve the family history information of the users to be screened from the electronic health records, set family history screening rules, screen the users to be screened based on the family history information, mark the target users, and put the target users into the screening assessment mechanism.

[0039] The feature generation module collects carotid plaque data and hemodynamic data of target users who are included in the screening and assessment mechanism. It calculates carotid plaque density using the carotid plaque data, generates user status features by combining the carotid hemodynamic data, and classifies target users according to the user status features.

[0040] The indicator calibration module targets users whose classification results are abnormal by detecting their blood samples to obtain red blood cell distribution width and brain natriuretic peptide (BNP) content. It calculates the red blood cell mutation rate based on the red blood cell distribution width and generates a calibration ratio based on the BNP content.

[0041] The risk assessment module uses a calibration ratio to correct the user status characteristics of abnormal states, and marks the target users as risky based on the correction results.

[0042] The technical effects and advantages of this invention are as follows:

[0043] This invention retrieves family history information of users to be screened from electronic health records, conducts preliminary screening based on preset family history screening rules, and marks target users. Target users are then included in the screening assessment mechanism, while non-target users are directly subjected to risk labeling. Carotid artery plaque ultrasound imaging and hemodynamic monitoring are performed on target users to calculate carotid artery plaque density parameters. These parameters are then fused with the carotid artery hemodynamic data to form user status feature values, used for preliminary classification of the risk status of target users. For target users with abnormal classification results, a biometric calibration module detects their blood samples, collecting erythrocyte distribution width and brain natriuretic peptide (BNP) levels. Based on the erythrocyte distribution width, the erythrocyte mutation rate is calculated, and a calibration ratio is generated based on the BNP levels. The calibration ratio is used to correct the user status feature values, forming corrected status feature values, which are then compared with preset risk thresholds to generate risk labels. By integrating family genetic information, imaging data, hemodynamic indicators, and biochemical markers, multi-dimensional dynamic risk assessment is achieved, which can improve the accuracy and sensitivity of risk screening and reduce the false negative and false positive rates. Attached Figure Description

[0044] Figure 1 This is a flowchart of a cardiovascular and cerebrovascular disease screening method based on collaborative data acquisition according to the present invention.

[0045] Figure 2 This is a schematic diagram of a cardiovascular and cerebrovascular disease screening system based on collaborative data acquisition according to the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] This invention retrieves family history information of users to be screened from electronic health records, conducts preliminary screening based on preset family history screening rules, and marks target users. Target users are then included in the screening assessment mechanism, while non-target users are directly subjected to risk labeling. Carotid artery plaque ultrasound imaging and hemodynamic monitoring are performed on target users to calculate carotid artery plaque density parameters. These parameters are then fused with the carotid artery hemodynamic data to form user status feature values, used for preliminary classification of the risk status of target users. For target users with abnormal classification results, a biometric calibration module detects their blood samples, collecting erythrocyte distribution width and brain natriuretic peptide (BNP) levels. Based on the erythrocyte distribution width, the erythrocyte mutation rate is calculated, and a calibration ratio is generated based on the BNP levels. The calibration ratio is used to correct the user status feature values, forming corrected status feature values, which are then compared with preset risk thresholds to generate risk labels. By integrating family genetic information, imaging data, hemodynamic indicators, and biochemical markers, multi-dimensional dynamic risk assessment is achieved, which can improve the accuracy and sensitivity of risk screening and reduce the false negative and false positive rates.

[0048] Example 1: A method for screening cardiovascular and cerebrovascular diseases based on collaborative data acquisition, such as... Figure 1 As shown, it includes the following steps:

[0049] Step S1: Retrieve the family history information of the users to be screened from the electronic health records, set family history screening rules, screen the users to be screened based on the family history information and mark the target users, and put the target users into the screening assessment mechanism.

[0050] Step S2: After entering the screening and assessment mechanism, collect the carotid plaque data and hemodynamic data of the target user, calculate the carotid plaque density using the carotid plaque data, generate user status characteristics by combining the carotid hemodynamic data, and classify the target user according to the user status characteristics.

