Coronary artery lesion risk prediction method, device and system and storage medium

Abstract

The invention discloses a coronary artery lesion risk prediction method, a coronary artery lesion risk processing device, a coronary artery lesion risk prediction system and a computer readable storage medium. Obtaining a point sequence according to high-frequency QRS wave group data obtained by the exercise electrocardiogram data; aiming at a point sequence corresponding to any electrocardiogram lead, taking any sampling point in a plurality of sampling points in the point sequence as a current reference sampling point in a traversal cycle, and traversing sampling points in a preset time interval from the current reference sampling point, determining whether a first sampling point and a second sampling point meeting a preset condition exist in the preset time interval, wherein the first sampling point is earlier than the second sampling point in time sequence; and if the preset condition is met, determining that the high-frequency QRS wave group data has waveform characteristics indicating the coronary artery lesion, thereby determining the possibility of the coronary artery lesion. Compared with a previous scheme, the scheme has the advantage that the qualitative accuracy of coronary artery lesion is improved.

Description

Technical Field

[0001] This application relates to the field of medical instrument technology, specifically to a method, processing device, system, and computer-readable storage medium for predicting the risk of coronary artery disease. Background Technology

[0002] With the continuous improvement of people's living standards and the increasing work pressure, heart disease is becoming increasingly prevalent among younger people, and heart health issues are receiving more and more attention. Therefore, accurately identifying heart health status is a crucial issue.

[0003] Currently, there are methods for assessing cardiac health based on electrocardiogram (ECG). Chinese patent application CN114742114A discloses a high-frequency QRS waveform curve analysis method, which assesses the degree of myocardial ischemia in coronary artery stenosis by analyzing ECG signals under exercise stress. However, the accuracy of this method can be further improved. Summary of the Invention

[0004] The purpose of this application is to provide a method, processing device, system, and computer-readable storage medium for predicting the risk of coronary artery lesions, which can more accurately and comprehensively predict and quantitatively analyze coronary artery lesions.

[0005] To achieve the above objectives, the first aspect of this application provides a method for predicting the risk of coronary artery disease, comprising: Acquire exercise electrocardiogram (ECG) data output from at least one ECG lead; The point sequence corresponding to each ECG lead is obtained from the high-frequency QRS complex data of each ECG lead obtained from the exercise ECG data. The point sequence includes multiple sampling points arranged in time sequence. For any ECG lead corresponding to a point sequence, in one iteration, any one of the multiple sampling points in the point sequence is taken as the current reference sampling point. The sampling points within a preset time interval starting from the current reference sampling point are traversed to determine whether there is a first sampling point and a second sampling point that meet the preset conditions within the preset time interval, wherein the first sampling point is earlier than the second sampling point in time. If a first sampling point and a second sampling point that meet the preset conditions are found, the waveform characteristics of the high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead are determined to indicate coronary artery lesions, thereby determining the possibility of coronary artery lesions. If it is determined that there is no first or second sampling point that meets the preset conditions, the current traversal loop ends and other sampling points in the point sequence other than the current reference sampling point are determined as new reference sampling points. For the new reference sampling point, perform the next traversal loop, repeat the traversal loop until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or all the sampling points that need to be traversed in the point sequence have been traversed.

[0006] In this embodiment of the application, the method for predicting the risk of coronary artery disease further includes: If, after all the sampling points that need to be traversed in the point sequence have been traversed, there are no first or second sampling points that meet the preset conditions, it is determined that the corresponding high-frequency QRS complex data does not contain waveform features that indicate coronary artery lesions. If the high-frequency QRS complex data corresponding to all electrocardiogram leads do not show waveform characteristics indicative of coronary artery disease, then the possibility of coronary artery disease is determined to be absent.

[0007] In this embodiment of the application, the preset conditions include: The relative decrease in amplitude between the first sampling point and the second sampling point reaches a first preset threshold, and the degree of voltage decrease between the first sampling point and the second sampling point reaches a second preset threshold; or The relative value of the amplitude decrease between the first sampling point and the second sampling point reaches the first preset threshold, or the degree of voltage decrease between the first sampling point and the second sampling point reaches the second preset threshold.

[0008] In this embodiment of the application, the preset time interval includes a preset duration. For any point sequence corresponding to an electrocardiogram lead, in one iteration, any one of the multiple sampling points in the point sequence is used as the current reference sampling point. The sampling points within the preset time interval starting from the current reference sampling point are traversed to determine whether there are first and second sampling points that meet preset conditions within the preset time interval, including: The first sampling point in the point sequence, which is the earliest in time, is taken as the current reference sampling point. The traversal starts from the next sampling point in time of the current reference sampling point to determine whether the current traversed sampling point and the current reference sampling point meet the preset conditions. If the current traversed sampling point and the current reference sampling point do not meet the preset conditions, determine whether the time interval between the current traversed sampling point and the current reference sampling point exceeds the preset duration; If the time interval between the current traversed sampling point and the current reference sampling point exceeds the preset duration, the traversal loop is terminated.

[0009] In this embodiment of the application, a new traversal loop is performed for the new reference sampling point, and the traversal loop is repeated until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or all the sampling points that need to be traversed in the point sequence have been traversed, including: The next traversal loop begins from the next sampling point in the time sequence of the new reference sampling point. The traversal loop is repeated until it is determined that there exists a current traversal sampling point and a current reference sampling point that meet the preset conditions, or the current traversal sampling point is the last sampling point in the point sequence.

