CN116616790B - Cardiac risk assessment method, apparatus, computer device and storage medium - Google Patents

Abstract

This application relates to a cardiac risk assessment method, device, computer equipment and storage medium. The method includes: obtaining ECG data corresponding to target ECG detection that matches clinical indications; the clinical indications are used to indicate Whether stress exercise ECG detection can be performed; the target ECG detection at least includes resting ECG detection; and the high-frequency components of the QRS wave group in the ECG data are analyzed according to the target ECG detection to obtain the corresponding reference features. ; Determine the heart risk assessment level and heart risk type according to the reference characteristics; the heart risk assessment level is used as a reference index for shunting the subject; the heart risk type is used to evaluate the possible existence of the subject's heart Reference indicators for risk types. Using this method can improve the accuracy of identifying cardiac health status in a non-invasive way.

Description

Cardiac risk assessment method, device, computer equipment and storage medium

Technical Field

The present application relates to the technical field of electrocardiogram data processing, and in particular to a cardiac risk assessment method, apparatus, computer equipment and storage medium.

Background Art

At present, cardiac activity-related information is usually analyzed based on the ST-T segment data that represents the cardiac repolarization phase in the electrocardiogram (ECG) to achieve cardiac risk assessment. However, many potential heart problems do not show any abnormalities in the ST-T segment data, resulting in low accuracy in cardiac risk assessment. If it is necessary to more accurately assess cardiac risk and thus more effectively identify heart health conditions, it needs to be achieved through invasive methods such as coronary angiography and myocardial biopsy. Invasive assessment methods will have more or less impact on the physical health of the subject.

Summary of the invention

Based on this, it is necessary to provide a cardiac risk assessment method, device, computer equipment and storage medium that can improve the accuracy of identifying cardiac health status in a non-invasive manner to address the above technical problems.

A method for assessing cardiac risk, the method comprising:

Acquire ECG data corresponding to a target ECG test that matches a clinical indication; the clinical indication is used to indicate whether a load exercise ECG test can be performed; the target ECG test at least includes a resting ECG test;

Analyzing the high-frequency components of the QRS complex in the electrocardiogram data according to the target electrocardiogram detection to obtain corresponding reference features;

The cardiac risk assessment level and the cardiac risk type are determined based on the reference characteristics; the cardiac risk assessment level is used as a reference indicator for triaging the subject; and the cardiac risk type is used as a reference indicator for assessing the possible risk type of the subject's heart.

In one embodiment, if the target ECG detection only includes resting ECG detection, the ECG data includes the age and resting ECG signal of the subject, and the reference feature includes a first resting reference feature; the high-frequency component of the QRS complex in the ECG data is analyzed according to the target ECG detection to obtain the corresponding reference feature, including:

Analyzing the high-frequency components of the QRS complex in the resting electrocardiogram signal to obtain a high-frequency QRS envelope curve;

A first resting reference feature is determined according to the high-frequency QRS envelope curve and the age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the number of first target leads and the number of second target leads; the first target number of leads refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target number of leads refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold.

In one embodiment, the reference feature further includes a second resting reference feature; and the step of analyzing the high frequency components of the QRS complex in the ECG data according to the target ECG detection to obtain the corresponding reference feature further includes:

Determining a second resting reference feature according to the high-frequency QRS envelope curve and the age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the first target lead number;

The determining of the cardiac risk assessment level and the cardiac risk type according to the reference feature comprises:

determining a first risk assessment level according to the first resting reference feature;

determining a second risk assessment level according to the second resting reference feature;

A cardiac risk assessment level and a cardiac risk type are determined according to the first risk assessment level and the second risk assessment level.

In one embodiment, the first resting reference feature also includes a target high-frequency morphology index and a target root mean square voltage; the second resting reference feature also includes the target high-frequency morphology index; the target high-frequency morphology index is the maximum value of the high-frequency morphology indexes corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltages corresponding to each resting lead.

In one embodiment, if the target ECG detection also includes load exercise ECG detection, the ECG data includes the age, resting ECG signal and exercise ECG signal of the subject, and the reference feature includes a first resting reference feature and a first exercise reference feature; the high-frequency components of the QRS complex in the ECG data are analyzed according to the target ECG detection to obtain the corresponding reference features, including:

Analyzing the high-frequency components of the QRS complex in the resting electrocardiogram signal to obtain a high-frequency QRS envelope curve;

Determine a first resting reference feature according to the high-frequency QRS envelope curve and the age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the number of first target leads and the number of second target leads; the first target lead number refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target lead number refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold;

Analyzing the high-frequency components of the QRS complex in the exercise electrocardiogram signal to obtain a high-frequency QRS waveform curve;

Determining the maximum heart rate of the subject according to the exercise electrocardiogram signal;

A first motion reference feature is determined according to the high-frequency QRS waveform curve, the age and the maximum heart rate; the first motion reference feature includes the number of motion-positive leads, the number of motion-critical leads, the number of third target leads, the number of fourth target leads and the number of fifth target leads; the third target lead number refers to the number of motion leads whose corresponding first amplitude decrease relative value is greater than or equal to the first relative value threshold; the fourth target lead number refers to the number of motion leads whose corresponding high-frequency QRS waveform curve shows a trend of repeated decrease and increase within a first time period; the fifth target lead number refers to the number of motion leads whose corresponding second amplitude decrease relative value is greater than or equal to the second relative value threshold.

In one embodiment, the reference feature further includes a second resting reference feature and a second motion reference feature; the step of analyzing the high frequency components of the QRS complex in the electrocardiogram data according to the target electrocardiogram detection to obtain the corresponding reference feature further includes:

Determining a second resting reference feature according to the high-frequency QRS envelope curve and the age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the first target lead number;

Determining a second motion reference feature according to the high-frequency QRS waveform curve, the age and the maximum heart rate; the second motion reference feature includes the number of motion-positive leads and the number of motion-critical leads;

The determining of the cardiac risk assessment level and the cardiac risk type according to the reference feature comprises:

determining a first risk assessment level according to the first resting reference feature;

determining a second risk assessment level according to the second rest reference feature and the second motion reference feature;

determining a third risk assessment level according to the first motion reference feature;

A cardiac risk assessment level and a cardiac risk type are determined according to the first risk assessment level, the second risk assessment level, and the third risk assessment level.

In one embodiment, the first resting reference feature also includes a target high-frequency morphology index and a target root mean square voltage; the second resting reference feature also includes the target high-frequency morphology index; the target high-frequency morphology index is the maximum value of the high-frequency morphology indexes corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltages corresponding to each resting lead; the first motion reference feature also includes a first relative amplitude decrease value and a second relative amplitude decrease value corresponding to each motion lead; the second motion reference feature also includes a second relative amplitude decrease value corresponding to each motion lead.

In one embodiment, the method further comprises:

Determining corresponding risk assessment features according to the cardiac risk type based on the electrocardiogram data;

A concern level of the corresponding cardiac risk type is determined according to the risk assessment characteristics.

A cardiac risk assessment device, comprising:

An acquisition module, used to acquire ECG data corresponding to a target ECG test that matches a clinical indication; the clinical indication is used to indicate whether a load exercise ECG test can be performed; the target ECG test at least includes a resting ECG test;

A feature determination module, used for analyzing the high-frequency components of the QRS complex in the ECG data according to the target ECG detection to obtain corresponding reference features;

The risk assessment module is used to determine the cardiac risk assessment level and the cardiac risk type according to the reference characteristics; the cardiac risk assessment level is used as a reference indicator for triaging the subject; the cardiac risk type is used as a reference indicator for assessing the possible risk type of the subject's heart.

A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in each method embodiment when executing the computer program.

A computer-readable storage medium stores a computer program, which implements the steps in various method embodiments when executed by a processor.

The above-mentioned cardiac risk assessment method, device, computer equipment and storage medium obtain the ECG data corresponding to the target ECG detection matching the clinical indication, and analyze the high-frequency components of the QRS wave group in the corresponding ECG data according to the target ECG detection to obtain the corresponding reference features, and evaluate and analyze the reference features, so as to efficiently and accurately obtain the cardiac risk assessment level that characterizes the risk of the subject's heart problems, and the cardiac risk type that characterizes the possible risk type of the heart. Thus, the risk size and risk type of the subject's heart problems can be efficiently and accurately assessed in a non-invasive manner for the doctor's reference, so that the doctor can efficiently and accurately identify the subject's heart health status in combination with clinical symptoms, and give further diagnosis and/or detection reference suggestions to achieve the guidance and diversion of the subject. In addition, the ECG data used to assess the subject's heart risk is collected during the target ECG detection process matching the subject's clinical indication, so as to obtain ECG data that can accurately assess the subject's heart risk as comprehensively as possible while ensuring the subject's safe detection, so as to further improve the accuracy of cardiac risk assessment, thereby further improving the accuracy of cardiac health status identification.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a process flow of a cardiac risk assessment method in one embodiment;

FIG2 is a schematic diagram of a high-frequency QRS waveform curve in one embodiment;

FIG3 is a schematic diagram of a high frequency QRS envelope curve in one embodiment;

FIG4 is a schematic flow chart of a cardiac risk assessment method in another embodiment;

FIG5 is a block diagram of a cardiac risk assessment device in one embodiment;

FIG. 6 is a diagram showing the internal structure of a computer device in one embodiment.