[0051] Step S3: For target users whose classification results are abnormal, detect their blood samples to obtain red blood cell distribution width and brain natriuretic hormone content, calculate the red blood cell mutation rate based on the red blood cell distribution width, and generate a calibration ratio based on the brain natriuretic hormone content.

[0052] Step S4: Correct the user status characteristics of abnormal states using the calibration ratio, and mark the target users as risk based on the correction results.

[0053] The specific implementation is as follows:

[0054] In step S1, the family history information of the user to be screened is retrieved from the electronic health record. The family history information includes the types of relatives with direct blood relations and abnormal characteristic events.

[0055] The family history screening rule is used to determine whether a user being screened is a target user. The family history screening rule is set as follows:

[0056] Different relationship weight coefficients are preset for different types of direct kinship relationships. For example, the relationship weight coefficient for parent-child relationships is set to 1, and the relationship weight coefficient for sibling relationships is set to 0.8. The specific relationship weight coefficients are set by professionals.

[0057] Different abnormal events are classified into event levels and event level factors are preset. For example, abnormal events are classified into medium-characteristic types and the event level factor for medium-characteristic types is set to 0.7. The specific classification of event levels and the preset event level factors are set by professionals.

[0058] The product of the relationship weight coefficient and the event level factor is used as the kinship feature score.

[0059] The scores of all relatives with direct blood ties to the user to be screened are summed up, and the sum is used as the comprehensive family history score.

[0060] The overall score reflects the total intensity of abnormal characteristic events among the relatives of the user to be screened by summing up the kinship characteristic scores;

[0061] The family history score is compared with a preset comprehensive score threshold to filter users to be screened.

[0062] If the family history score exceeds the preset comprehensive score threshold, the users to be screened will be marked as risk, and the risk-marked users to be screened will undergo key cardiac examinations.

[0063] Conversely, users to be screened are marked as target users, and these target users are included in the screening and evaluation mechanism.

[0064] It should be explained that the screening and assessment mechanism refers to the set of procedures for further identifying the risk status of target users in this embodiment. Specifically, it includes all the operations described in step S2, namely, collecting carotid plaque data and hemodynamic data of target users, calculating carotid plaque density based on carotid plaque data, generating user status feature values ​​in combination with hemodynamic data, and classifying target users according to these status feature values, thereby improving screening efficiency, reducing resource consumption, and avoiding invalid detection.

[0065] By accessing the family history information of users to be screened in electronic health records and setting family history screening rules, it is possible to pre-screen and mark users with high-risk family backgrounds, identify potential risk users in advance, improve the accuracy of screening, reduce resource waste, and improve the system's screening efficiency and tiered intervention capabilities by including target users in the subsequent evaluation mechanism.

[0066] It should be noted that electronic health records refer to a database of health information related to users stored on a digital platform, including the user's basic identity information and family history information related to their direct blood relatives; the preset comprehensive scoring threshold is the critical value used to determine the comprehensive score of family history, which can be adjusted according to the actual application scenario and screening needs, and is set by professionals, which will not be elaborated here.

[0067] In step S2, after the screening and assessment mechanism is entered, the carotid artery of the target user is scanned by an ultrasound imaging device to obtain the number of plaques in the carotid artery and the area of ​​each plaque. The number of plaques in the carotid artery and the area of ​​each plaque are used as the plaque data of the carotid artery.

[0068] The blood flow velocity in the carotid artery plaque area of ​​the target user is monitored by Doppler ultrasound. The maximum value of the blood flow velocity is compared with a preset blood flow velocity threshold. If the ratio of the maximum value of the blood flow velocity to the preset blood flow velocity threshold is less than 1, the ratio result is used as the hemodynamic data of the carotid artery; otherwise, the value of the hemodynamic data of the carotid artery is set to 1.

