Coronary heart disease risk assessment system

Abstract

The invention relates to the technical field of cardiovascular disease risk assessment, and discloses a coronary heart disease risk assessment system, which comprises a data acquisition module for acquiring and processing multi-dimensional data, a representation extraction module for screening coronary heart disease related representations and establishing coronary heart disease progress association rules, a representation judgment module for determining common disease characterization affiliation, and an evaluation module for evaluating the coronary heart disease risk. The risk assessment module removes duplication, eliminates characterization and determines the progress stage of the coronary heart disease, the period planning module determines the risk assessment period of the coronary heart disease, and the result output module outputs the progress stage and the risk assessment period of the coronary heart disease to the terminal. Through multi-module cooperation, co-disease characterization affiliation is judged, the progress stage is determined in combination with the association rule, and accurate determination of the progress stage and the evaluation period of the coronary heart disease and data security control are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cardiovascular disease risk assessment, in particular to a coronary heart disease risk assessment system. BACKGROUND

[0002] With the acceleration of medical informatization, coronary heart disease risk assessment needs to integrate multi-modal data such as electronic medical records, medical images, physiological data and patient self-reported information, and the demand for cross-institutional data interconnection is increasing. However, current data collection lacks standardized interfaces, making it difficult to integrate cross-institutional data. At the same time, the need for accurate division of coronary heart disease progression stages and related feature correlation has increased, and traditional assessment methods are difficult to adapt to multi-dimensional data processing and dynamic risk judgment, and cannot meet the requirements of clinical assessment of scientificity and continuity.

[0003] The traditional coronary heart disease risk assessment system data processing mechanism is imperfect, and the collected data is not subjected to standardized cross-validation, grading and de-identification processing, which can easily lead to low data quality or privacy leakage; the comorbidity feature attribution judgment does not combine the preset specific pathological indicators and disease onset time, which can easily confuse the coronary heart disease and the related features of the previous medical history; the coronary heart disease related features are not reasonably deduplicated as needed, and there is a lack of data access permission control, and the evaluation period cannot be accurately determined based on the coronary heart disease progression stage, which can lead to large evaluation result deviation and insufficient practicality.

[0004] The problems of non-standard data collection and processing, inaccurate comorbidity feature attribution judgment, unreasonable coronary heart disease related feature deduplication, lack of data security control and inaccurate evaluation period determination in the prior art affect the scientificity and reliability of coronary heart disease risk assessment.

[0005] Therefore, the present application provides a coronary heart disease risk assessment system to solve the above problems. SUMMARY

[0006] In view of this, the present application provides a coronary heart disease risk assessment system to solve the problems of non-standard data collection and processing, inaccurate comorbidity feature attribution judgment, unreasonable coronary heart disease related feature deduplication, lack of data security control and inaccurate evaluation period determination in the prior art coronary heart disease risk assessment system.

[0007] In one aspect, the present application provides a coronary heart disease risk assessment system, comprising: a data collection module configured to collect multi-dimensional data of a patient from a plurality of data sources through a standardized interface, and to cross-validate, grade and de-identify the multi-dimensional data by interfacing with a cross-institutional data platform. a characterization extraction module, configured to extract all characterizations from the processed multi-dimensional data and filter out characterizations related to coronary heart disease, and further configured to divide the progression stages of coronary heart disease according to unified clinical standards, and establish the coronary heart disease progression association rules between the progression stages of coronary heart disease and the characterizations related to coronary heart disease; a characterization judgment module, configured to obtain a comorbidity characterization with both coronary heart disease and a risk of previous medical history from the processed multi-dimensional data, and judge the attribution of the comorbidity characterization according to a preset specific pathological index; a risk assessment module, configured to eliminate the characterizations related to coronary heart disease according to the characterizations related to previous medical history of the patient, and further configured to determine the progression stage of coronary heart disease according to the attribution judgment result of the comorbidity characterization and the coronary heart disease progression association rules; if the attribution of the comorbidity characterization is judged to be the characterization related to coronary heart disease, the risk assessment module determines the progression stage of coronary heart disease according to the coronary heart disease progression association rules; if the attribution of the comorbidity characterization is judged to be the characterization related to previous medical history, the risk assessment module eliminates the characterizations related to coronary heart disease, and then determines the progression stage of coronary heart disease according to the coronary heart disease progression association rules; a cycle planning module, configured to determine the risk assessment cycle according to the progression stage of coronary heart disease; a result output module, configured to output the progression stage of coronary heart disease and the risk assessment cycle to a patient terminal.

[0008] Further, when the characterization judgment module judges the attribution of the comorbidity characterization according to the preset specific pathological index: if the comorbidity characterization is accompanied by a coronary heart disease specific pathological index, the characterization judgment module judges that the attribution of the comorbidity characterization is the characterization related to coronary heart disease; if the comorbidity characterization is only accompanied by a previous medical history specific pathological index, the characterization judgment module judges that the attribution of the comorbidity characterization is the characterization related to previous medical history; if the comorbidity characterization has neither a coronary heart disease specific pathological index nor a previous medical history specific pathological index, the characterization judgment module judges the attribution of the comorbidity characterization according to the onset time of the previous medical history of the patient and the first appearance time of the characterization related to coronary heart disease; The preset pathological index verification standard is a standard determined by referring to the diagnostic criteria of coronary heart disease and common previous medical history.