DETAILED DESCRIPTION

In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

The cardiac risk assessment method provided in the present application can be applied to a terminal, a server, or an interactive system including a terminal and a server, and is implemented through the interaction between the terminal and the server, which is not specifically limited here. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, electrocardiogram monitoring devices, and portable wearable devices, and the server can be implemented as an independent server or a server cluster consisting of multiple servers.

In one embodiment, as shown in FIG1 , a cardiac risk assessment method is provided, which is described by taking the method applied to a server as an example, and specifically includes the following steps:

S102, obtaining ECG data corresponding to a target ECG test that matches a clinical indication; the clinical indication is used to indicate whether a load exercise ECG test can be performed; the target ECG test at least includes a resting ECG test.

Among them, the clinical indication is used to indicate the ECG test applicable to the corresponding subject, specifically to indicate whether the corresponding subject can perform a load exercise ECG test, specifically including clinical data such as whether hypoglycemia, hypotension and acute attack of myocardial infarction are present, and whether vital signs are stable. If the vital signs are stable and hypoglycemia, hypotension and acute attack of myocardial infarction are excluded, it indicates that the corresponding subject can perform a load exercise ECG test, then the target ECG test matched by the clinical indication includes a resting ECG test and a load exercise ECG test, otherwise, it indicates that the corresponding subject cannot perform a load exercise ECG test, then the target ECG test matched by the clinical indication includes a resting ECG test. The subject is in a resting state during the resting ECG test. The subject is in a state of exercise during the load exercise ECG test to increase the subject's cardiac load through exercise. The target ECG test refers to the type of ECG test applicable to the subject. It can be understood that the specific content of the above clinical indications is only used as an example and is not used for specific limitation.

Specifically, the ECG data of the subject is obtained, which is the ECG data collected during the target ECG detection process. The target ECG detection matches the clinical indication of the subject, that is, the target ECG detection is determined by the clinical indication of the subject.

In one embodiment, if the target ECG detection includes a resting ECG detection, the acquired ECG data includes the age of the subject and the resting ECG signal collected during the resting ECG detection. If the target ECG detection includes a resting ECG detection and a load exercise ECG detection, the acquired ECG data includes the age of the subject, the resting ECG signal collected during the resting ECG detection, and the exercise ECG signal collected during the load exercise ECG detection.

S104, analyzing the high frequency components of the QRS complex in the ECG data according to the target ECG detection to obtain corresponding reference features.

Specifically, the ECG data includes multiple QRS complexes that reflect the changes in the left and right ventricular depolarization potentials and time, and each QRS complex includes high-frequency components and low-frequency components. Each target ECG detection corresponds to a corresponding reference feature, and by analyzing the high-frequency components of the QRS complex in the ECG data corresponding to each target ECG detection, the reference feature corresponding to the target ECG detection can be determined.

In one embodiment, if the target ECG detection includes a resting ECG detection, the high-frequency components of the QRS complex in the resting ECG signal are analyzed to obtain corresponding reference features. If the target ECG detection includes a resting ECG detection and a load exercise ECG detection, the high-frequency components of the QRS complex in the resting ECG signal are analyzed to obtain corresponding reference features, and the high-frequency components of the QRS complex in the exercise ECG signal are analyzed to obtain corresponding reference features. Among them, the reference features determined based on the resting ECG signal can be understood as resting reference features, and the resting reference features at least include a first resting reference feature and may also include a second resting reference feature. The reference features determined based on the exercise ECG signal can be understood as motion reference features, and the motion reference features at least include a first motion reference feature and may also include a second motion reference feature.

S106, determining a cardiac risk assessment level and a cardiac risk type based on the reference characteristics; the cardiac risk assessment level is used as a reference indicator for triaging the subject; the cardiac risk type is used as a reference indicator for assessing the possible risk type of the subject's heart.

Among them, the cardiac risk assessment level is used to characterize the risk of heart problems. For example, the higher the cardiac risk assessment level, the greater the risk of heart problems, so that doctors can guide and divert the subject in combination with the subject's clinical symptoms. The cardiac risk type is used to characterize the type of risk that the subject's heart may have, that is, it is used to characterize the type of heart problems that the subject may have, specifically including the risk of myocardial damage, the risk of sudden cardiac death, the risk of coronary artery stenosis, and at least one of the others. The risk of myocardial damage characterizes the possibility of myocardial damage, the risk of sudden cardiac death characterizes the possibility of sudden cardiac death, the risk of coronary artery stenosis characterizes the possibility of coronary artery stenosis, and the others characterize the possibility of no risk of any of myocardial damage, sudden cardiac death, and coronary artery stenosis. It can be understood that doctors can also divert subjects in combination with the cardiac risk assessment level, cardiac risk type, and clinical symptoms.

Specifically, the obtained reference features are evaluated and analyzed to obtain a corresponding cardiac risk assessment score, and the corresponding cardiac risk assessment level and cardiac risk type are determined based on the cardiac risk assessment score for reference by doctors, so that doctors can accurately understand the risk of the subject's heart problems and the possible risk types, thereby accurately identifying the subject's heart health status. If the target ECG test includes a resting ECG test, the cardiac risk assessment score is determined based on the resting reference index; if the target ECG test includes a resting ECG test and a load exercise ECG test, the cardiac risk assessment score is determined based on the resting reference index and the load exercise reference index, thereby determining the subject's cardiac risk assessment level and cardiac risk type.

In one embodiment, the reference features obtained are input into a corresponding preset risk assessment function or a pre-trained risk assessment model according to the target ECG detection to obtain a corresponding cardiac risk assessment score. It can be understood that the reference features are input into a pre-trained risk assessment model, and the risk assessment model can also directly output the corresponding cardiac risk assessment level and synchronously output the corresponding cardiac risk type, which is specifically determined by the training data set (including input features and corresponding output features) used to train the risk assessment model.

In one embodiment, the cardiac risk assessment score is compared with each preset score interval to determine the corresponding cardiac risk assessment level, and then determine the cardiac risk type. Taking the cardiac risk assessment level including six levels from the first level to the sixth level that increase in sequence as an example, a corresponding preset score interval is pre-configured for each cardiac risk assessment level, which are [0, 10], [11, 30], [31, 50], [51, 60], [61, 70], and [71, 100], respectively. If the cardiac risk assessment score is between [0, 10], the cardiac risk assessment level is determined to be the first level, and the cardiac risk type is determined to be others.

In one embodiment, if the output cardiac risk assessment level and cardiac risk type are provided for reference by the guiding doctor, the subject is triaged including suggesting the subject to go to the inpatient department for treatment, suggesting the subject to go to the outpatient department for treatment, and suggesting the subject to go home (no need for treatment). It can be understood that if the subject is recommended to go to the outpatient department or the inpatient department for treatment, the subject can also be diverted to the corresponding department or doctor according to the cardiac risk type. Among them, the guiding doctor is such as a doctor in a physical examination institution or a physical examination department, and also such as a doctor set up for the cardiovascular disease department in the hospital for guiding and diverting the subject who is about to go to the cardiovascular disease department for treatment, and no specific restrictions are made here. For example, taking the cardiac risk assessment level including six levels as an example, if the cardiac risk assessment level output for the subject is the first level, the guiding doctor can suggest the subject to go home, if the cardiac risk assessment level is the second level or the third level, the subject can be suggested to go to the outpatient department for treatment, otherwise, the subject is suggested to go to the inpatient department for treatment. It is worth mentioning that the guiding doctor also needs to guide and divert the subjects based on their clinical symptoms.

In one embodiment, if the output cardiac risk assessment level and cardiac risk type are for reference by outpatient doctors, the subjects are triaged including recommending hospitalization for the subjects, recommending regular checkups for the subjects, and recommending that the subjects seek medical attention again if they experience any discomfort symptoms. For example, taking the cardiac risk assessment level as an example, if the cardiac risk assessment level output for the subject is the first level, the outpatient doctor may recommend that the subject seek medical attention again if any discomfort symptoms occur. If the cardiac risk assessment level is the second or third level, the subject may be recommended to have regular checkups. Otherwise, the subject is recommended to be hospitalized. It is worth noting that outpatient doctors are also required to provide further diagnosis and treatment and/or testing reference suggestions based on the clinical symptoms of the subject.