[0010] In this embodiment, the preset time interval includes the window length of the window function. For any point sequence corresponding to an ECG lead, in one iteration, any one of the multiple sampling points in the point sequence is used as the current reference sampling point. The sampling points within the preset time interval starting from the current reference sampling point are traversed to determine whether there are first and second sampling points that meet preset conditions within the preset time interval, including: The window starts by taking the earliest sampling point in the point sequence as the starting point of the window. The sampling points in the point sequence contained in the window are traversed to determine the first sampling point with the largest root mean square voltage in the current window and the second sampling point with the smallest root mean square voltage that is located after the first sampling point in the time sequence. Determine whether the first sampling point and the second sampling point meet the preset conditions.

[0011] In this embodiment of the application, the next traversal loop is performed for the new reference sampling point until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or all the sampling points that need to be traversed in the point sequence have been traversed, including: The sampling points within the window after moving the preset step size are traversed again in the next loop. The loop is repeated until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or the end point of the current window reaches at least the last sampling point in the point sequence.

[0012] In this embodiment, the preset time interval includes the window length of the window function. For any point sequence corresponding to an ECG lead, in one iteration, any one of the multiple sampling points in the point sequence is used as the current reference sampling point. The sampling points within the preset time interval starting from the current reference sampling point are traversed to determine whether there are first and second sampling points that meet preset conditions within the preset time interval, including: The window starts by taking the earliest sampling point in the point sequence as the starting point and iterates through the sampling points in the point sequence contained in the window to determine whether there is a first sampling point and a second sampling point that meet the preset conditions in the current window.

[0013] In this embodiment of the application, a new traversal loop is performed for the new reference sampling point, and the traversal loop is repeated until it is determined that there are first and second sampling points that meet the preset conditions, or all sampling points that need to be traversed in the first point sequence have been traversed, including: The sampling points within the window after moving the preset step size are traversed again in the next loop. The loop is repeated until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or the end point of the current window reaches at least the last sampling point in the point sequence.

[0014] In this embodiment of the application, the method for predicting the risk of coronary artery disease further includes: Given a first sampling point and a second sampling point that satisfy preset conditions, determine the first-order difference of the mean square voltage of the sampling points between the first sampling point and the second sampling point in the point sequence, so as to determine the minimum sampling point corresponding to the minimum value and at least one maximum sampling point corresponding to the maximum value. Calculate the first root mean square voltage difference between the maximum and minimum sampling points; Calculate the second root mean square voltage difference between the first sampling point and the second sampling point; If the ratio of the first root mean square voltage difference to the second root mean square voltage difference is greater than or equal to the third threshold, the first sampling point and the second sampling point will be redefined as not meeting the preset conditions.

[0015] In this embodiment of the application, the method for predicting the risk of coronary artery disease further includes: Positive indicators in leads are identified by high-frequency QRS complex data with waveform characteristics that indicate coronary artery lesions. When the positive indicators in the leads are positive or borderline, the type of coronary artery disease is determined to be coronary artery stenosis; When a positive lead indicates a negative result, the type of coronary artery disease is determined to be coronary sclerosis.

[0016] A second aspect of this application provides a processing apparatus, including a processor and a memory, wherein instructions are stored in the memory, and the processor is configured to call and execute the instructions from the memory to implement the aforementioned method for predicting the risk of coronary artery lesions.

[0017] A third aspect of this application provides a cardiac health status assessment system, comprising: An electrocardiogram (ECG) signal acquisition device includes at least one electrode forming at least one ECG lead for acquiring ECG signals from a subject; and The aforementioned processing device.

[0018] A fourth aspect of this application provides a machine-readable storage medium storing instructions for causing a machine to perform the aforementioned coronary artery lesion risk prediction method.

[0019] The technical solution provided in this application, compared to focusing only on the steep drop wave in the first three minutes of exercise, determines the presence of a steep drop wave in the high-frequency QRS complex data throughout the entire load exercise detection process. This can improve the accuracy of the qualitative analysis of coronary artery lesions, thereby improving the sensitivity of coronary artery lesion identification without significantly reducing or even reducing specificity.

[0020] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0021] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 The diagram illustrates a high-frequency QRS waveform curve during motion, showing a steep drop in waveform after 3 minutes of motion.

[0022] Figure 2 An example flowchart illustrating a method for predicting the risk of coronary artery lesions according to an embodiment of this application is shown.

[0023] Figure 3 An example flowchart of a method for predicting the risk of coronary artery lesions according to another embodiment of this application is illustrated.

[0024] Figure 4 An example flowchart of a method for predicting the risk of coronary artery lesions according to another embodiment of this application is illustrated.

[0025] Figure 5 An example flowchart of a method for predicting the risk of coronary artery lesions according to another embodiment of this application is illustrated.

[0026] Figure 6 An example block diagram of a cardiac health status assessment system according to an embodiment of this application is shown schematically. Detailed Implementation

[0027] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0028] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0029] The coronary artery lesion risk prediction method provided in this 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, without specific limitations. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, electrocardiogram monitoring devices, and portable wearable devices, and the server can be a standalone server or a server cluster composed of multiple servers.

[0030] In the embodiments of this application, "exercise ECG data" or its equivalents refer to ECG data collected from the subject during exercise ECG testing. Exercise ECG testing increases the cardiac load by a certain amount of exercise to collect ECG data from the subject, and analyzes the subject's cardiac health status based on the collected ECG data. It is widely used in the detection of heart and cardiovascular diseases. Exercise ECG data may include multiple QRS complexes reflecting changes in the depolarization potential and time of the left and right ventricles. Each QRS complex is a set of Q waves, R waves, and S waves in the ECG. Based on the QRS complexes in the exercise ECG data, corresponding high-frequency QRS waveform curves can be obtained. The high-frequency QRS waveform curve can essentially be a curve obtained by connecting multiple sampling points in time sequence. The high-frequency QRS waveform curve is also called the high-frequency QRS time-intensity curve, which can be used to characterize the root mean square voltage of the high-frequency components of the subject's QRS complex over time during the entire exercise ECG testing process, that is, to reflect the energy change trend during the entire exercise ECG testing process. The high-frequency QRS waveform curve is presented through the high-frequency QRS waveform diagram. In the high-frequency QRS waveform diagram, the horizontal axis can be time, which corresponds to the detection time of the load exercise ECG detection process, and the unit can be min (minute). The vertical axis can be root mean square voltage (RMS voltage), which can also be understood as intensity or amplitude, and the unit can be uV (microvolt).