[0009] Further, when the characterization judgment module judges the attribution of the comorbidity characterization according to the onset time of the previous medical history of the patient and the first appearance time of the characterization related to coronary heart disease: if the onset time of the previous medical history is earlier than the first appearance time of the characterization related to coronary heart disease, the characterization judgment module judges that the attribution of the comorbidity characterization is the characterization related to previous medical history; If the onset time of the previous medical history is later than the first appearance time of the coronary heart disease related feature, the feature judgment module determines that the comorbidity feature belongs to the coronary heart disease related feature; If the onset time of the previous medical history is the same as the first appearance time of the coronary heart disease related feature, the feature judgment module collects the duration of the comorbidity feature, and determines the attribution of the comorbidity feature according to the duration and the preset time length.

[0010] Further, when the feature judgment module determines the attribution of the comorbidity feature according to the duration and the preset time length: If the duration is in the first preset time length, the feature judgment module determines that the comorbidity feature belongs to the previous medical history related feature; If the duration is in the second preset time length, the feature judgment module determines that the comorbidity feature belongs to the coronary heart disease related feature; If the duration is not in the preset time length, the feature judgment module marks the comorbidity feature as an observed feature; The first preset time length is determined according to the common duration of the comorbidity feature in the previous medical history, and the second preset time length is determined according to the common duration of the comorbidity feature in the coronary heart disease.

[0011] Further, the coronary heart disease progression correlation rule is: The feature extraction module divides the coronary heart disease related features into early, medium and late stages according to their appearance time in the coronary heart disease progression stage.

[0012] Further, the coronary heart disease progression correlation rule further includes: If the coronary heart disease related feature is a single feature, the feature extraction module associates the coronary heart disease related feature to the corresponding progression stage; If the coronary heart disease related feature is multiple features, the feature extraction module associates the coronary heart disease related feature to the later progression stage in the multiple corresponding progression stages.

[0013] Further, it further includes a feature verification module, which is used to verify the judgment result when the feature judgment module determines that the comorbidity feature belongs to the previous medical history related feature.

[0014] Further, when the feature verification module verifies the judgment result: The feature verification module analyzes the disease type of the previous medical history according to the comorbidity feature, and obtains all features of the disease type; determine whether the comorbidity feature is caused by the previous medical history according to all features of the disease type and all features extracted by the feature extraction module; If all the characteristics of the disease type are covered in all the characteristics, it is determined that the comorbidity characteristic is caused by the past medical history; On the contrary, if all the characteristics of the disease type are not covered in all the characteristics, it is determined that the comorbidity characteristic is not caused by the past medical history, and the comorbidity characteristic is attributed to the coronary heart disease related characteristic.

[0015] Further, an access control module is further included, and the access control module is configured to set a data access right according to a preset post level.

[0016] Further, when the access control module sets the data access right according to the preset post level, the access control module acquires post information input by an access person. Further, when the access control module sets the data access right according to the preset post level, the access control module acquires post information input by an access person. If the post information of the access person is a first type of post, the access control module sets that the access person can access all data of the risk assessment system. If the post information of the access person is a second type of post, the access control module sets that the access person can access data of the risk assessment module and the cycle planning module.

[0017] Compared with the prior art, the present application has the following beneficial effects: The present application guarantees the integrity and accuracy of multi-dimensional data and protects patient data privacy through the standardized interface of the data acquisition module, cross-agency interfacing and cross-validation, and de-identification processing; provides a reliable basis for evaluation by screening coronary heart disease related characteristics and establishing progress association rules through the characteristic extraction module; improves evaluation accuracy by determining the progress stage after combining the risk assessment module and removing the comorbidity characteristic attribution through the characteristic judgment module; determines the evaluation cycle according to the progress stage through the cycle planning module to adapt to clinical needs; and the result output module facilitates the patient to know the situation, thereby improving the scientificity and practicality of coronary heart disease risk assessment as a whole. BRIEF DESCRIPTION OF DRAWINGS

[0018] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not intended to limit the scope of the application. Moreover, the same reference numerals are used throughout the same figures. In the drawings: Figure 1 A functional block diagram of a coronary heart disease risk assessment system according to an embodiment of the present application. DETAILED DESCRIPTION

[0019] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms without being limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thoroughly and completely comprehended, and so that the scope of the present disclosure will be accurately conveyed to those skilled in the art. It should be noted that the embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0020] In the existing coronary heart disease risk assessment system, there are problems such as non-standard data collection and processing, inaccurate comorbidity representation attribution judgment, unreasonable coronary heart disease related representation deduplication, lack of data security control, and inaccurate assessment cycle determination, which affect the scientificity and reliability of coronary heart disease risk assessment. Therefore, the present application provides a coronary heart disease risk assessment system to solve the above problems.

[0021] Reference Figure 1 In some embodiments of the present application, a coronary heart disease risk assessment system comprises a data collection module, a representation extraction module, a representation judgment module, a risk assessment module, a cycle planning module, and a result output module.

[0022] Specifically, the data collection module collects multi-dimensional data of patients from multiple data sources such as medical imaging systems, electronic medical record systems, biological sensors, and patient self-report data platforms through standardized interfaces, and is also used to interface with cross-institutional data platforms to retrieve historical diagnosis and treatment data of patients in different hospitals, such as electrocardiogram in A hospital and blood lipid test report in B hospital, to complete multi-dimensional data, and to perform cross-validation, grading, and de-identification processing on multi-dimensional data.