The above-mentioned cardiac risk assessment method obtains the ECG data corresponding to the target ECG test that matches the clinical indications, and analyzes the high-frequency components of the QRS wave group in the corresponding ECG data according to the target ECG test to obtain the corresponding reference features, and evaluates and analyzes the reference features, so as to efficiently and accurately obtain the cardiac risk assessment level that characterizes the risk of the subject's heart problems, and the cardiac risk type that characterizes the possible risk type of the heart. Thus, the risk size and risk type of the subject's heart problems can be efficiently and accurately assessed in a non-invasive manner for the doctor's reference, so that the doctor can efficiently and accurately identify the subject's heart health status in combination with clinical symptoms, and give further diagnosis and/or testing reference suggestions to achieve guidance and diversion of the subject. In addition, the ECG data used to assess the subject's heart risk is collected during the target ECG test that matches the subject's clinical indications, so as to obtain ECG data that can accurately assess the subject's heart risk as comprehensively as possible while ensuring the subject's safe test, thereby further improving the accuracy of cardiac risk assessment, thereby further improving the accuracy of cardiac health status identification.

In one embodiment, if the target ECG detection only includes resting ECG detection, the ECG data includes the age and resting ECG signal of the subject, and the reference feature includes a first resting reference feature; S104 includes: analyzing the high-frequency components of the QRS complex in the resting ECG signal to obtain a high-frequency QRS envelope curve; determining the first resting reference feature based on the high-frequency QRS envelope curve and age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target lead number and the second target lead number; the first target lead number refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target lead number refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold.

Among them, the number of resting positive leads refers to the number of resting leads in which the corresponding lead positive index indicates positive, which can be used to assess the risk of myocardial ischemia in the resting state, and the two are positively correlated. The number of resting critical leads refers to the number of resting leads in which the corresponding lead positive index indicates critical, which can be used to assess the critical risk of myocardial ischemia in the resting state. Combined with this reference feature, myocardial ischemia can be more accurately identified. The lead positive index indicates positive, which is used to characterize that the high-frequency morphological index of the corresponding resting lead is greater than or equal to the second index threshold, and the age of the subject is greater than or equal to the preset age, or, it is used to characterize that the high-frequency morphological index of the corresponding resting lead is greater than or equal to the third index threshold, and the age of the subject is less than the preset age. The lead positive index indicates critical, which is used to characterize that the high-frequency morphological index of the corresponding resting lead is greater than the second index threshold and less than the third index threshold, and the age of the subject is less than the preset age. The first index threshold, the second index threshold and the third index threshold can be customized, for example, the first index threshold is 20%, the second index threshold is 8%, and the third index threshold is 15%. The first voltage threshold can be customized according to actual conditions, such as 4uV (microvolts). The preset age can be customized, such as 50 years old.

Specifically, each QRS complex in the resting ECG signal is aligned, averaged, and high-frequency filtered in sequence to obtain high-frequency QRS complex data (high-frequency band data of the QRS complex), or the QRS complex in the resting ECG signal is high-frequency filtered, aligned, and averaged in sequence to obtain high-frequency QRS complex data, or the high-frequency ECG signal is extracted from the resting ECG signal by analyzing the resting ECG signal, and then the QRS complex in the high-frequency ECG signal is aligned and averaged in sequence to obtain high-frequency QRS complex data, which is not specifically limited here. A high-frequency QRS envelope curve can be formed based on the high-frequency QRS complex data, and thus, the high-frequency QRS envelope curve can be obtained by performing the above data processing on the resting ECG signal. The high-frequency QRS envelope curve corresponding to each resting lead is analyzed to obtain the total area of each amplitude reduction region on the high-frequency QRS envelope curve as the first total area, and the total area below the high-frequency QRS envelope curve as the second total area, and the ratio of the first total area to the second total area is used as the high-frequency morphological index corresponding to the resting lead. According to the age of the subject and the high-frequency morphological index corresponding to each resting lead, the positive index of the corresponding lead is determined respectively, and the resting leads indicated as positive by the positive index of the corresponding lead are screened and counted to obtain the number of resting positive leads, the resting leads indicated as critical by the positive index of the corresponding lead are screened and counted to obtain the number of resting critical leads, and the resting leads with the corresponding high-frequency morphological index greater than or equal to the first index threshold are screened and counted to obtain the first target lead number. The root mean square of the high-frequency QRS envelope curve (that is, the high-frequency QRS wave group data) corresponding to each resting lead is calculated to obtain the root mean square voltage of the corresponding resting lead, and the resting leads with the corresponding root mean square voltage less than or equal to the first voltage threshold are screened and counted to obtain the second target lead number.

In one embodiment, if the reference feature includes a first resting reference feature, the cardiac risk assessment score includes a first risk assessment score. The first resting reference feature determined based on the subject's age and resting ECG signal is input into a first risk assessment function preconfigured for resting ECG detection or a pre-trained first risk assessment model to obtain a corresponding first risk assessment score, the first risk assessment score is compared with each preset score interval to obtain a first risk assessment level, the first risk type is determined according to the first risk assessment level, the first risk assessment level is determined as a cardiac risk assessment level, and the first risk type is determined as a cardiac risk type. For example, if the first risk assessment level is the first level, the first risk type is others, otherwise, the first risk type is myocarditis risk.

In one embodiment, the expression of the first risk assessment function preconfigured for resting ECG detection is as follows:

in, is the first risk assessment score, is the number of resting critical leads, is the number of resting positive leads, is the first target lead number, is the second target lead number.

In one embodiment, the first resting reference feature also includes a target high frequency morphology index and a target root mean square voltage, wherein the target high frequency morphology index is the maximum value of the high frequency morphology indexes corresponding to each resting lead, and the target root mean square voltage is the minimum value of the root mean square voltage corresponding to each resting lead. In this embodiment, the expression of the first risk assessment function preconfigured for resting ECG detection can also be expressed as follows:

in, is an additional score determined based on the target high frequency morphology index, is the additional fraction determined based on the target RMS voltage, and is obtained based on the following expressions:

in, is the target high-frequency morphological index, is the target RMS voltage in uV (microvolt).

In the above embodiment, if the clinical indication indicates that the subject cannot undergo stress exercise ECG testing, the first resting reference feature is determined based on the resting ECG signal collected during the resting ECG testing process and the age of the subject, so as to accurately assess the cardiac risk assessment level and cardiac risk type based on the first resting reference feature.

In one embodiment, the reference feature also includes a second resting reference feature; S104 also includes: determining the second resting reference feature based on the high-frequency QRS envelope curve and age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the number of first target leads; S106 includes: determining a first risk assessment level based on the first resting reference feature; determining a second risk assessment level based on the second resting reference feature; determining a cardiac risk assessment level and a cardiac risk type based on the first risk assessment level and the second risk assessment level.

The second resting reference feature is a subset of the first resting reference feature, so when the first resting reference feature is determined, the second resting reference feature can also be determined synchronously. The first resting reference feature is used to assess the possibility of myocarditis, and the second resting reference feature is used to assess the possibility of sudden cardiac death.

In one embodiment, if the reference feature includes a first resting reference feature and a second resting reference feature, the cardiac risk assessment score includes a first risk assessment score and a second risk assessment score. The first resting reference feature is input into a first risk assessment function preconfigured for resting ECG detection or a pre-trained first risk assessment model to obtain a corresponding first risk assessment score, and the first risk assessment score is compared with each preset score interval to obtain a first risk assessment level. The second resting reference feature is input into a second risk assessment function preconfigured for resting ECG detection or a second risk assessment model to obtain a corresponding second risk assessment score, and the second risk assessment score is compared with each preset score interval to obtain a second risk assessment level. The maximum value (highest level) of the first risk assessment level and the second risk assessment level is determined as the cardiac risk assessment level. The first risk type is determined according to the first risk assessment level, and the second risk type is determined according to the second risk assessment level, and the cardiac risk type is determined according to the first risk type and the second risk type.

For example, if the first risk assessment level is the second level and the second risk assessment level is the first level, the cardiac risk assessment level is determined to be the second level, the first risk type is determined to be myocarditis risk, the second risk type is determined to be others, and the cardiac risk type is determined to be myocarditis risk. If the first risk assessment level is the third level and the second risk assessment level is the second level, the cardiac risk assessment level is determined to be the third level, the first risk type is determined to be myocarditis risk, the second risk type is determined to be sudden cardiac death risk, and the cardiac risk type is determined to be myocarditis risk and sudden cardiac death risk.

In one embodiment, the expression of the second risk assessment function preconfigured for resting ECG detection is as follows:

in, is a second risk assessment score determined based on a second resting reference feature, is the number of resting critical leads, is the number of resting positive leads, is the first target lead number.