[0031] In one embodiment, the stress-exertion ECG detection process includes multiple stages, specifically three stages: a resting stage, an exercise stage, and a recovery stage. The exercise ECG data includes ECG data from each stage. It is understood that the stage division is not limited to this and can be adjusted according to actual circumstances. In this embodiment, the pre-exercise stage is the resting stage, the exercise stage is the exercise stage, and the post-exercise stage is the recovery stage.

[0032] In the embodiments of this application, "resting ECG data" or its equivalents can refer to ECG data collected during resting ECG testing. The subject is in a resting state during the resting ECG testing process. Resting ECG data may include multiple QRS complexes. High-frequency QRS complex data can be obtained by sequentially aligning, averaging, and high-frequency filtering the QRS complexes in the resting ECG data; alternatively, high-frequency QRS complex data can be obtained by sequentially performing high-frequency filtering, alignment, and averaging the QRS complexes in the resting ECG data; or, high-frequency ECG data can be obtained by analyzing the resting ECG data, extracting it, and then sequentially aligning and averaging the QRS complexes in the high-frequency ECG data. No specific limitations are made here. High-frequency QRS complex data corresponds to a high-frequency QRS envelope curve; a corresponding high-frequency QRS envelope curve can be formed based on the high-frequency QRS complex data. Therefore, it is evident that a corresponding high-frequency QRS envelope curve can be obtained by processing resting ECG data.

[0033] In the embodiments of this application, "conventional ECG data" or its equivalents can refer to ECG data that can be obtained from conventional electrocardiogram (ECG) signals. In the embodiments of this application, conventional ECG data can be obtained by analyzing and processing resting ECG data. For example, resting ECG data can be low-pass filtered to remove high-frequency components and retain low-frequency effective ECG components. The result after this filtering process is conventional ECG data.

[0034] Coronary artery disease can be categorized as large vessel disease of the coronary arteries, specifically including coronary artery stenosis and coronary artery sclerosis. When exercise electrocardiogram data indicates the possibility of coronary artery stenosis or coronary artery sclerosis, the presence of coronary artery disease is considered a possibility.

[0035] The method disclosed in Chinese Patent Application No. CN114742114A, filed by the applicant of this application, is used to determine the possibility of coronary artery stenosis. This application discloses a high-frequency QRS waveform curve analysis method, including acquiring a high-frequency QRS waveform curve (high-frequency QRS waveform curve) corresponding to exercise electrocardiogram data; selecting a high-frequency QRS waveform curve within a first time period as a first reference waveform curve; selecting the point with the minimum root-mean-square voltage from the first reference waveform curve as a first reference point, and the point earlier than the first reference point and with the largest root-mean-square voltage as a second reference point; determining a first amplitude decrease relative value based on the root-mean-square voltages of the first and second reference points; if the first amplitude decrease relative value is greater than or equal to a first preset threshold, determining that the high-frequency QRS waveform curve meets preset conditions, i.e., the high-frequency QRS waveform curve exhibits waveform characteristics indicative of coronary artery stenosis, thereby judging or determining the possibility of coronary artery stenosis.

[0036] The high-frequency QRS waveform analysis method disclosed in the above embodiments mainly focuses on whether there is a situation where the relative value of the first amplitude decrease is greater than or equal to a first preset threshold in the QRS waveform curve segment corresponding to the first 3 minutes of exercise. That is, it mainly determines whether there is a steep drop waveform in the high-frequency QRS waveform curve during the first 3 minutes of exercise. If so, it determines that there is coronary artery stenosis. However, in reality, the location (time point) where the relative value of the first amplitude decrease is greater than or equal to the first preset threshold in the high-frequency QRS waveform curve may be after 3 minutes, that is, the high-frequency QRS waveform curve shows a steep drop waveform after 3 minutes of exercise (e.g., Figure 1 As shown in the diagram, the method in the above embodiment may overlook this situation and fail to detect the presence of coronary artery stenosis. Furthermore, this method may also overlook the identification of the possibility of coronary artery sclerosis.

[0037] In view of this, the inventors of this application have proposed an innovative method for predicting the risk of coronary artery disease through further research. Compared with previously disclosed methods, the method provided in this embodiment can improve the accuracy of judging the possibility of coronary artery stenosis and can also judge the possibility of coronary artery sclerosis.

[0038] In the embodiments of this application, the general inventive concept for determining whether coronary artery lesions exist may include obtaining high-frequency QRS complex data based on exercise electrocardiogram data.