[0023] It can be understood that the standardized interface is a unified data access interface that meets the industry specifications for medical data exchange, which can adapt to different data sources such as the data formats of medical imaging systems and electronic medical record systems, and solves the problem of incompatible data formats across systems.

[0024] Specifically, the multi-dimensional data is a comprehensive data set, including patient biomarker data such as blood lipids and blood glucose, medical imaging data such as coronary CT, physiological data such as heart rate and blood pressure, lifestyle data such as smoking history, and past medical history data.

[0025] Specifically, the cross-validation process is to compare the patient's self-reported data (such as smoking history) with objective data (such as the amount of activity recorded by the sensor, the record of smoking cessation intervention in the medical record), and eliminate contradictory data; hierarchical processing is to divide data levels according to data types (image data, physiological data, medical history data) and quality (such as image data according to resolution, artifact level, physiological data according to fluctuation range), and preferentially use high-quality data; de-identification processing includes deleting or replacing information in the data that can directly identify the patient's identity, such as name, ID number, and only keeping anonymous identification to avoid patient privacy leakage.

[0026] Specifically, the representation extraction module extracts all representations from the processed multi-dimensional data and filters out coronary heart disease related representations, and also divides the progression stages of coronary heart disease according to unified clinical standards, and establishes the coronary heart disease progression association rules of coronary heart disease progression stages and coronary heart disease related representations.

[0027] Specifically, the representation extraction module extracts all quantifiable or describable representations from the processed multi-dimensional data: such as "blood pressure 140 / 90 mmHg", "chest pain frequency 2 times per week", "coronary CT shows stenosis 30%", "5 years of history of hypertension". And filter out representations related to the pathology of coronary heart disease, such as blood pressure, blood lipids, blood sugar, body mass index, electrocardiogram abnormalities, myocardial enzyme spectrum, etc. Then the filtered coronary heart disease related representations are corresponding to the progression stages of coronary heart disease, forming the coronary heart disease progression association rules. According to the unified clinical standard, the progression stages of coronary heart disease are divided into early, medium and late stages.

[0028] Specifically, the representation judgment module obtains comorbidity representations with both coronary heart disease and past medical history risks from the processed multi-dimensional data, and judges the attribution of the comorbidity representations according to the pre-set specific pathological indicators.

[0029] Specifically, the pre-set specific pathological indicators include coronary heart disease specific pathological indicators and past medical history specific pathological indicators. The coronary heart disease specific pathological indicators include coronary heart disease troponin, coronary stenosis, diastolic blood pressure threshold of hypertension, such as elevated troponin I (cTnI) (myocardial injury marker), coronary angiography showing stenosis ≥50%; past medical history specific pathological indicators include hypertension, diabetes, hyperlipidemia, stroke, etc., such as "non-same-day 3 blood pressure ≥140 / 90 mmHg" for hypertension, "fasting blood glucose ≥7.0 mmol / L" for diabetes.

[0030] Specifically, the risk assessment module eliminates the coronary heart disease related representations according to the patient's past medical history related representations, and determines the progression stage of coronary heart disease according to the attribution judgment result of the comorbidity representations and the coronary heart disease progression association rules: If the comorbidity representation is determined to be a coronary heart disease related representation, the risk assessment module determines the coronary heart disease progression stage according to the coronary heart disease progression association rule. If the comorbidity representation is determined to be a past medical history related representation, the risk assessment module determines the coronary heart disease progression stage according to the coronary heart disease progression association rule after eliminating the coronary heart disease related representation.

[0031] Specifically, the elimination of duplication is as follows: firstly, the coronary heart disease related representation is compared with the past medical history related representation of the patient, and the non-core representation that is repeated is deleted, for example, in the repeated representation of "elevated blood pressure", the record of past history of hypertension is deleted, and only the association information of the coronary heart disease risk factor is retained; secondly, the comorbidity representation attributed to the past medical history related representation in the coronary heart disease related representation is eliminated.

[0032] It can be understood that if the comorbidity representation is attributed to "coronary heart disease related representation", the risk assessment module directly calls the "coronary heart disease progression association rule" established by the representation extraction module, compares the comorbidity representation with the association rule, and determines the current coronary heart disease progression stage of the patient; if the comorbidity representation is attributed to "past medical history related representation", the risk assessment module first completes the elimination of duplication of the coronary heart disease related representation, and then calls the "coronary heart disease progression association rule", compares the coronary heart disease related representation after elimination of duplication with the rule, and then determines the coronary heart disease progression stage of the patient.

[0033] Specifically, the cycle planning module determines the risk assessment cycle according to the coronary heart disease progression stage.

[0034] It can be understood that the risk assessment cycle is the next assessment time.

[0035] Specifically, the risk assessment cycle corresponding to the early stage of the coronary heart disease progression stage is: assessment once every 6 months; the risk assessment cycle corresponding to the middle stage is: assessment once every 3 months; and the risk assessment cycle corresponding to the late stage is: assessment once a month, so as to ensure that high-risk patients are monitored frequently and low-risk patients are not over-assessed.

[0036] Specifically, the result output module outputs the coronary heart disease progression stage and the risk assessment cycle to the patient terminal.

[0037] It can be understood that the result output module outputs the sorted results through the hospital APP, short message, self-service terminal and other terminals accessible by the patient, so as to ensure that the patient knows the coronary heart disease risk state and subsequent monitoring arrangement in time.