In the above embodiment, if the clinical indication indicates that the subject cannot undergo stress exercise ECG testing, the first resting reference feature and the second resting reference feature are determined based on the resting ECG signal and the age of the subject, so as to more accurately evaluate the cardiac risk assessment level and cardiac risk type in combination with the first resting reference feature and the second resting reference feature.

In one embodiment, the first resting reference feature also includes a target high-frequency morphology index and a target root mean square voltage; the second resting reference feature also includes a target high-frequency morphology index; the target high-frequency morphology index is the maximum value of the high-frequency morphology indexes corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltages corresponding to each resting lead.

In one embodiment, the expression of the second risk assessment function preconfigured for resting ECG detection can also be expressed as follows:

in, is an additional score determined based on the target high-frequency morphology index, based on the following expression:

in, is the target high-frequency morphology index.

In the above embodiment, based on the first resting reference feature and the second resting reference feature including the additional reference feature, it is possible to more accurately identify and distinguish the possible risk types of the heart, and more accurately assess the risk of heart problems, so that the doctor can more accurately identify the heart health status of the subject in combination with the clinical symptoms, and thus give reference suggestions for further diagnosis, treatment, testing, etc., so as to achieve more accurate guidance and diversion.

In one embodiment, if the target ECG detection also includes load exercise ECG detection, the ECG data includes the age, resting ECG signal and exercise ECG signal of the subject, and the reference features include a first resting reference feature and a first exercise reference feature; S104 includes: analyzing the high-frequency components of the QRS complex in the resting ECG signal to obtain a high-frequency QRS envelope curve; determining the first resting reference feature according to the high-frequency QRS envelope curve and the age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target lead number and the second target lead number; the first target lead number refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target lead number refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold; analyzing A high-frequency QRS waveform curve is obtained from the high-frequency components of the QRS wave group in the exercise ECG signal; the maximum heart rate of the subject is determined according to the exercise ECG signal; the first exercise reference feature is determined according to the high-frequency QRS waveform curve, age and maximum heart rate; the first exercise reference feature includes the number of exercise-positive leads, the number of exercise-critical leads, the number of third target leads, the number of fourth target leads and the number of fifth target leads; the number of third target leads refers to the number of exercise leads whose corresponding first amplitude decrease relative value is greater than or equal to the first relative value threshold; the number of fourth target leads refers to the number of exercise leads whose corresponding high-frequency QRS waveform curve shows a trend of repeated decrease and increase within a first time period; the number of fifth target leads refers to the number of exercise leads whose corresponding second amplitude decrease relative value is greater than or equal to the second relative value threshold.

Among them, the exercise ECG signal is the ECG signal collected during the load exercise ECG detection process. The load exercise ECG detection process includes multiple stages, which can specifically include three stages, namely, the resting stage, the exercise stage and the recovery stage. The exercise ECG signal includes the ECG signals of each stage. The division of the stages is not limited to this, and can be specifically divided according to actual conditions. The maximum heart rate refers to the maximum heart rate of the subject during the entire load exercise ECG detection process. The number of exercise positive leads refers to the number of exercise leads where the corresponding lead positive index indicates positive, which can be used to assess the risk of myocardial ischemia under load exercise, and the two are positively correlated. The number of critical exercise leads refers to the number of exercise leads where the corresponding lead positive index indicates critical, which can be used to assess the critical risk of myocardial ischemia under load exercise. Combined with this reference feature, the myocardial ischemia under load exercise can be more accurately assessed.

The lead positive indicator indicates a positive value, which is used to characterize that the first amplitude decrease relative value of the corresponding motion lead is greater than the third relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is less than the preset age, and the maximum heart rate of the subject is greater than 80% of the target heart rate, or, it is used to characterize that the first amplitude decrease relative value of the corresponding motion lead is greater than the fourth relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is less than the preset age, and the maximum heart rate of the subject is less than or equal to 80% of the target heart rate, or, it is used to characterize that the first amplitude decrease relative value of the corresponding motion lead is greater than the fourth relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is greater than or equal to the preset age, and the maximum heart rate of the subject is greater than 80% of the target heart rate, or, it is used to characterize that the first amplitude decrease relative value of the corresponding motion lead is greater than the fifth relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is greater than or equal to the preset age, and the maximum heart rate of the subject is less than or equal to 80% of the target heart rate. The first relative value threshold, the second relative value threshold, the third relative value threshold, the fourth relative value threshold, the fifth relative value threshold, the preset absolute value threshold, and the preset age are customized according to the actual situation, such as 55%, 40%, 60%, 50%, 40%, 1uV (microvolt), and 50 years old. The target heart rate is determined according to the age of the subject, such as target heart rate = (220-subject age) × 85%.

The lead positive indicator is critical, which is used to characterize that the first amplitude decrease relative value of the corresponding motion lead is greater than 90% of the third relative value threshold and less than or equal to the third relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is less than the preset age, and the maximum heart rate of the subject is greater than 80% of the target heart rate, or the first amplitude decrease relative value of the corresponding motion lead is greater than 90% of the fourth relative value threshold and less than or equal to the fourth relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is less than the preset age, and the maximum heart rate of the subject is less than or equal to 80% of the target heart rate. %, or, the first amplitude decrease relative value used to characterize the corresponding motion lead is greater than 90% of the fourth relative value threshold and less than or equal to the fourth relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is greater than or equal to the preset age, and the maximum heart rate of the subject is greater than 80% of the target heart rate, or, the first amplitude decrease relative value used to characterize the corresponding motion lead is greater than 90% of the fifth relative value threshold and less than or equal to the fifth relative value threshold, the absolute value of the amplitude decrease is greater than the preset absolute value threshold, the age of the subject is greater than or equal to the preset age, and the maximum heart rate of the subject is less than or equal to 80% of the target heart rate.

The first time period includes a period of time before exercise, during exercise and after exercise, the period of time before exercise is in the resting stage, the period of time during exercise includes the entire exercise stage, and the period of time after exercise is in the recovery stage. The period of time before exercise, during exercise and after exercise are continuous time periods. The second time period includes a period of time before exercise and a period of time during exercise, or the second time period includes a period of time during exercise, the period of time before exercise and the period of time during exercise are continuous time periods, and the period of time during exercise can be customized according to actual conditions, such as the first 3 minutes of exercise. The first time period may specifically include the second time period, that is, the second time period is a sub-interval of the first time period. The second amplitude decrease relative value is used to characterize the descending slope or steepness of the high-frequency QRS waveform curve in the second time period. If the second amplitude decrease relative value exceeds the second relative value threshold, it indicates that the descending slope or steepness is large enough, which indicates the possibility of coronary stenosis. The first amplitude decrease relative value can be understood as the relative value of the amplitude decrease of the high-frequency QRS waveform curve in the first time period, and the second amplitude decrease relative value can be understood as the relative value of the amplitude decrease of the high-frequency QRS waveform curve in the second time period.

Specifically, according to the first resting reference feature determination method disclosed in one or more embodiments of the present application, the corresponding first resting reference feature is determined according to the resting ECG signal and the age of the subject, which will not be repeated here. The exercise ECG signal includes the QRS wave group corresponding to each heartbeat of the subject during the entire load exercise ECG detection process. The exercise ECG signal is divided into multiple ECG signal subsets according to the timing and the preset moving step size through a window function, and each ECG signal subset includes QRS wave groups corresponding to multiple heartbeats. For each ECG signal subset, the QRS wave groups corresponding to the multiple heartbeats included therein are aligned, averaged and high-frequency filtered in turn to obtain the corresponding high-frequency QRS wave group data (high-frequency band data of the QRS wave group), and the high-frequency QRS wave group data is root-mean-squared to obtain the corresponding root-mean-square voltage, which is used as the root-mean-square voltage/intensity/amplitude corresponding to the ECG signal subset. The root mean square voltage/intensity/amplitude corresponding to each subset of ECG signals is smoothed according to the timing to obtain the high-frequency QRS waveform curve corresponding to the exercise ECG signal. Therefore, the high-frequency QRS waveform curve can be understood as a high-frequency QRS time-intensity curve. Among them, the window length and the preset moving step of the window function can be customized according to actual needs. For example, the window length is set to 10 seconds, and the preset moving step is set to 10 seconds or one heartbeat cycle. One heartbeat cycle refers to the time interval between two adjacent heartbeats, which is not specifically limited here. According to the timing, it refers to the order of detection time promoted by the signal acquisition time/load exercise ECG detection process. According to the method of extracting heart rate sequence from ECG signal disclosed in the prior art, the heart rate sequence of the subject is extracted from the exercise ECG data, and the maximum value from the heart rate sequence is selected as the maximum heart rate of the subject.