[0069] Carotid hemodynamic data reflects the blood flow velocity in the narrowed area or blockage point of the carotid artery. An abnormally high blood flow velocity indicates local narrowing of the carotid artery.

[0070] Calculate carotid plaque density using carotid plaque data: Where m is the number of patches. Let j be the area of ​​the j-th patch. This represents the projected area of ​​the carotid artery. Carotid plaque density;

[0071] Among them, the carotid artery projection area refers to the area of ​​the image region of the carotid artery wall and lumen obtained by scanning the carotid artery with an ultrasound imaging device to obtain a two-dimensional image.

[0072] Carotid plaque density is the proportion of plaque within the lumen of the carotid artery, reflecting the degree of hardening of the carotid artery wall.

[0073] It should be explained that the preset blood flow velocity threshold is used to indicate that if the blood flow velocity exceeds the preset blood flow velocity threshold, it indicates that the blood flow velocity is abnormal. The specific setting should be done by professionals.

[0074] Carotid plaque density and carotid hemodynamic data are used to generate user state features through a weighted algorithm: ,in, As a preset weighting factor, Carotid plaque density, For carotid artery hemodynamic data, User status characteristics;

[0075] The larger the user status characteristics, the more the function and structure of the target user's carotid artery deviates from the normal physiological range, and the higher the potential risk level.

[0076] User status characteristics are compared with preset feature thresholds to classify target users:

[0077] If the user's status characteristics are greater than the preset characteristic threshold, the target user will be classified as an abnormal state; otherwise, the target user will be classified as a normal state.

[0078] After target users are included in the screening and assessment mechanism, plaque data and hemodynamic data of their carotid arteries are collected and analyzed jointly from two dimensions: vascular structural integrity and blood flow function. This allows for the construction of user status feature values ​​and classification, effectively compensating for the one-sidedness of screening based on a single structural or hemodynamic indicator, improving the sensitivity of identifying carotid artery abnormalities, reducing the false positive rate, and providing accurate stratification basis for subsequent more detailed detection or intervention.

[0079] It should be noted that ultrasound imaging devices utilize the principle of high-frequency sound waves propagating and reflecting in human tissues. They emit ultrasound waves through a probe and receive the reflected signals, which are then processed to generate two-dimensional or three-dimensional images of the carotid artery and its internal structures. This allows for the acquisition of the number and area of ​​plaques in the carotid artery. Doppler ultrasound is an ultrasound technology based on the Doppler effect, which monitors blood flow velocity by detecting changes in ultrasound frequency in response to blood flow. Preset feature thresholds are set by professionals based on their experience and actual needs. Preset weighting factors refer to the numerical coefficients assigned to carotid artery plaque density during the process of fusing carotid artery plaque density and hemodynamic data using a weighted algorithm to generate user status feature values. These values ​​are pre-set by professionals according to actual screening needs.

[0080] In step S3, for target users whose classification results are abnormal, blood sample testing and data analysis are performed. Peripheral venous blood samples are collected from the target user, and the red blood cell distribution width is detected using a fully automated blood analyzer. The level of brain natriuretic peptide (BNP) is also detected using immunofluorescence analysis.

[0081] Red blood cell distribution width reflects the heterogeneity of red blood cell volume distribution in peripheral blood, quantitatively describes the degree of variability in red blood cell volume distribution, and is calculated based on the mean red blood cell volume and distribution standard deviation.

[0082] The red blood cell volume in the blood samples of target users with abnormal classification results is detected one by one using a fully automated blood analyzer to obtain a red blood cell volume dataset. Based on this dataset, the mean corpuscular volume (MCV) is calculated using the following formula:

[0083] ;

[0084] in, Mean corpuscular volume (MCV) Let represent the volume of the i-th red blood cell, where i is the index value of the red blood cells in the blood sample, and n is the total number of red blood cells in the blood sample.