[0038] It can be seen that the embodiments of the present application solve the problem of "data island" of traditional systems through standardized interfaces and cross-agency docking, and multi-dimensional data coverage is more comprehensive; cross-validation improves data accuracy, and de-identification processing can balance data quality and privacy security; by screening coronary heart disease related manifestations and establishing "manifestation-progression stage" association rules, the problems of manifestation confusion and subjective stage division in traditional systems are solved, providing standardized basis for evaluation; by distinguishing the attribution of comorbidity manifestations through specific pathological indicators, the confusion between coronary heart disease and previous disease history manifestations is avoided, and the evaluation deviation caused by manifestation misjudgment in traditional systems is reduced; in risk assessment, the progression stage is determined by de-duplication and scene division, ensuring that the evaluation result fits the actual condition of the patient; by combining the progression stage and risk level to set the evaluation period, precise monitoring of "high risk high frequency, low risk low frequency" is realized, avoiding excessive medical treatment or insufficient monitoring; finally, the results are output to the patient terminal, improving the patient's right to know about their own condition and helping patients cooperate with subsequent treatment and monitoring, and overall improving the scientificity, practicality and clinical adaptability of coronary heart disease risk assessment.

[0039] Reference Figure 1 In some embodiments of the present application, when the manifestation judgment module determines the attribution of the comorbidity manifestation according to the pre-set specific pathological indicators: If the comorbidity manifestation is accompanied by coronary heart disease specific pathological indicators, the manifestation judgment module determines that the comorbidity manifestation is attributed to coronary heart disease related manifestations; If the comorbidity manifestation is only accompanied by previous disease history specific pathological indicators, the manifestation judgment module determines that the comorbidity manifestation is attributed to previous disease history related manifestations; If the comorbidity manifestation has neither coronary heart disease specific pathological indicators nor previous disease history specific pathological indicators, the manifestation judgment module determines the attribution of the comorbidity manifestation according to the onset time of the patient's previous disease history and the first appearance time of the coronary heart disease related manifestations; The pre-set pathological indicator verification standard is a standard determined by referring to the diagnostic criteria of coronary heart disease and common previous disease history.

[0040] Specifically, if the comorbidity manifestation is accompanied by coronary heart disease specific pathological indicators, the manifestation judgment module first compares the relevance of the comorbidity manifestation and the coronary heart disease specific pathological indicators: for example, the comorbidity manifestation is "chest pain", and it is checked whether there are coronary heart disease specific pathological indicators (such as troponin I = 0.12 ng / mL, exceeding the verification standard threshold) in the multi-dimensional data; Then confirm the causal relationship between the indicators and the manifestations: determine whether the coronary heart disease specific indicators are directly related to the comorbidity manifestations (such as elevated troponin is a direct cause of myocardial injury leading to chest pain), and exclude cases where "indicators are irrelevant to manifestations" (such as occasional slight elevation of troponin is irrelevant to chest pain); Final determination of attribution: If there is an associated coronary heart disease specific pathological indicator, the representation judgment module will attribute the comorbidity representation (such as chest pain) to "coronary heart disease related representation", and record the judgment basis (such as "chest pain is attributed to coronary heart disease related, based on troponin I = 0.12 ng / mL > 0.04 ng / mL").

[0041] Specifically, when the comorbidity representation is accompanied by a specific pathological indicator of past medical history, the representation judgment module first determines the relevance of the comorbidity representation and the specific pathological indicator of past medical history: for example, the comorbidity representation is "elevated blood pressure", and the specific indicator of past medical history (such as blood pressure = 150 / 95 mmHg, which meets the verification standard of hypertension) is checked in the multidimensional data; Then exclude coronary heart disease specific indicators: confirm that there are no coronary heart disease specific pathological indicators (such as no troponin elevation, no coronary stenosis image) in the multidimensional data, to ensure that there are only past medical history indicators; Finally determine the attribution: if there is only an associated specific pathological indicator of past medical history, the representation judgment module will attribute the comorbidity representation (such as elevated blood pressure) to "past medical history related representation", and record the judgment basis (such as "elevated blood pressure is attributed to hypertension related, based on blood pressure = 150 / 95 mmHg > 140 / 90 mmHg, and no coronary heart disease indicators").

[0042] Specifically, when the comorbidity representation has no pre-set specific pathological indicator, the representation judgment module first determines that there is no specific pathological indicator: after checking the multidimensional data, it is confirmed that there are no coronary heart disease specific pathological indicators and no specific pathological indicators of past medical history (for example, the comorbidity representation is "fatigue", there are no coronary heart disease indicators such as troponin and coronary imaging, and there are no past medical history indicators such as blood pressure and blood sugar); Then retrieve time data: extract the "onset time of patient's past medical history" (such as diagnosed with hypertension in January 2020) and the "first appearance time of coronary heart disease related representation" (such as first recorded dyslipidemia in March 2022) from the electronic medical record data; According to the time comparison and attribution judgment: sort the two times, and the subsequent steps will further determine the attribution based on the "time sequence"; Finally record the state to be judged: temporarily mark the comorbidity representation as "to be further judged", and wait for the time comparison result to output the final attribution.

[0043] It can be understood that when the comorbidity representation has no specific pathological indicator (such as "fatigue", "dizziness" and other non-specific symptoms), the attribution needs to be determined based on the clinical logic of the time sequence of disease occurrence.