[0039] Specifically, in this embodiment, exercise ECG data can be analyzed to obtain high-frequency QRS complex data. Specifically, this can be achieved by analyzing the high-frequency components of the QRS complex in the exercise ECG data. More specifically, the exercise ECG data includes ECGs (electrocardiograms) corresponding to each heartbeat during the entire exercise ECG testing process, and the ECGs include QRS complexes. The exercise ECG data is divided into multiple subsets according to the time sequence and a preset step size using a window function. Each subset includes ECGs corresponding to multiple heartbeats. For each subset of ECG data, the ECG or QRS complexes corresponding to multiple heartbeats are sequentially aligned, averaged, and bandpass filtered to obtain the corresponding high-frequency QRS complexes (high-frequency bands of the QRS complex). The root mean square (RMS) of these high-frequency QRS complexes is then calculated to obtain the corresponding RMS voltage, which serves as the RMS voltage for that subset of ECG data. This yields a time-intensity data point set arranged in chronological order, with each data point corresponding to a time and a RMS voltage (or intensity or amplitude). The high-frequency QRS complex data can include this set of data points. It can be understood that the window length and preset movement step size of the window function can be customized according to actual needs. For example, the window length can be set to 10 seconds, and the preset movement step size can be 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. "In chronological order" refers to the order in which the signal acquisition time / the detection time progresses during the exercise ECG detection process.

[0040] High-frequency QRS complex data can include a set of data points arranged in time sequence (i.e., chronological order), where each data point corresponds to a time point (time stamp) and an amplitude value (or intensity value), which can be, for example, an RMS voltage value. The high-frequency QRS waveform curve can essentially be a curve obtained by connecting these data points in time sequence, or it can be a curve obtained by smoothing the curve to remove minor fluctuations.

[0041] A point sequence for each ECG lead can be obtained from high-frequency QRS complex data. This point sequence can include multiple data points sampled or selected from a set of data points in the high-frequency QRS complex data at a set sampling period (sampling interval) or sampling frequency. Each sample point in the point sequence is iterated to determine whether the high-frequency QRS complex data contains waveform features indicative of coronary artery disease, such as a steeply drooping wave. If a steeply drooping wave is present, the possibility of coronary artery disease can be determined.

[0042] Figure 2 An example flowchart illustrating a method for predicting the risk of coronary artery disease according to an embodiment of this application is shown. Figure 2 As shown, a method for predicting the risk of coronary artery disease may include the following steps.

[0043] In step S101, exercise electrocardiogram data output through at least one electrocardiogram lead is acquired.

[0044] Specifically, during electrocardiogram (ECG) acquisition, at least one electrode pad can be used for signal acquisition on the subject. For example, using 10 electrode pads, the pads can be distributed on the chest and limbs, forming 12 ECG leads (such as V1, V2, V3, V4, V5, V6, I, II, III, aVL, aVF, and aVR), corresponding to 12 sets of ECG data output. It should be understood that 10 electrode pads are only an example and are not intended to limit the specific number of electrode pads; more or fewer electrode pads can be used according to actual needs. Acquiring ECG signals from subjects undergoing strenuous exercise yields exercise ECG data.

[0045] In step S102, the point sequence (which can be called high-frequency QRS time-intensity point sequence) corresponding to each ECG lead is obtained based on the high-frequency QRS complex data corresponding to each ECG lead obtained from the exercise ECG data. The point sequence may include multiple sampling points arranged in time sequence.

[0046] In step S103, for any point sequence corresponding to an ECG lead, any one of the multiple sampling points in the point sequence is used as a reference sampling point. The sampling points within a preset time interval starting from the reference sampling point are traversed to determine whether there exists an earlier first sampling point and a later second sampling point that satisfy preset conditions within the preset time interval. The sampling points within this preset time interval include the reference sampling point. In one example, the preset conditions may be that the relative amplitude decrease between the first and second sampling points reaches a first preset threshold and the voltage decrease between the first and second sampling points reaches a second preset threshold (hereinafter referred to as the first preset condition). In another example, the preset conditions may be that the relative amplitude decrease between the first and second sampling points reaches the first preset threshold or the voltage decrease between the first and second sampling points reaches the second preset threshold (hereinafter referred to as the second preset condition).

[0047] The relative value of the amplitude decrease can be, for example,: (RMS voltage at the first sampling point - RMS voltage at the second sampling point) / RMS voltage at the first sampling point. In one example, the degree of voltage decrease can be the absolute value of the voltage decrease, which can be the RMS voltage at the first sampling point minus (-) the RMS voltage at the second sampling point. In an alternative example, the degree of voltage decrease can be the ratio of the absolute value of the voltage decrease to the maximum voltage value, i.e., (RMS voltage at the first sampling point - RMS voltage at the second sampling point) / maximum voltage value. This maximum voltage value can be understood as the maximum power, which can be used to reflect the subject's maximum cardiac pumping function, i.e., the peak value of the high-frequency QRS complex data (or high-frequency QRS waveform curve) will not exceed this maximum voltage value. This maximum voltage value can be determined based on the maximum value of the RMS voltage in the high-frequency QRS complex data corresponding to all ECG leads (e.g., the maximum RMS voltage rounded up to an even number). In this example, the second preset threshold can be in the form of a percentage, just like the first preset threshold.

[0048] In step S104, if preset conditions are met, the possibility of coronary artery disease is determined. During the traversal, if a first sampling point and a second sampling point that meet the preset conditions appear, it indicates that the high-frequency QRS complex data has waveform characteristics indicative of coronary artery disease, such as a steep drop wave. In this case, the possibility of coronary artery disease can be determined.

[0049] In step S105, if the preset conditions are not met, step S103 is repeated after changing the reference sampling point. In this step, if no first or second sampling point meeting the preset conditions appears after traversing all sampling points within the preset time interval, the current traversal loop ends. At this time, the reference sampling point can be changed, and the next traversal loop can be entered. For example, the adjacent sampling points of the current reference sampling point can be used as the new reference sampling point. The sampling points within the preset time interval starting from the new reference sampling point can be traversed using the new reference sampling point as the reference to determine whether there are first and second sampling points meeting the preset conditions. If there are first and second sampling points meeting the preset conditions, it indicates that the high-frequency QRS complex data has waveform characteristics indicating coronary artery lesions, and the traversal ends. If no first or second sampling point meeting the preset conditions appears after the current traversal loop ends, the next traversal loop is entered until a first or second sampling point meeting the preset conditions appears, or all sampling points in the point sequence that need to be traversed have been traversed. If the preset conditions are still not met after all the sampling points to be traversed, then it is determined that there are no waveform characteristics indicating coronary artery lesions.