Further, a point with the largest root mean square voltage is selected from the high-frequency QRS waveform curve within the first time period as the first reference point, and a point with a time later than the first reference point and the smallest root mean square voltage is selected from the high-frequency QRS waveform curve within the first time period as the second reference point, and the root mean square voltage of the first reference point is subtracted from the root mean square voltage of the second reference point to obtain the first amplitude decrease absolute value, and the ratio of the first amplitude decrease absolute value to the root mean square voltage of the first reference point is determined as the first amplitude decrease relative value. The motion leads whose corresponding first amplitude decrease relative value is greater than or equal to the first relative value threshold are screened and counted to obtain the third target lead number. The lead positive index of the corresponding high-frequency QRS waveform curve is determined according to the first amplitude decrease relative value, the first amplitude decrease absolute value, the age of the subject and the maximum heart rate, as the lead positive index corresponding to the corresponding motion lead. The motion leads whose corresponding lead positive index indicates positive are screened and counted to obtain the number of motion positive leads, and the motion leads whose corresponding lead positive index indicates critical are screened and counted to obtain the number of motion critical leads. The change trend of the high-frequency QRS waveform curve corresponding to each motion lead in the first time period is analyzed to screen and count the motion leads whose corresponding high-frequency QRS waveform curve shows a downward and upward repeated fluctuation trend in the first time period to obtain the fourth target lead number.

A point with the smallest root mean square voltage is selected from the high-frequency QRS waveform curve in the second time period as the third reference point, and a point with a time earlier than the third reference point and the largest root mean square voltage is selected from the high-frequency QRS waveform curve in the second time period as the fourth reference point. The root mean square voltage of the fourth reference point is subtracted from the root mean square voltage of the third reference point to obtain the second amplitude decrease absolute value in the second time period, and the ratio of the second amplitude decrease absolute value to the root mean square voltage of the fourth reference point is used as the second amplitude decrease relative value. The motion leads whose corresponding second amplitude decrease relative value is greater than or equal to the second relative value threshold are screened and counted to obtain the fifth target lead number.

In one embodiment, the high-frequency QRS waveform curve presents a downward and upward repeated fluctuation trend in the first time period, which means that the total number of times the high-frequency QRS curve presents a downward trend in the first time period is greater than or equal to two times, and the upward trend and the downward trend appear alternately. For example, the high-frequency QRS waveform curve presents a "W" type or "inverted N" type waveform in the first time period.

In one embodiment, if the reference feature includes a first resting reference feature and a first motion reference feature, the cardiac risk assessment score includes a first risk assessment score and a third risk assessment score. According to one or more embodiments of the present application, the corresponding first risk assessment level and the first risk type are determined according to the first resting reference feature. The first motion reference feature is input into a risk assessment function preconfigured for load exercise ECG detection or a pre-trained risk assessment model to obtain a corresponding third risk assessment score, and the third risk assessment score is compared with each preset score interval to obtain a third risk assessment level, and the third risk type is determined according to the third risk assessment level. The highest level of the first risk assessment level and the third risk assessment level is determined as the cardiac risk assessment level, and the cardiac risk type is determined according to the first risk type and the third risk type. For example, if the first risk assessment level is the first level and the third risk assessment level is the fifth level, the first risk type is others, and the third risk type is coronary stenosis risk, then the cardiac risk assessment level is determined to be the fifth level, and the cardiac risk type is determined to be coronary stenosis risk.

In one embodiment, the expression of the risk assessment function preconfigured for stress exercise ECG detection is as follows:

in, is the third risk assessment score, is the number of motion critical leads, is the number of motor-positive leads, is the third target lead number, is the fourth target lead number, is the fifth target lead number.

In one embodiment, the first motion reference feature also includes a first amplitude drop relative value and a second amplitude drop relative value corresponding to each motion lead. In this embodiment, the expression of the risk assessment function preconfigured for stress exercise ECG detection can also be expressed as follows:

in, is an additional score determined based on the relative value of the first amplitude decrease corresponding to each motion lead, is an additional score determined based on the relative value of the second amplitude decrease corresponding to each motion lead, and is obtained based on the following expressions:

In the above embodiment, if the clinical indication indicates that the subject can undergo stress exercise ECG testing, a first resting reference feature is determined based on the resting ECG signal collected during the resting ECG testing and the age of the subject, and a first motion reference feature is determined based on the motion ECG signal collected during the stress exercise ECG testing and the age of the subject, so as to accurately assess the cardiac risk assessment level and the cardiac risk type based on the first resting reference feature and the first motion reference feature.

In one embodiment, the reference feature also includes a second resting reference feature and a second motion reference feature; S104 also includes: determining the second resting reference feature according to the high-frequency QRS envelope curve and age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the number of first target leads; determining the second motion reference feature according to the high-frequency QRS waveform curve, age and maximum heart rate; the second motion reference feature includes the number of motion positive leads and the number of motion critical leads; S106 includes: determining the first risk assessment level according to the first resting reference feature; determining the second risk assessment level according to the second resting reference feature and the second motion reference feature; determining the third risk assessment level according to the first motion reference feature; determining the cardiac risk assessment level and the cardiac risk type according to the first risk assessment level, the second risk assessment level and the third risk assessment level.

Among them, the second resting reference feature is a subset of the first resting reference feature, so when the first resting reference feature is determined, the second resting reference feature can also be synchronously determined. Similarly, the second motion reference feature is a subset of the first motion reference feature, so when the first motion reference feature is determined, the second motion reference feature can also be synchronously determined. The first resting reference feature is used to assess the possibility of myocarditis, the second resting reference feature and the second motion reference feature are used to jointly assess the possibility of sudden cardiac death, and the first motion reference feature is used to assess the possibility of coronary artery stenosis.

In one embodiment, if the reference feature includes a first resting reference feature, a second resting reference feature, a first motion reference feature, and a second motion reference feature, the cardiac risk assessment score includes a first risk assessment score, a second risk assessment score, and a third risk assessment score. According to one or more embodiments of the present application, the first risk assessment score is determined according to the first resting reference feature, and then the first risk assessment level and the first risk type are determined, and the third risk assessment score is determined according to the first motion reference feature, and then the third risk assessment level and the third risk type are determined. The second resting reference feature and the second motion reference feature are input into a risk assessment function or a risk assessment model configured for resting ECG detection and load exercise ECG detection to obtain a second risk assessment score, and the second risk assessment score is compared with each preset score interval to obtain a second risk assessment level, and the second risk type is determined according to the second risk assessment level. Further, the highest level among the first risk assessment level, the second risk assessment level, and the third risk assessment level is determined as the cardiac risk assessment level, and the cardiac risk type is determined according to the first risk type, the second risk type, and the third risk type.

For example, if the first risk assessment level is the second level, the second risk assessment level is the fourth level, and the third risk assessment level is the third level, the first risk type is myocarditis risk, the second risk type is sudden cardiac death risk, and the third risk type is coronary artery stenosis risk, then the cardiac risk assessment level is determined to be the fourth level, and the cardiac risk types are determined to be myocarditis risk, sudden cardiac death risk, and coronary artery stenosis risk.

In one embodiment, the expression of the risk assessment function for the joint configuration of resting ECG detection and stress exercise ECG detection is as follows:

in, is a second risk assessment score determined jointly based on the second rest reference feature and the second motion reference feature, is the number of resting critical leads, is the number of resting positive leads, is the first target lead number, is the number of motion critical leads, is the number of motion-positive leads.

In the above embodiment, if the clinical indication indicates that the subject is capable of undergoing stress exercise ECG testing, the first resting reference feature and the second resting reference feature are determined based on the resting ECG signal and the subject's age, and the first motion reference feature and the second motion reference feature are determined based on the exercise ECG signal and the subject's age, so as to more accurately assess the cardiac risk assessment level and the cardiac risk type by combining the first resting reference feature, the second resting reference feature, the first motion reference feature and the second motion reference feature.

In one embodiment, the first resting reference feature also includes a target high-frequency morphology index and a target root mean square voltage; the second resting reference feature also includes a target high-frequency morphology index; the target high-frequency morphology index is the maximum value of the high-frequency morphology indexes corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltages corresponding to each resting lead; the first motion reference feature also includes a first amplitude decrease relative value and a second amplitude decrease relative value corresponding to each motion lead; the second motion reference feature also includes a second amplitude decrease relative value corresponding to each motion lead.

In one embodiment, the expression of the risk assessment function for the joint configuration of resting ECG detection and stress exercise ECG detection can also be expressed as follows:

in, is an additional score determined based on the target high frequency morphology index, is an additional score determined based on the relative value of the second amplitude decrease corresponding to each motion lead, and is obtained based on the following expressions:

in, is the target high-frequency morphology index.

In the above embodiment, based on the first resting reference feature, the second resting reference feature, the first motion reference feature and the second motion reference feature including the additional reference feature, it is possible to more accurately identify and distinguish the possible risk types of the heart, and more accurately assess the risk of heart problems, so that doctors can more accurately identify the heart health status of the subjects in combination with clinical symptoms, and thus provide reference suggestions for further diagnosis, treatment, testing, etc., to achieve more accurate guidance and diversion.