[0044] It can be understood that, by the explicit logic of "specific indicators first", the embodiments of the present application replace the judgment mode of the traditional system relying on the subjective experience of doctors (such as classifying as coronary heart disease only based on "chest pain"), reduce misjudgment caused by experience differences, and improve the objectivity of attribution judgment; both the conventional scene of "having indicators" and the special scene of "having no indicators" (such as non-specific manifestations such as fatigue and fatigue) are covered through "time compensation", avoiding the impasse of the traditional system when there are no indicators, and improving the applicability of the system. It meets the clinical diagnosis and treatment specifications and lays a foundation for subsequent risk assessment.

[0045] Reference Figure 1 In some embodiments of the present application, the representation judgment module determines the attribution of the comorbidity representation according to the onset time of the patient's past medical history and the first appearance time of the coronary heart disease related representation: If the onset time of the past medical history is earlier than the first appearance time of the coronary heart disease related representation, the representation judgment module determines that the comorbidity representation is attributed to the past medical history related representation; If the onset time of the past medical history is later than the first appearance time of the coronary heart disease related representation, the representation judgment module determines that the comorbidity representation is attributed to the coronary heart disease related representation; If the onset time of the past medical history is the same as the first appearance time of the coronary heart disease related representation, the representation judgment module collects the duration of the comorbidity representation and determines the attribution of the comorbidity representation according to the duration and the preset time length.

[0046] Specifically, the onset time of the past medical history is the specific time when the patient is clinically diagnosed with a past medical history (such as hypertension, diabetes), which is usually recorded in the diagnosis record of the electronic medical record (such as "diagnosed with hypertension in March 2018"), which is the starting time of the objective existence of the disease. The first appearance time of the coronary heart disease related representation is: the time when the patient's body first detects the coronary heart disease related representation (such as dyslipidemia, chest pain), which can be extracted from the collection record of multidimensional data (such as "first detected LDL-C elevation in May 2020"), reflecting the starting time of the coronary heart disease related pathological change. The duration of the comorbidity representation is: the duration (such as "dizziness from January to March 2023, for a total of 2 months") of the comorbidity representation (such as dizziness, blood sugar fluctuation) from the first appearance to the current evaluation, which embodies the time span characteristics of the representation.

[0047] Specifically, the preset time length is: the duration range of the comorbidity representation commonly seen in coronary heart disease and past medical history based on clinical big data or guidelines, used for auxiliary judgment in the same time scenario.

[0048] It can be seen that the above embodiments provide objective judgment basis for non-specific representations such as fatigue and dizziness through the time dimension when there is no clear indicator for comorbidity representation, covering more clinical scenarios; using the clinical rule of "disease first, cause second" and the pathological correlation of "duration-disease type", the judgment result is in line with the natural process of disease development (such as long-term hypertension appearing before coronary heart disease, and its related representation is more likely to come from hypertension), reducing the misjudgment without basis.

[0049] With reference to Figure 1 In some embodiments of the present application, when the representation judgment module determines the attribution of the comorbidity representation according to the duration and the preset time length: If the duration is within the first preset time length, the representation judgment module determines that the comorbidity representation is attributed to the past medical history related representation; If the duration is within the second preset time length, the representation judgment module determines that the comorbidity representation is attributed to the coronary heart disease related representation; If the duration is not within the preset time length, the representation judgment module marks the comorbidity representation as a to-be-observed representation; The first preset time length is determined according to the common duration of the comorbidity representation in the past medical history, and the second preset time length is determined according to the common duration of the comorbidity representation in coronary heart disease.

[0050] Specifically, the preset time length includes the first preset time length and the second preset time length; the first preset time length is a time interval determined by statistics of the typical duration range of the comorbidity representation in the corresponding past medical history in the clinic, and the second preset time length is a time interval determined by statistics of the typical duration range of the comorbidity representation in coronary heart disease in the clinic, for example: if the comorbidity representation is dizziness, the first preset time length is the typical duration of dizziness caused by hypertension, which is 1-2 weeks; and the second preset time length is the duration of dizziness caused by coronary heart disease, which is several minutes to 1 hour.

[0051] Specifically, when the duration of the comorbidity representation does not fall within the first preset time length or the second preset time length, it needs to be marked as "to be observed" due to the lack of clear time characteristics to support attribution judgment, and after subsequent supplementary collection of the duration change data of the representation (such as rechecking the duration after 1 week), the comorbidity representation is re-executed for attribution judgment, such as dizziness caused by hypertension, which usually lasts for 1-2 weeks (first preset time length), and dizziness caused by coronary heart disease, which usually lasts for several minutes to 1 hour (second preset time length).

[0052] It can be seen that the above embodiment provides an objective judgment standard for non-specific comorbidity manifestations (such as dizziness and fatigue) by comparing the duration with the preset time length when the “past medical history is consistent with the appearance time of the coronary heart disease manifestations”. The embodiment covers more clinical special scenarios. The manifestations that are not within the preset time length are not forcibly attributed, but are marked as “to be observed” and a review mechanism is set, which avoids incorrect conclusions caused by insufficient data and leaves room for re-judgment after subsequent supplementary data, balancing the rigor of judgment and the actual needs of clinical practice.

[0053] With reference to Figure 1 In some embodiments of the present application, the coronary heart disease progression association rule is: The manifestation extraction module divides the coronary heart disease related manifestations into early, medium and late stages according to their appearance time in the coronary heart disease progression stage.