[0050] There are multiple ways or rules for traversing the sampling points. Figure 3An example flowchart illustrating a method for predicting the risk of coronary artery disease according to an embodiment of this application is shown. Figure 3As shown, in this embodiment, for the high-frequency QRS complex data corresponding to each ECG lead, the earliest sampling point in the time sequence can be used as the reference sampling point (which can be called the initial reference point). The next sampling point in the time sequence of the initial reference point is used as the traversal starting point, and the traversal is performed according to the time sequence. The relative value of amplitude decrease and the degree of voltage decrease between the initial reference point (the first sampling point in this embodiment) and the currently traversed sampling point (the second sampling point in this embodiment) are calculated. It is determined whether the initial reference point and the currently traversed sampling point meet preset conditions. If these two sampling points meet the preset conditions, it can be determined that the high-frequency QRS complex data corresponding to the lead has waveform characteristics indicating coronary artery disease, such as a steep drop wave, thereby determining the possibility of coronary artery disease, and the traversal stops. If the preset conditions are not met, it is determined whether the time interval between the currently traversed sampling point and the initial reference point exceeds a preset duration, and whether the currently traversed sampling point is the last sampling point in the point sequence. If the time interval between the current traversed sampling point and the initial reference point does not exceed a preset duration and the current traversed sampling point is not the last sampling point, then switch to the next sampling point of the current traversed sampling point, and use this next sampling point as the new current traversed sampling point. The same operation is then performed on this new current traversed sampling point. Specifically, the relative value of amplitude decrease and the degree of voltage decrease between the initial reference point and the new current traversed sampling point are calculated. It is then determined whether the initial reference point and the new current traversed sampling point meet preset conditions. If the preset conditions are met, it can be determined that the high-frequency QRS complex data corresponding to this lead has a steep drop wave, thus confirming the possibility of coronary artery lesions, and the traversal stops. If the preset conditions are not met, the above operation can be repeated to traverse subsequent sampling points. If the preset conditions are still not met, then traversal can continue until the time interval between the traversed current traversed sampling point and the initial reference point reaches a preset duration (in this embodiment, the preset duration is the preset time interval mentioned above), or the current traversed sampling point is the last sampling point in the point sequence. If the time interval between the current traversed sampling point and the initial reference point reaches a preset duration, or if the current traversed sampling point is the last sampling point in the point sequence, then the process can switch to the next sampling point after the initial reference point, using this next sampling point as the reference sampling point, and proceeding with the next traversal loop in the same manner as described above. If the preset conditions are still not met, the traversal loop can continue until the preset conditions are met, or until all sampling points in the point sequence except the last sampling point have been used as reference sampling points to complete the traversal. In other words, if the preset conditions are still not met after the previous sampling point of the last sampling point has been used as the reference sampling point to complete the traversal, it indicates that the high-frequency QRS complex data corresponding to that ECG lead does not contain waveform characteristics indicative of coronary artery lesions, such as steep descent waves.

[0051] In this embodiment, the preset duration can range from, for example, 1 minute to 3 minutes. If the preset condition is that the relative amplitude decrease between the first sampling point and the second sampling point reaches a first preset threshold and the voltage decrease between the first sampling point and the second sampling point reaches a second preset threshold (first preset condition), then the first preset threshold can range from 30% to 40%. In this case, if the voltage decrease is the absolute value of the voltage decrease, then the second preset threshold can range from 1 uV to 2 uV. If the voltage decrease is the ratio of the absolute value of the voltage decrease to the maximum voltage value, then the second preset threshold can range from 10% to 20%.

[0052] If the preset condition is that the relative decrease in amplitude between the first and second sampling points reaches a first preset threshold, or the voltage decrease between the first and second sampling points reaches a second preset threshold (the first preset condition), then the value of the first preset threshold can range from 50% to 60%. In this case, the value of the second preset threshold is related to the preset duration. Generally speaking, the value of the second preset threshold is positively correlated with the preset duration; that is, the longer the preset duration, the larger the second preset threshold. For example, taking a preset duration ranging from 1 minute to 3 minutes as an example, if the preset duration is 1 minute, then when the voltage decrease is the absolute value of the voltage decrease, the value of the second preset threshold ranges from 1 uV to 2 uV; when the voltage decrease is the ratio of the absolute value of the voltage decrease to the maximum voltage value, the value of the second preset threshold ranges from 10% to 20%. If the preset duration is 2 minutes, then when the voltage drop is the absolute value of the voltage drop, the second preset threshold ranges from 2 uV to 3 uV; when the voltage drop is the ratio of the absolute value of the voltage drop to the maximum voltage value, the second preset threshold ranges from 20% to 30%. If the preset duration is 3 minutes, when the voltage drop is the absolute value of the voltage drop, the second preset threshold ranges from 3 uV to 4 uV; when the voltage drop is the ratio of the absolute value of the voltage drop to the maximum voltage value, the second preset threshold ranges from 30% to 40%. Those skilled in the art will understand that the specific values ​​of the preset thresholds above are exemplary and are used to illustrate the correlation between the second preset threshold and the preset duration. For a shorter preset duration, using a larger second preset threshold may misidentify a steeply drooping wave as a non-steeply drooping wave; conversely, for a longer preset duration, using a smaller second preset threshold may misidentify a non-steeply drooping wave as a steeply drooping wave, thus reducing the accuracy of identification. The specific values ​​can be adjusted according to the actual application.