In one embodiment, the cardiac risk assessment method further includes: determining corresponding risk assessment features according to the cardiac risk type based on the electrocardiogram data; and determining a concern level of the corresponding cardiac risk type based on the risk assessment features.

Specifically, the attention level represents the different degree of attention, which can be used to indicate the different possibilities of the subject to have the corresponding cardiac risk type. After the cardiac risk type is determined according to the ECG data, the corresponding risk assessment feature is determined according to the ECG data according to each determined cardiac risk type, and the corresponding attention level is determined according to the risk assessment feature corresponding to each cardiac risk type, for the doctor's reference in the diagnosis process, so that the doctor can accurately identify the subject's cardiac health status according to the cardiac risk type, the attention level and the clinical symptoms, and thus give corresponding reference suggestions for diagnosis and treatment. It can be understood that at least one of the cardiac risk assessment level and the risk assessment feature corresponding to the cardiac risk type can also be output, so that the doctor can also comprehensively consider these reference indicators to identify the cardiac health status. In this embodiment, the output cardiac risk type, attention level, etc. are provided for reference by outpatient doctors or inpatient doctors, so that the doctor can accurately identify the subject's cardiac health status in combination with clinical symptoms, and thus give corresponding reference suggestions for diagnosis and treatment or testing.

In one embodiment, the risk assessment feature corresponding to each cardiac risk type is input into a risk assessment model pre-trained for the cardiac risk type, and the risk assessment model outputs the corresponding level of concern, or the risk assessment feature corresponding to each cardiac risk type is input into a risk assessment function pre-configured for the cardiac risk type or a pre-trained risk assessment model to obtain a risk assessment score corresponding to the cardiac risk type, and then determine the corresponding level of concern, or each risk assessment feature corresponding to each cardiac risk type is compared with a corresponding preset reference interval to determine a reference level corresponding to each risk assessment feature, and the corresponding level of concern is determined based on each reference level. It can be understood that if the risk assessment feature only includes features for quantitative analysis, the corresponding level of concern is determined based on the risk assessment feature in the above-mentioned evaluation method. If the risk assessment feature includes features for qualitative analysis and features for quantitative analysis, then when it is determined that the corresponding cardiac risk type exists based on the features for qualitative analysis, the corresponding level of concern can be determined based on the features for quantitative analysis in the above-mentioned evaluation method. For a risk assessment model that uses the level of concern as an output feature, the features for qualitative analysis and the features for quantitative analysis can also be used as input features and input into the corresponding risk assessment model to obtain the corresponding level of concern.

In one embodiment, the ECG data only includes a resting ECG signal. If the cardiac risk type includes the risk of myocardial injury, the risk assessment feature corresponding to the risk of myocardial injury is determined according to the resting ECG signal. If the cardiac risk type also includes the risk of sudden cardiac death, the risk assessment feature corresponding to the risk of sudden cardiac death is also determined according to the resting ECG signal. The risk assessment feature corresponding to the risk of myocardial injury includes a first risk assessment feature for assessing the risk of myocarditis and a second risk assessment feature for assessing the risk of heart failure. The first risk assessment feature includes the number of resting positive leads, which is used to determine the type of myocarditis. If the type of myocarditis is acute myocarditis, the first risk assessment feature also includes a target high-frequency morphology index. If the type of myocarditis is chronic myocarditis, the first risk assessment feature also includes a target high-frequency morphology index and an average peak voltage of limb leads. If the type of myocarditis is fulminant myocarditis, the first risk assessment feature also includes a QRS duration, a target high-frequency morphology index, an average peak voltage of limb leads, and a high-frequency morphology index corresponding to each resting lead. Among them, the target high-frequency morphology index is the maximum value of the high-frequency morphology index corresponding to each resting lead, the average peak voltage of the limb leads is the average value of the peak voltage of each limb lead, and the resting leads include limb leads and chest leads. It can be understood that the first risk assessment feature may also include the sixth target lead number, which is used to qualitatively analyze the myocarditis situation. The sixth target lead number is the number of resting leads whose corresponding wave peak number exceeds the first quantity threshold and the corresponding lead positive index indicates positive. The first quantity threshold is, for example, 3. The second risk assessment feature includes the QRS duration, the average peak voltage of the limb leads and the high-frequency morphology index corresponding to each resting lead, and may also include the number of resting positive leads, which is used to quantitatively analyze the risk of heart failure, and may also include the sixth target lead number, which is used to qualitatively analyze the heart failure situation.

In this embodiment, the risk assessment features corresponding to the risk of sudden cardiac death include the QRS duration and the high-frequency morphology index corresponding to each resting lead, and also include at least one of the number of resting positive leads, arrhythmia assessment index, target peak voltage and tachycardia assessment index, which are used to quantitatively analyze the risk of sudden cardiac death, and may also include the number of seventh target leads for qualitative analysis of the risk of sudden cardiac death, and may also include QRS fragmentation wave index for more accurate qualitative analysis in combination with the number of seventh target leads. The arrhythmia assessment index includes frequent arrhythmias and occasional arrhythmias. The target peak voltage is the minimum value of the peak voltages corresponding to each resting lead. The tachycardia assessment index is used to indicate whether there is a possibility of ventricular tachycardia. The number of seventh leads is the number of resting leads whose corresponding wave peak number is greater than or equal to the second number threshold. The second number threshold is, for example, 4. The arrhythmia assessment index and the tachycardia assessment index are determined based on the low-frequency electrocardiogram obtained from the resting electrocardiogram signal. The QRS fragmentation index is used to indicate whether there is a QRS fragmentation wave in the low-frequency electrocardiogram.

In one embodiment, the ECG data includes a resting ECG signal and an exercise ECG signal. If the cardiac risk type includes the risk of coronary artery stenosis, the corresponding risk assessment feature is determined based on the exercise ECG signal. If the cardiac risk type includes the risk of myocardial injury, the corresponding risk assessment feature is determined based on the resting ECG signal. If the cardiac risk type includes the risk of sudden cardiac death, the corresponding risk assessment feature is determined based on the resting ECG signal and the exercise ECG signal. The risk assessment feature corresponding to the risk of coronary artery stenosis includes at least one of a first relative value of amplitude decrease and a first absolute value of amplitude decrease, which is used to quantitatively analyze the coronary artery stenosis. It may also include the number of exercise-positive leads, which is used to quantitatively analyze the myocardial ischemia. It may also include a second relative value of amplitude decrease, which is used to qualitatively analyze the coronary artery stenosis. The risk assessment feature corresponding to the risk of myocardial injury is consistent with that in the above-mentioned corresponding embodiments and will not be repeated here. The risk assessment characteristics corresponding to the risk of sudden cardiac death include risk assessment characteristics determined based on resting ECG signals and risk assessment characteristics determined based on exercise ECG signals. The risk assessment characteristics determined based on resting ECG signals are consistent with those in the above-mentioned corresponding embodiments and will not be repeated here. The risk assessment characteristics determined based on exercise ECG signals include the number of motion-positive leads and may also include the relative value of the second amplitude decrease.

In the above embodiment, corresponding risk assessment features are determined according to the subject's cardiac risk type based on the ECG data, and the concern level corresponding to the cardiac risk type is further determined based on the risk assessment features for the doctor's reference, so that the doctor can more accurately identify the subject's cardiac health status in combination with clinical symptoms.

In one or more embodiments of the present application, the resting leads used to collect resting ECG signals during resting ECG detection and the motion leads used to collect motion ECG signals during load exercise ECG detection may be the same or different in type, quantity, and position, and are not specifically limited here.

In one implementation, as shown in FIG2 , a schematic diagram of a high-frequency QRS waveform is provided. The high-frequency QRS waveform is used to characterize the change trend of the root mean square of the high-frequency component of the subject's QRS wave group over time during the entire load exercise ECG detection process, that is, to reflect the energy change trend during the entire load exercise ECG detection process. FIG2 illustrates the high-frequency QRS waveform corresponding to the limb lead aVF. The horizontal axis is time, corresponding to the detection time of the load exercise ECG detection process, and the unit is min (minutes). The vertical axis is the root mean square voltage (RMS voltage), which can also be understood as intensity or amplitude, and the unit is uV (microvolt). Among them, the relative value of the first amplitude decrease and the absolute value of the first amplitude decrease used to determine the positive indicator of the corresponding lead are 55% and 2.9uV, respectively.

In one implementation, as shown in FIG3 , a schematic diagram of a high-frequency QRS envelope curve is provided. The high-frequency QRS envelope curve reflects the shape diagram obtained by averaging all high-frequency QRS complexes (high-frequency components of the QRS complex) in the resting ECG signal, and is specifically presented as a single high-frequency QRS complex envelope curve. FIG3 illustrates the high-frequency QRS envelope curve corresponding to chest lead V5, where the horizontal axis is time, corresponding to the duration of the QRS complex, in ms (milliseconds), and the vertical axis is voltage, in uV (microvolts). Among them, the high-frequency morphology index of the corresponding lead is 7.1%.