[0085] Based on the mean corpuscular volume (MCV), the standard deviation of the MCV distribution is further calculated using the following formula:

[0086] ;

[0087] in, This represents the standard deviation of the red blood cell volume distribution.

[0088] The red blood cell distribution width is calculated based on the mean corpuscular volume and standard deviation of red blood cell distribution. The formula for calculation is as follows:

[0089] ;

[0090] in, The red blood cell distribution width reflects the dispersion of red blood cell volume.

[0091] Red blood cell distribution width (RBV) and brain natriuretic peptide (BNP) levels of healthy reference populations were extracted from the central health checkup database. Based on the red blood cell volume dataset and the mean and standard deviation of RBV from the healthy reference population, a Gaussian distribution deviation function was introduced to calculate the RBV variation rate as a derived higher-order indicator of RBV. The calculation formula is as follows:

[0092] ;

[0093] in, For red blood cell mutation rate, This represents the mean red blood cell distribution width in a healthy reference population. The standard deviation of red blood cell distribution width for a healthy reference population.

[0094] The Gaussian distribution deviation function, combined with the normalized sum of squared deviations of red blood cell volume and reference value, amplifies the contribution of extreme abnormal red blood cell populations, thereby enhancing the sensitivity to anomalies.

[0095] Simultaneously, the brain natriuretic peptide (BNP) levels in blood samples from target users were determined using immunofluorescence analysis. To reduce the influence of the nonlinear distribution of BNP levels, a natural logarithmic transformation was performed, followed by normalization using BNP levels from a healthy reference population. The resulting BNP influence factor was calculated using the following formula:

[0096] ;

[0097] in, Factors affecting brain natriuretic peptides, The content of brain natriuretic peptide hormone, and These represent the mean and standard deviation of the natural logarithmic values ​​of brain natriuretic peptide (BNP) levels in a healthy reference population.

[0098] Based on the influence factors of erythrocyte variability and brain natriuretic peptide, the calibration ratio is calculated using the hyperbolic tangent function. The calculation formula is as follows:

[0099] ;

[0100] in, For calibration ratio, These are the weighting coefficients corresponding to the influencing factors of erythrocyte mutation rate and brain natriuretic peptide, respectively. Interactive weights are used to capture the amplifying effect of synergistic abnormalities in erythrocyte mutation rate and brain natriuretic peptide factors on risk. For bias terms, and It was obtained through training with large-scale sample data, which will not be elaborated here.

[0101] The hyperbolic tangent function makes the calibration ratio range (-1, 1), and the larger the value, the more abnormal the target user's state.

[0102] It should be noted that the fully automated blood analyzer is an in vitro diagnostic device based on electrical impedance tomography, light scattering, and flow cytometry technologies, used for multi-parameter detection and analysis of blood cells in collected peripheral venous blood samples; immunofluorescence analysis is an in vitro diagnostic technology based on antigen-antibody specific binding reaction, combined with fluorescence signal detection to achieve quantitative detection of targeted analytes; the Gaussian distribution deviation function is a mathematical function used to characterize the degree of statistical deviation of the test data set relative to the distribution of the healthy reference population; the central health checkup database is a standardized health data set jointly established by national and regional health management institutions, collected through multi-center collaborative efforts, and contains multi-dimensional test data from the healthy checkup population.

[0103] In step S4, based on the calibration ratio obtained in step S3 The user status characteristics of target users in abnormal states are corrected to obtain corrected user status characteristics. These user status characteristics are a comprehensive index calculated by fusing carotid plaque density and carotid hemodynamic data, used to characterize the target user's cardiovascular and cerebrovascular disease risk level. The correction process employs a linear calibration algorithm, and the specific calculation formula is as follows:

[0104] ;

[0105] in, For the corrected user state characteristics, These are user status feature values.

[0106] In this formula, the user state feature value is multiplied by... To achieve proportional adjustment, so that the user state feature value is within... It is enhanced in time, This weakens the risk assessment error and thus corrects it.