[0053] The above processing can be performed on the high-frequency QRS complex data corresponding to all ECG leads. If at least one ECG lead's high-frequency QRS complex data shows waveform characteristics indicative of coronary artery disease, such as a steep descent wave, then the possibility of coronary artery disease can be determined. Conversely, if none of the high-frequency QRS complex data corresponding to all ECG leads shows waveform characteristics indicative of coronary artery disease, then the absence of coronary artery disease can be determined.

[0054] Although the above embodiments process multiple sampling points in a point sequence in a time sequence from early to late, those skilled in the art will understand that the above method is also applicable to multiple sampling points in a time sequence from late to early. In this case, the relative value of amplitude decrease and the degree of voltage decrease between the currently traversed sampling point and the reference sampling point are calculated (in this case, the currently traversed sampling point is earlier than the reference sampling point in time).

[0055] Figure 4 An example flowchart illustrating a method for predicting the risk of coronary artery disease according to another embodiment of this application is shown. Figure 4As shown, in this embodiment, a window function can be preset. The window of this window function includes a start point and an end point, and the distance between the start point and the end point is the window length (in this embodiment, the window length is the preset time interval mentioned above). The value range of the window length can be, for example, 1 minute to 3 minutes. For the high-frequency QRS complex data corresponding to each ECG lead, the window function is used to traverse the point sequence of the high-frequency QRS complex data. Specifically, the earliest sampling point in the time sequence can be used as the starting point of the window. The sampling point with the largest RMS voltage among the multiple sampling points in the point sequence included in the window (the first sampling point in this embodiment) and the sampling point with the smallest RMS voltage among the sampling points that are time-series after the first sampling point (the second sampling point in this embodiment) can be determined. The relative value of the amplitude decrease and the degree of voltage decrease between the first sampling point and the second sampling point are calculated. It is determined whether the first sampling point and the second sampling point meet the preset conditions. If these two sampling points meet the preset conditions, it can be determined that there is a steep drop wave in the high-frequency QRS complex data corresponding to the lead, and the possibility of coronary artery lesions can be determined, and the traversal stops. If the preset conditions are not met, the window of the window function can be slid backward by a preset step size, and the above operation can be repeated. If the preset conditions are still not met, the window can be slid backward by the preset step size again until the preset conditions are met, or until the end of the window reaches or exceeds the last sampling point in the point sequence. That is, if the preset conditions are not met even after the window has slid by the preset step size and its end reaches or exceeds the last sampling point, it indicates that there are no waveform characteristics of coronary artery lesions, such as steep descent waves, in the high-frequency QRS complex data of that ECG lead. In this embodiment, the preset step size can be, for example, an integer multiple of the time interval (e.g., sampling period) between two adjacent sampling points, preferably, the time interval between two adjacent sampling points, such as 10 seconds. Of course, those skilled in the art will understand that setting the preset step size to other values ​​is also feasible.

[0056] In this embodiment, if the preset condition is that the relative decrease in amplitude between the first sampling point and the second sampling point reaches a first preset threshold and the degree of voltage drop between the first sampling point and the second sampling point reaches a second preset threshold (first preset condition), then the value range of the first preset threshold can be 30% to 40%. In this case, if the degree of voltage drop is the absolute value of the voltage drop, then the value range of the second preset threshold can be 1 uV to 2 uV. If the degree of voltage drop is the ratio of the absolute value of the voltage drop to the maximum voltage value, then the value range of the second preset threshold can be 10% to 20%.

[0057] If the preset condition is that the relative decrease in amplitude between the first and second sampling points reaches a first preset threshold, or the voltage decrease between the first and second sampling points reaches a second preset threshold (the first preset condition), then the value of the first preset threshold can range from 50% to 60%. In this case, the value of the second preset threshold is related to the window length. Generally speaking, the value of the second preset threshold is positively correlated with the window length; that is, the longer the window length, the larger the second preset threshold. For example, taking a window length ranging from 1 minute to 3 minutes as an example, if the window length is 1 minute, and the voltage decrease is the absolute value of the voltage decrease, then the value of the second preset threshold ranges from 1 uV to 2 uV; if the voltage decrease is the ratio of the absolute value of the voltage decrease to the maximum voltage value, then the value of the second preset threshold ranges from 10% to 20%. If the window length is 2 minutes, the second preset threshold ranges from 2 uV to 3 uV when the voltage drop is the absolute value of the voltage drop, and from 20% to 30% when the voltage drop is the ratio of the absolute voltage drop to the maximum voltage value. If the window length is 3 minutes, the second preset threshold ranges from 3 uV to 4 uV when the voltage drop is the absolute value of the voltage drop, and from 30% to 40% when the voltage drop is the ratio of the absolute voltage drop to the maximum voltage value. Those skilled in the art will understand that the specific values ​​of the above preset thresholds are exemplary and intended to illustrate the correlation between the second preset threshold and the window length. For shorter window lengths, using a larger second preset threshold may misidentify steeply drooping waves as non-steeply drooping waves; for longer window lengths, using a smaller second preset threshold may misidentify non-steeply drooping waves as steeply drooping waves, thus reducing recognition accuracy. The specific values ​​can be adjusted according to actual applications.

[0058] The above processing can be performed on the high-frequency QRS complex data corresponding to all ECG leads. If at least one ECG lead's high-frequency QRS complex data shows waveform characteristics indicative of coronary artery disease, such as a steep descent wave, then the possibility of coronary artery disease can be determined. Conversely, if none of the high-frequency QRS complex data corresponding to all ECG leads shows waveform characteristics indicative of coronary artery disease, then the absence of coronary artery disease can be determined.