[0054] The coronary heart disease progression association rule further includes: If the coronary heart disease related manifestation is a single manifestation, the manifestation extraction module associates the coronary heart disease related manifestation to the corresponding progression stage; If the coronary heart disease related manifestation is multiple manifestations, the manifestation extraction module associates the coronary heart disease related manifestation to the later progression stage among the multiple corresponding progression stages.

[0055] It can be understood that the pathological development of coronary heart disease has a clear sequence, and different stages of pathological changes will give rise to specific manifestations (such as early metabolic abnormalities, medium functional symptoms, and late severe symptoms). Therefore, the “appearance time of the manifestation” can directly correspond to the progression stage, for example, dyslipidemia (early appearance) corresponds to the early stage, chest pain after exercise (medium appearance) corresponds to the medium stage, and resting chest pain (late appearance) corresponds to the late stage.

[0056] It can be understood that in clinical practice, patients often have multiple manifestations belonging to different stages at the same time (such as early dyslipidemia + medium chest pain after exercise), and the manifestations of the later stage can better reflect the “true severity” of the current condition - the medium coronary stenosis has already affected myocardial blood supply, and the risk is higher than that of early lipid streaks. If only the early manifestations are associated with the stage, the risk of the disease will be underestimated. Therefore, when there are multiple manifestations, the later stage is preferentially associated to ensure that the clinical decision-making pays more attention to the higher risk pathological state.

[0057] It can be seen that the above embodiment has clear rule design (single manifestation directly corresponds, multiple manifestations select later stage), does not require complex calculation or additional parameters, and can be quickly understood and applied by clinical users (such as primary care physicians), thereby reducing the operation threshold of the system and improving the clinical practicability.

[0058] With reference to Figure 1In some embodiments of the present application, a characterization verification module is further included, which is configured to verify the judgment result when the characterization judgment module judges the comorbidity characterization as belonging to the past medical history related characterization.

[0059] Specifically, when the characterization verification module verifies the judgment result: The characterization verification module analyzes the disease type of the past medical history according to the comorbidity characterization, and obtains all characterizations of the disease type; According to all characterizations of the disease type and all characterizations extracted by the characterization extraction module, it is judged whether the comorbidity characterization is caused by the past medical history; If all characterizations of the disease type are covered in all characterizations, it is judged that the comorbidity characterization is caused by the past medical history; On the contrary, if all characterizations of the disease type are not covered in all characterizations, it is judged that the comorbidity characterization is not caused by the past medical history, and the comorbidity characterization is attributed to the coronary heart disease related characterization.

[0060] It can be understood that the pathological changes of any disease (especially chronic diseases) do not only manifest as a single characterization, but exist in the form of "a set of interrelated characterization clusters" - a single characterization may occur accidentally or be misjudged, but the core pathology of the disease will drive multiple typical characterizations to exist simultaneously. For example: "Blood pressure rise" caused by true hypertension is usually accompanied by other typical characterizations of hypertension (such as dizziness, headache); if only "blood pressure rise" exists without other typical characterizations of hypertension, it is more likely to be a manifestation of other diseases (such as early vascular dysfunction of coronary heart disease).

[0061] Therefore, when the comorbidity characterization is initially judged as "belonging to the past medical history related characterization", it is necessary to verify whether "all typical characterizations of the past medical history exist in the characterization set of the patient" to confirm whether the comorbidity characterization is really caused by the past medical history, so as to avoid attribution errors caused by "single characterization misjudgment", forming a closed-loop logic of "initial judgment-secondary verification".

[0062] The specific implementation is as follows: Determine the disease type of the past medical history: the characterization verification module receives the output data of the characterization judgment module, including the comorbidity characterization: blood pressure rise (155 / 95 mmHg), initial attribution: past medical history (hypertension) related characterization The characterization verification module calls the data of the multidimensional database: confirm that Zhang Mou was diagnosed with "essential hypertension" in 2018 through electronic medical record, the disease type is "essential hypertension", and other types such as secondary hypertension are excluded to ensure the accuracy of the disease type.

[0063] The characterization verification module calls the system preset disease-all characterization database to extract all typical characterizations of "essential hypertension" to form a "to-be-verified characterization set": Blood pressure index: systolic pressure ≥ 140 mmHg and / or diastolic pressure ≥ 90 mmHg measured 3 times on different days (or 24-hour average blood pressure ≥ 130 / 80 mmHg by ambulatory blood pressure monitoring); Symptom characterization: dizziness, headache (mostly frontal or occipital pain); Long-term complication-related characterization (if the disease has been present for more than 5 years): mild arteriosclerosis of the fundus oculi (visible on ophthalmic examination).

[0064] The characterization verification module obtains all actual characterizations of Zhang from the characterization extraction module (derived from multi-dimensional data, including biomarkers, symptoms, and imaging data) to form a "patient actual characterization set": Blood pressure index: blood pressure measured at this visit was 155 / 95 mmHg (consistent with the high blood pressure value standard); Symptom characterization: occasional dizziness (Zhang reported "dizziness 2 times per week for 1 hour each time in the past week"), no headache symptoms (reported "never had frontal or occipital pain"); Complication characterization: ophthalmic examination showed "no arteriosclerosis of the fundus oculi" (Zhang has had hypertension for 5 years, which does not meet the long-term complication standard); Other characterizations: occasional chest tightness (more obvious after exercise), LDL-C = 4.3 mmol / L (dyslipidemia), no ST-T changes on electrocardiogram.