As shown in FIG4 , in one embodiment, a cardiac risk assessment method is provided, the method specifically comprising the following steps:

S402, obtaining ECG data corresponding to a target ECG test that matches a clinical indication; the clinical indication is used to indicate whether a load exercise ECG test can be performed; the target ECG test at least includes a resting ECG test.

S404, if the target ECG detection only includes resting ECG detection, the ECG data includes the age of the subject and the resting ECG signal; and the high-frequency components of the QRS complex in the resting ECG signal are analyzed to obtain a high-frequency QRS envelope curve.

S406, determine the first resting reference feature and the second resting reference feature according to the high-frequency QRS envelope curve and age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target number of leads, the second target number of leads, the target high-frequency morphology index and the target root mean square voltage; the first target number of leads refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target number of leads refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold; the target high-frequency morphology index is the maximum value of the high-frequency morphology index corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltage corresponding to each resting lead; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target number of leads and the target high-frequency morphology index.

S408: Determine a first risk assessment level according to the first rest reference feature.

S410: Determine a second risk assessment level according to a second rest reference feature.

S412, determining a cardiac risk assessment level and a cardiac risk type according to the first risk assessment level and the second risk assessment level; the cardiac risk assessment level is used as a reference indicator for triaging the subject; the cardiac risk type is used as a reference indicator for assessing the possible risk type of the subject's heart.

S414, if the target ECG test also includes load exercise ECG test, the ECG data includes the age of the subject, resting ECG signal and exercise ECG signal; the high-frequency components of the QRS wave group in the resting ECG signal are analyzed to obtain a high-frequency QRS envelope curve.

S416, determine the first resting reference feature and the second resting reference feature according to the high-frequency QRS envelope curve and age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target number of leads, the second target number of leads, the target high-frequency morphology index and the target root mean square voltage; the first target number of leads refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target number of leads refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold; the target high-frequency morphology index is the maximum value of the high-frequency morphology index corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltage corresponding to each resting lead; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target number of leads and the target high-frequency morphology index.

S418, analyzing the high-frequency components of the QRS wave group in the exercise electrocardiogram signal to obtain a high-frequency QRS waveform curve.

S420, determining the maximum heart rate of the subject according to the exercise electrocardiogram signal.

S422, determining the first motion reference feature and the second motion reference feature according to the high-frequency QRS waveform curve, age and maximum heart rate; the first motion reference feature includes the number of motion-positive leads, the number of motion-critical leads, the number of third target leads, the number of fourth target leads and the number of fifth target leads, as well as the first relative amplitude decrease value and the second relative amplitude decrease value corresponding to each motion lead; the third target lead number refers to the number of motion leads whose corresponding first relative amplitude decrease value is greater than or equal to the first relative value threshold; the fourth target lead number refers to the number of motion leads whose corresponding high-frequency QRS waveform curve shows a trend of repeated fluctuation of decrease and increase within a first time period; the fifth target lead number refers to the number of motion leads whose corresponding second relative amplitude decrease value is greater than or equal to the second relative value threshold; the second motion reference feature includes the number of motion-positive leads and the number of motion-critical leads, as well as the second relative amplitude decrease value corresponding to each motion lead.

S424: Determine a first risk assessment level according to the first rest reference feature.

S426: Determine a second risk assessment level according to the second rest reference feature and the second motion reference feature.

S428: Determine a third risk assessment level according to the first motion reference feature.

S430: Determine a cardiac risk assessment level and a cardiac risk type according to the first risk assessment level, the second risk assessment level, and the third risk assessment level.

S432, determining corresponding risk assessment features according to the cardiac risk type based on the ECG data.

S434, determining the level of concern of the corresponding cardiac risk type according to the risk assessment characteristics.

In the above embodiment, if the clinical indication indicates that the subject cannot undergo stress exercise ECG testing, the subject's age and the resting ECG signal collected during the resting ECG testing process are obtained and analyzed to obtain the first resting reference feature and the second resting reference feature, and then the corresponding cardiac risk assessment level and cardiac risk type are determined. If the clinical indication indicates that the subject can undergo stress exercise ECG testing, the subject's age and the resting ECG signal collected during the resting ECG testing process are obtained and analyzed to obtain the first resting reference feature and the second resting reference feature, and the age and the exercise ECG signal collected during the stress exercise ECG testing process are obtained and analyzed to obtain the first exercise reference feature and the second exercise reference feature, and the corresponding cardiac risk assessment level and cardiac risk type are obtained by comprehensively considering each reference feature. By outputting the cardiac risk assessment level that characterizes the risk of the subject's heart problems and the cardiac risk type that characterizes the possible risk type of the heart for the doctor's reference, the doctor can efficiently and accurately identify the subject's heart health status in combination with clinical symptoms, and give further diagnosis and/or testing reference suggestions to achieve general screening, guidance and triage of the subject. Based on the ECG data, the corresponding risk assessment features are determined according to the subject's heart risk type, and then the corresponding level of concern for the heart risk type is determined for the doctor's reference, so that the doctor can more accurately identify the subject's heart health status in combination with clinical symptoms, and give further diagnosis and treatment reference suggestions.

It should be understood that, although the various steps in the flowcharts of Fig. 1 and Fig. 4 are sequentially displayed according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in Fig. 1 and Fig. 4 may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

In one embodiment, as shown in FIG5 , a cardiac risk assessment device 500 is provided, comprising: an acquisition module 501 , a feature determination module 502 and a risk assessment module 503 , wherein:

The acquisition module 501 is used to acquire ECG data corresponding to a target ECG test that matches a clinical indication; the clinical indication is used to indicate whether a load exercise ECG test can be performed; the target ECG test at least includes a resting ECG test;

The feature determination module 502 is used to analyze the high frequency components of the QRS complex in the ECG data according to the target ECG detection to obtain corresponding reference features;

The risk assessment module 503 is used to determine the cardiac risk assessment level and cardiac risk type according to the reference characteristics; the cardiac risk assessment level is used as a reference indicator for triaging the subject; the cardiac risk type is used as a reference indicator for assessing the possible risk type of the subject's heart.

In one embodiment, if the target ECG detection only includes resting ECG detection, the ECG data includes the age and resting ECG signal of the subject, and the reference feature includes a first resting reference feature; the feature determination module 502 is also used to analyze the high-frequency components of the QRS complex in the resting ECG signal to obtain a high-frequency QRS envelope curve; determine the first resting reference feature based on the high-frequency QRS envelope curve and age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target lead number and the second target lead number; the first target lead number refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target lead number refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold.

In one embodiment, the reference feature also includes a second resting reference feature; the feature determination module 502 is also used to determine the second resting reference feature based on the high-frequency QRS envelope curve and age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the number of first target leads; the risk assessment module 503 is also used to determine the first risk assessment level based on the first resting reference feature; determine the second risk assessment level based on the second resting reference feature; determine the cardiac risk assessment level and the cardiac risk type based on the first risk assessment level and the second risk assessment level.

In one embodiment, the first resting reference feature also includes a target high-frequency morphology index and a target root mean square voltage; the second resting reference feature also includes a target high-frequency morphology index; the target high-frequency morphology index is the maximum value of the high-frequency morphology indexes corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltages corresponding to each resting lead.

In one embodiment, if the target ECG detection also includes load exercise ECG detection, the ECG data includes the age, resting ECG signal and exercise ECG signal of the subject, and the reference features include a first resting reference feature and a first exercise reference feature; the feature determination module 502 is also used to analyze the high-frequency components of the QRS wave group in the resting ECG signal to obtain a high-frequency QRS envelope curve; determine the first resting reference feature according to the high-frequency QRS envelope curve and age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, the first target lead number and the second target lead number; the first target lead number refers to the number of resting leads whose corresponding high-frequency morphology index is greater than or equal to the first index threshold; the second target lead number refers to the number of resting leads whose corresponding root mean square voltage is less than or equal to the first voltage threshold ; Analyze the high-frequency components of the QRS wave group in the exercise ECG signal to obtain a high-frequency QRS waveform curve; determine the maximum heart rate of the subject based on the exercise ECG signal; determine the first motion reference feature based on the high-frequency QRS waveform curve, age and maximum heart rate; the first motion reference feature includes the number of motion-positive leads, the number of motion-critical leads, the number of third target leads, the number of fourth target leads and the number of fifth target leads; the number of third target leads refers to the number of motion leads whose corresponding first amplitude decrease relative value is greater than or equal to the first relative value threshold; the number of fourth target leads refers to the number of motion leads whose corresponding high-frequency QRS waveform curve shows a trend of repeated decrease and increase within a first time period; the number of fifth target leads refers to the number of motion leads whose corresponding second amplitude decrease relative value is greater than or equal to the second relative value threshold.