[0059] Although the window function used in the above embodiments starts from the earliest temporally sampled point in the point sequence, those skilled in the art will understand that it can also start from the latest temporally sampled point in the point sequence.

[0060] Figure 5 An example flowchart illustrating a method for predicting the risk of coronary artery disease according to another embodiment of this application is shown. Figure 5 As shown, in this embodiment, a window function can be preset. The window of this window function includes a start point and an end point, and the distance between the start point and the end point is the window length (in this embodiment, the window length is the preset time interval mentioned above). The value range of the window length can be, for example, 1 minute to 3 minutes. For the high-frequency QRS complex data corresponding to each ECG lead, the start point of the window can be aligned with the earliest sampling point in the time sequence (i.e., these two points coincide in time). Using this first sampling point as a reference point, each sampling point within the window is traversed sequentially. Specifically, the relative value of amplitude decrease and the degree of voltage decrease between the reference point and the currently traversed sampling point within the window can be calculated. It is determined whether the reference point and the currently traversed sampling point meet preset conditions. If these two sampling points meet the preset conditions, it can be determined that the high-frequency QRS complex data corresponding to the lead has waveform characteristics indicating coronary artery lesions, such as a steep drop wave, and the possibility of coronary artery lesions can be determined, and the traversal stops. If the preset conditions are not met, the next sampling point within the window is traversed. If the reference point and the currently traversed sampling point meet the preset conditions, the traversal stops; otherwise, it continues until the last sampling point within the traversal window. If the preset conditions are still not met after all sampling points within the window have been traversed, the traversal loop ends, and the window of the window function can be slid backward (time-wise backward) by a preset step size, and the above operation is repeated (the next traversal loop). If the preset conditions are still not met after the traversal loop ends, the window can be slid backward by the preset step size again until the preset conditions are met, or until the end of the window reaches or exceeds the last sampling point in the point sequence. That is, if the preset conditions are not met after the window has slid by the preset step size and its end reaches or exceeds the last sampling point, it indicates that there are no waveform characteristics indicating coronary artery lesions, such as steep descent waves, for the high-frequency QRS complex data of that ECG lead. In this embodiment, the preset step size can be, for example, an integer multiple of the time interval (e.g., sampling period) between two adjacent sampling points, preferably, the time interval between two adjacent sampling points, such as 10 seconds. Of course, those skilled in the field can understand that setting the preset step size to other values ​​is also feasible.

[0061] In this embodiment, if the preset condition is that the relative decrease in amplitude between the first sampling point and the second sampling point reaches a first preset threshold and the degree of voltage drop between the first sampling point and the second sampling point reaches a second preset threshold (first preset condition), then the value range of the first preset threshold can be 30% to 40%. In this case, if the degree of voltage drop is the absolute value of the voltage drop, then the value range of the second preset threshold can be 1 uV to 2 uV. If the degree of voltage drop is the ratio of the absolute value of the voltage drop to the maximum voltage value, then the value range of the second preset threshold can be 10% to 20%.

[0062] If the preset condition is that the relative decrease in amplitude between the first and second sampling points reaches a first preset threshold, or the voltage decrease between the first and second sampling points reaches a second preset threshold (the first preset condition), then the value of the first preset threshold can range from 50% to 60%. In this case, the value of the second preset threshold is related to the window length. Generally speaking, the value of the second preset threshold is positively correlated with the window length; that is, the longer the window length, the larger the second preset threshold. For example, taking a window length ranging from 1 minute to 3 minutes as an example, if the window length is 1 minute, and the voltage decrease is the absolute value of the voltage decrease, then the value of the second preset threshold ranges from 1 uV to 2 uV; if the voltage decrease is the ratio of the absolute value of the voltage decrease to the maximum voltage value, then the value of the second preset threshold ranges from 10% to 20%. If the window length is 2 minutes, the second preset threshold ranges from 2 uV to 3 uV when the voltage drop is the absolute value of the voltage drop, and from 20% to 30% when the voltage drop is the ratio of the absolute voltage drop to the maximum voltage value. If the window length is 3 minutes, the second preset threshold ranges from 3 uV to 4 uV when the voltage drop is the absolute value of the voltage drop, and from 30% to 40% when the voltage drop is the ratio of the absolute voltage drop to the maximum voltage value. Those skilled in the art will understand that the specific values ​​of the above preset thresholds are exemplary and intended to illustrate the correlation between the second preset threshold and the window length. For shorter window lengths, using a larger second preset threshold may misidentify steeply drooping waves as non-steeply drooping waves; for longer window lengths, using a smaller second preset threshold may misidentify non-steeply drooping waves as steeply drooping waves, thus reducing recognition accuracy. The specific values ​​can be adjusted according to actual applications.

[0063] The above processing can be performed on the high-frequency QRS complex data corresponding to all ECG leads. If at least one ECG lead's high-frequency QRS complex data shows waveform characteristics indicative of coronary artery disease, such as a steep descent wave, then the possibility of coronary artery disease can be determined. Conversely, if none of the high-frequency QRS complex data corresponding to all ECG leads shows waveform characteristics indicative of coronary artery disease, then the absence of coronary artery disease can be determined.

[0064] Although the window function used in the above embodiments starts from the earliest temporally sampled point in the point sequence, those skilled in the art will understand that it can also start from the latest temporally sampled point in the point sequence.