[0107] After the correction is completed, the corrected user status characteristics will be... With preset risk threshold The comparison is performed, and when the corrected user status feature value is greater than or equal to the risk threshold, the target user is marked for risk and subjected to key heart detection.

[0108] If the corrected user status characteristic value is lower than the risk threshold, the target user will not be marked as a risk.

[0109] The above process dynamically adjusts the original user status characteristics based on the calibration ratio derived from biomarkers, effectively reducing the probability of misjudgment of risks caused by early data acquisition errors or feature extraction deviations, while enhancing the sensitivity and specificity of the overall risk screening system.

[0110] It should be noted that the risk threshold is a user status characteristic value boundary derived by professionals based on historical big data analysis and clinical validation studies. It is used to distinguish between high-risk and low-risk status of target users to ensure that the sensitivity and specificity of risk screening are optimally balanced. This will not be elaborated on here.

[0111] Example 2: A cardiovascular and cerebrovascular disease screening system based on collaborative data acquisition, such as... Figure 2 As shown, the system includes a family history screening module, a feature generation module, an indicator calibration module, and a risk assessment module. The functions of each module are as follows:

[0112] The family history screening module is used to retrieve the family history information of the users to be screened from the electronic health records, set family history screening rules, screen the users to be screened based on the family history information, mark the target users, and put the target users into the screening assessment mechanism.

[0113] The feature generation module collects carotid plaque data and hemodynamic data of target users who are included in the screening and assessment mechanism. It calculates carotid plaque density using the carotid plaque data, generates user status features by combining the carotid hemodynamic data, and classifies target users according to the user status features.

[0114] The indicator calibration module targets users whose classification results are abnormal by detecting their blood samples to obtain red blood cell distribution width and brain natriuretic peptide (BNP) content. It calculates the red blood cell mutation rate based on the red blood cell distribution width and generates a calibration ratio based on the BNP content.

[0115] The risk assessment module uses a calibration ratio to correct the user status characteristics of abnormal states, and marks the target users as risky based on the correction results.

[0116] Finally, it should be noted that in this paper, relational terms such as first and second are used only 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.

[0117] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0118] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