[0065] The characterization verification module compares and verifies and makes attribution judgments: The characterization verification module compares the "to-be-verified characterization set" with the "patient actual characterization set" and determines whether it is "fully covered". The results are as follows in Table 1:

[0066] Table 1 As shown in Table 1 above, "headache" in the "to-be-verified characterization set" does not appear in the "patient actual characterization set", and there is no other clinical basis to explain the absence of "headache" (such as Zhang's history of taking painkillers), therefore, the full characterization of essential hypertension is not covered in all patient characterizations, and the verification is not passed.

[0067] The characterization verification module corrects the attribution of the comorbidity characterization "elevated blood pressure" according to the verification conclusion: Original attribution: past hypertension history-related characterization Corrected attribution: coronary heart disease-related characterization Amendment according to record: "The patient Zhang's hypertension all manifestations (including headache) are not fully covered, and the elevated blood pressure is more likely to be related to early vascular dysfunction of coronary heart disease. The revised attribution is coronary heart disease related manifestations" The revised attribution result is synchronized to the risk assessment module by the manifestation test module for subsequent "coronary heart disease related manifestations deduplication" and "progress stage judgment".

[0068] It can be understood that the preset disease-all manifestations database is a structured data set that accurately associates "specific diseases" with "all manifestations (symptoms, signs, biomarkers, etc.) that may occur in that disease" based on authoritative medical evidence.

[0069] As can be seen, the above embodiment adds a secondary verification link after the preliminary judgment of the manifestation judgment module, constructing a closed-loop process of "preliminary attribution-clinical logic verification-result revision", which makes up for the limitations of single judgment logic and makes the attribution result more in line with the true pathological state of the disease.

[0070] Reference Figure 1 In some embodiments of the present application, an access control module is further included, which is configured to set data access permissions according to preset post level.

[0071] Specifically, when the access control module sets data access permissions according to the preset post level: The access control module obtains the post information input by the access person; If the post information of the access person is a first type of post, the access control module sets the access person to access all data of the risk assessment system; If the post information of the access person is a second type of post, the access control module sets the access person to access data of the risk assessment module and the periodic planning module.

[0072] Specifically, the preset post level is a hierarchical system based on the division of medical post functions, reflecting the actual demand degree of different posts for system data in clinical work, which is usually directly related to the title and responsibility range of the hospital (for example, "the first type of post" corresponds to the high permission demand post, and "the second type of post" corresponds to the medium permission demand post).

[0073] The first type of post refers to the post that undertakes core decision-making and system management responsibilities in coronary heart disease risk assessment and treatment, and its work needs to fully master patient data to develop overall plans, for example: chief physician, director of cardiology department, system administrator (responsible for system maintenance).

[0074] The second type of post refers to a post that undertakes an execution or auxiliary role in the risk assessment of coronary heart disease, and the work thereof only needs to obtain assessment results and subsequent planning data, without needing to access original sensitive information, such as a primary physician (executing a diagnosis and treatment plan) and a specialist nurse (tracking a patient assessment period).

[0075] It can be understood that the overall data of the risk assessment system covers all patient-related information and system configuration data stored in the system, including: original multi-dimensional data (such as patient ID number, biomarker detection results, image reports), intermediate processing data (such as comorbidity characterization attribution results, coronary heart disease progression stage judgment basis), and final output data (such as risk level assessment results, system parameter configuration logs).

[0076] The data of the risk assessment module and the cycle planning module only include the output results of the two modules and necessary execution information, such as the early, medium, and late stages of coronary heart disease progression generated by the risk assessment module, and the risk assessment cycle generated by the cycle planning module, such as once every 3 months, without original sensitive data (such as complete medical records of patients, privacy information).

[0077] It can be seen that the above embodiments prevent low-privilege post personnel from accessing sensitive data (such as patient ID numbers and original biomarker detection results) through strict post permission division, reduce the risk of data leakage and misuse, avoid “excessive operation” (such as nurses modifying risk assessment parameters without authorization), strictly match the operation range of each post with clinical responsibilities, and reduce medical errors caused by chaotic permissions (such as developing a treatment plan based on incorrect data).

[0078] It can be seen that the coronary heart disease risk assessment system in the above embodiments, through the whole-process design of “data integration, accurate characterization judgment, disease progression stage, individualized cycle, and permission protection”, ultimately achieves three core goals: for doctors, providing “accurate and comprehensive” disease assessment results to assist in developing accurate treatment plans and improving diagnosis and treatment efficiency; for patients, protecting privacy and safety, clarifying disease and review plans, and improving disease management participation; and for hospitals, optimizing medical resource allocation, reducing misdiagnosis and missed diagnosis rates, and meeting the development trend of “precision medicine” and “smart medicine”.