In one embodiment, the reference feature also includes a second resting reference feature and a second motion reference feature; the feature determination module 502 is also used to determine the second resting reference feature based on the high-frequency QRS envelope curve and age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the number of first target leads; the second motion reference feature is determined based on the high-frequency QRS waveform curve, age and maximum heart rate; the second motion reference feature includes the number of motion positive leads and the number of motion critical leads; the risk assessment module 503 is also used to determine the first risk assessment level based on the first resting reference feature; determine the second risk assessment level based on the second resting reference feature and the second motion reference feature; determine the third risk assessment level based on the first motion reference feature; determine the cardiac risk assessment level and the cardiac risk type based on the first risk assessment level, the second risk assessment level and the third risk assessment level.

In one embodiment, the first resting reference feature also includes a target high-frequency morphology index and a target root mean square voltage; the second resting reference feature also includes a target high-frequency morphology index; the target high-frequency morphology index is the maximum value of the high-frequency morphology indexes corresponding to each resting lead; the target root mean square voltage is the minimum value of the root mean square voltages corresponding to each resting lead; the first motion reference feature also includes a first amplitude decrease relative value and a second amplitude decrease relative value corresponding to each motion lead; the second motion reference feature also includes a second amplitude decrease relative value corresponding to each motion lead.

In one embodiment, the feature determination module 502 is further used to determine corresponding risk assessment features according to the cardiac risk type based on the ECG data; the risk assessment module 503 is further used to determine the concern level of the corresponding cardiac risk type based on the risk assessment features.

For the specific definition of the cardiac risk assessment device, please refer to the definition of the cardiac risk assessment method above, which will not be repeated here. Each module in the above cardiac risk assessment device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in FIG6 . The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store ECG data corresponding to target ECG detection that matches clinical indications. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a cardiac risk assessment method is implemented.

Those skilled in the art will understand that the structure shown in FIG. 6 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims (10)

1. A cardiac risk assessment method, characterized in that the method includes: Obtain the ECG data corresponding to the target ECG test that matches the clinical indication; the clinical indication is used to indicate whether stress exercise ECG test can be performed; the target ECG test at least includes resting ECG test; Analyze the high-frequency components of the QRS wave complex in the electrocardiogram data according to the target electrocardiogram detection to obtain corresponding reference features; The cardiac risk assessment level and cardiac risk type are determined according to the reference characteristics; the cardiac risk assessment level is used as a reference index for shunting the subject; the cardiac risk type is used to evaluate the possible risk of the subject's heart Type of reference indicators; Wherein, if the target ECG detection also includes stress exercise ECG detection, the ECG data includes the subject's age, resting ECG signal and exercise ECG signal, and the reference feature includes a first Resting reference features and first movement reference features; analyzing the high-frequency components of the QRS wave complex in the ECG data according to the target ECG detection to obtain corresponding reference features, including: Analyze the high-frequency components of the QRS wave complex in the resting ECG signal to obtain a high-frequency QRS envelope curve; The first resting reference feature is determined according to the high-frequency QRS envelope curve and the age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, and the number of first target leads. and the second target lead number; the first target lead number refers to the number of resting leads whose corresponding high-frequency morphological index is greater than or equal to the first index threshold; the second target lead number refers to the corresponding average The number of resting leads whose root-square voltage is less than or equal to the first voltage threshold; Analyze the high-frequency components of the QRS complex in the exercise ECG signal to obtain a high-frequency QRS waveform curve; Determine the maximum heart rate of the subject according to the exercise ECG signal; The first motion reference feature is determined according to the high-frequency QRS waveform curve, the age and the maximum heart rate; the first motion reference feature includes the number of motion positive leads, the number of motion critical leads, and the number of third target leads. , the fourth target lead number and the fifth target lead number; the third target lead number refers to the number of motion leads whose corresponding first amplitude drop relative value is greater than or equal to the first relative value threshold; The number of four target leads refers to the number of motion leads whose corresponding high-frequency QRS waveform curve shows a downward and upward fluctuation trend in the first period of time; the number of fifth target leads refers to the relative value of the corresponding second amplitude decrease. The number of motor leads greater than or equal to the second relative value threshold. 2. The method according to claim 1, wherein if the target ECG detection only includes resting ECG detection, the ECG data includes the subject's age and resting ECG signal. , the reference feature includes a first resting reference feature; and analyzing the high-frequency components of the QRS wave complex in the ECG data according to the target ECG detection to obtain the corresponding reference feature includes: Analyze the high-frequency components of the QRS wave complex in the resting ECG signal to obtain a high-frequency QRS envelope curve; The first resting reference feature is determined according to the high-frequency QRS envelope curve and the age; the first resting reference feature includes the number of resting positive leads, the number of resting critical leads, and the number of first target leads. and the second target lead number; the first target lead number refers to the number of resting leads whose corresponding high-frequency morphological index is greater than or equal to the first index threshold; the second target lead number refers to the corresponding average The number of resting leads whose root-square voltage is less than or equal to the first voltage threshold. 3. The method according to claim 2, wherein the reference feature further includes a second resting reference feature; and the high-frequency QRS wave complex in the ECG data is detected according to the target ECG. The components are analyzed to obtain corresponding reference characteristics, which also include: A second resting reference feature is determined according to the high-frequency QRS envelope curve and the age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the Number of first target leads; Determining the cardiac risk assessment level and cardiac risk type based on the reference characteristics includes: Determine a first risk assessment level based on the first resting reference characteristic; Determine a second risk assessment level based on the second resting reference characteristic; The cardiac risk assessment level and the cardiac risk type are determined according to the first risk assessment level and the second risk assessment level. 4. The method according to claim 3, wherein the first resting reference feature further includes a target high-frequency morphological index and a target root mean square voltage; the second resting reference feature further includes the target High-frequency morphology index; the target high-frequency morphology index is the maximum value among the high-frequency morphology indexes corresponding to each resting lead; the target root-mean-square voltage is the root-mean-square voltage corresponding to each resting lead. minimum value. 5. The method of claim 1, wherein the reference features further include a second resting reference feature and a second motion reference feature; The high-frequency components of the QRS wave complex are analyzed to obtain corresponding reference features, which also include: A second resting reference feature is determined according to the high-frequency QRS envelope curve and the age; the second resting reference feature includes the number of resting positive leads, the number of resting critical leads and the Number of first target leads; Determine a second motion reference feature according to the high-frequency QRS waveform curve, the age and the maximum heart rate; the second motion reference feature includes the number of motion positive leads and the number of motion critical leads; Determining the cardiac risk assessment level and cardiac risk type based on the reference characteristics includes: Determine a first risk assessment level based on the first resting reference characteristic; Determine a second risk assessment level based on the second resting reference feature and the second motion reference feature; Determine a third risk assessment level based on the first motion reference characteristic; The cardiac risk assessment level and cardiac risk type are determined according to the first risk assessment level, the second risk assessment level and the third risk assessment level. 6. The method of claim 5, wherein the first resting reference feature further includes a target high-frequency morphological index and a target root mean square voltage; the second resting reference feature further includes the target High-frequency morphology index; the target high-frequency morphology index is the maximum value among the high-frequency morphology indexes corresponding to each resting lead; the target root-mean-square voltage is the root-mean-square voltage corresponding to each resting lead. minimum value; the first motion reference feature also includes the first relative amplitude decrease value and the second relative amplitude decrease value corresponding to each motion lead; the second motion reference feature also includes the second relative amplitude decrease value corresponding to each motion lead. value. 7. The method according to any one of claims 1 to 6, characterized in that the method further includes: Determine corresponding risk assessment characteristics according to the cardiac risk type according to the ECG data; The level of concern for the corresponding cardiac risk type is determined based on the risk assessment characteristics. 8. A cardiac risk assessment device, characterized in that the device includes: The acquisition module is used to obtain the ECG data corresponding to the target ECG detection that matches the clinical indications; the clinical indications are used to indicate whether stress exercise ECG detection can be performed; the target ECG detection at least includes static ECG testing; A feature determination module, configured to analyze the high-frequency components of the QRS wave complex in the ECG data according to the target ECG detection to obtain corresponding reference features; A risk assessment module, configured to determine a cardiac risk assessment level and a cardiac risk type based on the reference characteristics; the cardiac risk assessment level serves as a reference index for triaging a subject; and the cardiac risk type serves as a reference indicator for assessing the subject A reference indicator of the type of risk that may exist for your heart. 9. A computer device, comprising a memory and a processor, the memory stores a computer program, characterized in that when the processor executes the computer program, the method of any one of claims 1 to 7 is implemented. step. 10. A computer-readable storage medium with a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.