Claims

1. A method for screening cardiovascular and cerebrovascular diseases based on collaborative data acquisition, characterized in that: Includes the following steps: Step S1: Retrieve the family history information of the users to be screened from the electronic health records, set family history screening rules, screen the users to be screened based on the family history information and mark the target users, and put the target users into the screening assessment mechanism. In step S1, the family history information of the user to be screened is retrieved from the electronic health record. The family history information includes the types of relatives with direct blood relations and abnormal characteristic events. The family history screening rules include: setting different relationship weight coefficients for different types of direct kinship relationships, classifying different abnormal characteristic events into event levels and setting event level factors, and using the product of the relationship weight coefficient and the event level factor as the kinship characteristic score. In step S1, the characteristic scores of all relatives who have a direct blood relationship with the user to be screened are summed up, and the summed result is used as the comprehensive family history score. The family history score is compared with a preset comprehensive score threshold to filter users to be screened. If the family history score exceeds the preset comprehensive score threshold, the users to be screened will be marked as risk, and the risk-marked users to be screened will undergo key cardiac examinations. Conversely, users to be screened are marked as target users, and these target users are included in the screening and evaluation mechanism. Step S2: After entering the screening and assessment mechanism, collect the carotid plaque data and hemodynamic data of the target user, calculate the carotid plaque density using the carotid plaque data, generate user status characteristics by combining the carotid hemodynamic data, and classify the target user according to the user status characteristics. In step S2, after the screening and assessment mechanism is entered, the carotid artery of the target user is scanned by an ultrasound imaging device to obtain the number of plaques in the carotid artery and the area of ​​each plaque. The number of plaques in the carotid artery and the area of ​​each plaque are used as plaque data for the carotid artery; Blood flow velocity in the carotid artery plaque area of ​​the target user is monitored by Doppler ultrasound, and the maximum blood flow velocity is compared with a preset blood flow velocity threshold. If the ratio of the maximum blood flow velocity to the preset blood flow velocity threshold is less than 1, then the ratio result is used as the hemodynamic data of the carotid artery; otherwise, the hemodynamic data of the carotid artery is set to 1. Calculate carotid plaque density using carotid plaque data: Where m is the number of patches. Let j be the area of ​​the j-th patch. This represents the projected area of ​​the carotid artery. Carotid plaque density; Carotid plaque density and carotid hemodynamic data are used to generate user state features through a weighted algorithm: ,in, As a preset weighting factor, Carotid plaque density, For carotid artery hemodynamic data, User status characteristics; User status characteristics are compared with preset feature thresholds to classify target users: If the user's status characteristics are greater than the preset characteristic threshold, the target user will be classified as an abnormal state. Conversely, the target user is classified as normal. Step S3: For target users whose classification results are abnormal, detect their blood samples to obtain red blood cell distribution width and brain natriuretic hormone content, calculate the red blood cell mutation rate based on the red blood cell distribution width, and generate a calibration ratio based on the brain natriuretic hormone content. In step S3, for target users whose classification results are abnormal, the red blood cell volume in each blood sample is detected, and the red blood cell volume dataset is merged to calculate the average red blood cell volume. Calculate the standard deviation of red blood cell volume distribution based on mean corpuscular volume; The ratio of mean erythrocyte volume to the standard deviation of erythrocyte volume distribution is used as the erythrocyte distribution width. Based on the red blood cell volume dataset and the mean and standard deviation of red blood cell distribution width, the red blood cell mutation rate is calculated using the Gaussian distribution deviation function. The brain natriuretic peptide (BNP) content in blood samples was measured and subjected to natural logarithmic transformation and normalization to obtain the BNP influence factor. Based on the influence factors of erythrocyte variability and brain natriuretic peptide, the calibration ratio is calculated using the hyperbolic tangent function. The calculation formula is as follows: ； in, For red blood cell mutation rate, Factors affecting brain natriuretic peptides, For calibration ratio, These are the weighting coefficients corresponding to the influencing factors of erythrocyte mutation rate and brain natriuretic peptide, respectively. For interaction weights, For bias terms; Step S4: Correct the user status characteristics of abnormal states using the calibration ratio, and mark the target users as risky based on the correction results; In step S4, the user status characteristics of the target user in the abnormal state are multiplied by 1 and the sum of the calibration ratio to obtain the corrected user status characteristics. When the corrected user status feature value is greater than or equal to the risk threshold, the target user is marked for risk and subjected to key cardiac detection. If the corrected user status characteristic value is lower than the risk threshold, the target user will not be marked as a risk.

2. A cardiovascular and cerebrovascular disease screening system based on collaborative acquisition, used to implement the cardiovascular and cerebrovascular disease screening method based on collaborative acquisition as described in claim 1, characterized in that: It includes a family history screening module, a feature generation module, an indicator calibration module, and a risk assessment module; The functions of each module are as follows: The family history screening module is used to retrieve the family history information of the users to be screened from the electronic health records, set family history screening rules, combine the family history information to screen the users to be screened and mark the target users, and put the target users into the screening assessment mechanism. After the feature generation module enters the screening and evaluation mechanism, it collects carotid plaque data and hemodynamic data of the target user, calculates carotid plaque density using the carotid plaque data, generates user status features by combining the carotid hemodynamic data, and classifies the target user according to the user status features. The indicator calibration module targets users whose classification results are abnormal by detecting their blood samples to obtain red blood cell distribution width and brain natriuretic peptide hormone content. It calculates the red blood cell mutation rate based on the red blood cell distribution width and generates a calibration ratio based on the brain natriuretic peptide hormone content. The risk assessment module uses a calibration ratio to correct the user status characteristics of abnormal states, and marks the target users as risky based on the correction results.

Images