[0079] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0080] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A coronary heart disease risk assessment system, characterized in that, include: The data acquisition module is used to collect multi-dimensional patient data from multiple data sources through standardized interfaces. It is also used to connect to cross-institutional data platforms and perform cross-validation, grading, and de-identification processing on the multi-dimensional data. The characterization extraction module is used to extract all characters from the processed multidimensional data and filter out coronary heart disease-related characters. It is also used to classify the stages of coronary heart disease progression according to unified clinical standards and to establish the association rules between the stages of coronary heart disease progression and coronary heart disease-related characters. The characterization judgment module is used to obtain comorbid characterizations that combine coronary heart disease and past medical history risk from the processed multi-dimensional data, and to determine the attribution of the comorbid characterizations based on preset specific pathological indicators. The risk assessment module is used to eliminate duplicate coronary artery disease-related characteristics based on the patient's past medical history; it is also used to determine the stage of coronary artery disease progression based on the attribution results of comorbid characteristics and the association rules for coronary artery disease progression. If the comorbidity symptoms are determined to be coronary artery disease-related symptoms, the risk assessment module then determines the stage of coronary artery disease progression based on the coronary artery disease progression association rules. If the comorbidity symptoms are determined to be related to past medical history, the risk assessment module will remove duplicates from the coronary heart disease-related symptoms and determine the stage of coronary heart disease progression according to the coronary heart disease progression association rules. The cycle planning module is used to determine the risk assessment cycle based on the stage of coronary artery disease progression. The results output module is used to output the coronary heart disease progression stage and risk assessment cycle to the patient terminal.

2. The coronary heart disease risk assessment system according to claim 1, characterized in that, When the characterization judgment module determines the attribution of the comorbidity characterization based on preset specific pathological indicators: If comorbid symptoms are accompanied by coronary heart disease-specific pathological indicators, the comorbidity judgment module will determine that the comorbid symptoms are classified as coronary heart disease-related symptoms. If the comorbidity symptoms are only accompanied by pathological indicators specific to past medical history, the comorbidity judgment module will determine that the comorbidity symptoms are classified as symptoms related to past medical history. If the comorbidity symptoms do not have specific pathological indicators for coronary heart disease or specific pathological indicators for past medical history, the comorbidity judgment module determines the attribution of the comorbidity symptoms based on the onset time of the patient's past medical history and the first appearance time of the coronary heart disease-related symptoms. The pre-defined pathological indicator verification criteria were determined by referring to the diagnostic criteria for coronary heart disease and common past medical histories.

3. The coronary heart disease risk assessment system according to claim 2, characterized in that, The characterization judgment module determines the attribution of comorbidity characteristics based on the patient's past medical history, specifically the onset time and the first appearance time of coronary heart disease-related characteristics. If the onset time of the past medical history is earlier than the first appearance time of coronary heart disease-related symptoms, the comorbidity judgment module determines that the comorbidity symptoms are classified as comorbidity symptoms related to the past medical history. If the onset of the previous medical history is later than the first appearance of coronary heart disease-related symptoms, the comorbidity judgment module will determine that the comorbidity symptoms are classified as coronary heart disease-related symptoms. If the onset time of the previous medical history is the same as the first appearance time of the coronary heart disease-related symptoms, the symptom judgment module collects the duration of the comorbid symptoms and determines the attribution of the comorbid symptoms based on the duration and the preset duration.

4. The coronary heart disease risk assessment system according to claim 3, characterized in that, When the characterization determination module determines the attribution of comorbidity characterization based on the duration and a preset duration: If the duration is within the first preset duration, the characterization judgment module determines that the comorbid characterization belongs to the characterization related to past medical history. If the duration is within the second preset duration, the characterization judgment module determines that the comorbid characterization belongs to coronary heart disease-related characterization; If the duration is not within the preset duration, the characterization judgment module will mark the comorbidity characterization as a characterization to be observed; The first preset duration is determined based on the common duration of comorbid symptoms in the patient's medical history, and the second preset duration is determined based on the common duration of comorbid symptoms in coronary heart disease.

5. The coronary heart disease risk assessment system according to claim 4, characterized in that, The association rule for the progression of coronary artery disease is as follows: The characterization extraction module categorizes coronary heart disease-related characteristics into early, middle, and late stages based on their occurrence time during the progression of coronary heart disease.

6. The coronary heart disease risk assessment system according to claim 5, characterized in that, The association rules for the progression of coronary artery disease also include: If the coronary artery disease-related characteristic is a single characteristic, the characteristic extraction module will associate the coronary artery disease-related characteristic with the corresponding progression stage; If there are multiple coronary artery disease-related characteristics, the characteristic extraction module will associate the coronary artery disease-related characteristics with the later progression stage among the multiple corresponding progression stages.

7. A coronary heart disease risk assessment system according to claim 6, characterized in that, It also includes a characterization verification module, which is used to verify the judgment result when the characterization judgment module judges the comorbid characterization as a characterization related to past medical history.

8. The coronary heart disease risk assessment system according to claim 7, characterized in that, When the characterization verification module verifies the judgment result: The characterization testing module analyzes the disease type in the past medical history based on the comorbidity characterization and obtains all characterizations of the disease type. Based on all the characteristics of the disease type and all the characteristics extracted by the characteristic extraction module, determine whether the comorbidity characteristics are caused by past medical history; If all the characteristics of the disease type are included in all characteristics, then the comorbid characteristics are determined to be caused by past medical history. Conversely, if all the manifestations of the disease type are not covered in all manifestations, it is determined that the comorbid manifestations are not caused by past medical history and are classified as coronary heart disease-related manifestations.

9. A coronary heart disease risk assessment system according to claim 8, characterized in that, It also includes an access control module, which is used to set data access permissions according to preset job levels.

10. A coronary heart disease risk assessment system according to claim 9, characterized in that, The access control module is used to set data access permissions according to preset job levels: The access control module obtains the job information input by the visitor; If the visitor's job information is a Category 1 job, the access control module will set the visitor to access all data in the risk assessment system; If the visitor's job information is a second-category job, the access control module will set the data that the visitor can access in the risk assessment module and the cycle planning module.

Images