[0065] In the above embodiments, if two sampling points are determined to meet the preset conditions during the traversal process, but the high-frequency QRS complex data (or high-frequency QRS waveform curve) may show significant fluctuations between these two sampling points, the cause of this situation may be coronary artery spasm or other factors such as microcirculatory disturbance. To eliminate such interference factors, in the preferred embodiment of this application, after determining the first and second sampling points that meet the preset conditions, the first-order difference of the RMS voltage / amplitude / intensity of the sampling points between the first and second sampling points is calculated to determine the third sampling point (minimum point) corresponding to the minimum value and at least one fourth sampling point (maximum point) corresponding to the maximum value. Here, the minimum point refers to the point where the first-order difference result changes from negative to positive, and the maximum point refers to the point where the first-order difference result changes from positive to negative. For any fourth sampling point among at least one fourth sampling point, calculate the absolute value of the RMS voltage difference between the fourth sampling point and the third sampling point (referred to as the first RMS voltage difference), and calculate the RMS voltage difference between the first sampling point and the second sampling point (referred to as the second RMS voltage difference). If the ratio of the first RMS voltage difference to the second RMS voltage difference is greater than or equal to a third preset threshold, it is determined that there is a significant fluctuation amplitude, and it is determined that there are no waveform characteristics indicating coronary artery lesions between the two sampling points (the first sampling point and the second sampling point) in the high-frequency QRS complex data (it is considered that the first sampling point and the second sampling point do not meet the preset conditions). Then, the traversal continues according to the case where the preset conditions are not met as described in the above embodiment. If the ratio is less than the third preset threshold, the influence of significant fluctuation amplitude can be excluded, and the first sampling point and the second sampling point meet the preset conditions, and it can be determined that there is a steep drop wave. When there are multiple fourth sampling points, it is preferable that the ratio obtained for each fourth sampling point is less than the third preset threshold, so that the influence of significant fluctuation amplitude can be excluded. In one example, the third preset threshold can be, for example, 40%, 50%, etc. Using this method to eliminate the interference of significant fluctuation amplitude in determining the presence of steep drop waves can further improve the specificity of the assessment.

[0066] Compared to the methods for determining the presence of coronary artery stenosis disclosed in prior patent applications, the coronary artery lesion risk prediction method provided in this application can significantly improve sensitivity without significantly reducing or even reducing specificity.

[0067] In a further embodiment of this application, the type of coronary artery lesion can be further determined, i.e., coronary artery stenosis or coronary artery sclerosis. In this embodiment, the coronary artery lesion risk prediction method may further include the following steps.

[0068] In step S106, the positive indicators of the leads corresponding to the high-frequency QRS complex data with waveform characteristics indicating coronary artery lesions are determined.

[0069] Lead-positive indicators corresponding to each ECG lead can be obtained from high-frequency QRS complex data. These indicators can be obtained using methods disclosed in existing technologies. For example, in one instance, the relative and absolute values ​​of amplitude decrease can be calculated from high-frequency QRS complex data to form the lead-positive indicators. A predefined function can be used to process the high-frequency QRS complex data to calculate the relative and absolute values ​​of amplitude decrease between two reference points where the RMS voltage decreases most rapidly and significantly in the high-frequency QRS waveform during, before, and after exercise. The absolute value of amplitude is obtained by subtracting the RMS voltages of the two reference points, while the relative value of amplitude decrease is calculated as a percentage of the absolute value of amplitude relative to the RMS voltage value of the reference point with the highest RMS voltage. The relative and absolute values ​​of amplitude decrease primarily assess changes in cardiac blood flow during exercise.

[0070] In step S107, if the positive indicator in the first lead is positive, the type of coronary artery lesion is determined to be coronary artery stenosis. Specifically, if a borderline situation (i.e., a state between positive and negative) is considered, a positive indicator in the first lead includes both positive and borderline situations.

[0071] In step S108, if the positive indicator in the first lead is negative, the type of coronary artery lesion is determined to be coronary sclerosis.

[0072] As described above, when the possibility of coronary artery disease is determined, the maximum relative amplitude decrease and the maximum voltage decrease are determined from the relative amplitude decrease and voltage decrease (e.g., the absolute voltage decrease, or the ratio of the absolute voltage decrease to the maximum voltage value) for each electrocardiogram lead. The coronary artery disease is then quantitatively assessed based on the maximum relative amplitude decrease and the maximum voltage decrease.

[0073] For example, coronary artery lesions can be classified into multiple risk levels (e.g., D1, D2, D3, etc.). Each risk level corresponds to a range of relative values ​​for the maximum amplitude decrease and a range of ranges for the maximum voltage decrease. The risk level is determined based on the ranges where the relative value for the maximum amplitude decrease and the maximum voltage decrease fall. If the risk level corresponding to the relative value for the maximum amplitude decrease differs from the risk level corresponding to the maximum voltage decrease, the more severe risk level is taken as the final risk level.

[0074] In this application embodiment, a processor is provided, configured to execute the coronary artery lesion risk prediction method described in the above embodiments.

[0075] In this application embodiment, a machine-readable storage medium is provided, on which instructions are stored, which are used to cause a machine to perform the coronary artery lesion risk prediction method described in the above embodiments.

[0076] In this application embodiment, a processing device is provided, including a processor and a memory, wherein instructions are stored in the memory, and the processor is configured to call and execute the instructions from the memory to implement the coronary artery lesion risk prediction method described in the above embodiments.

[0077] Examples of processors may include, but are not limited to, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), and application-specific integrated circuits (ASICs).

[0078] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0079] Figure 6 An example block diagram of a cardiac health status assessment system according to an embodiment of this application is illustrated. Figure 6 As shown in the embodiments of this application, a cardiac health status assessment system is provided, which may include: An electrocardiogram (ECG) signal acquisition device 100 includes at least one electrode forming at least one ECG lead for acquiring ECG signals from a subject; and The aforementioned processing device 200.

[0080] The ECG signal acquisition device 100 may also include a display configured to display various data and information, including but not limited to, subject information such as age, height, weight, etc., and ECG signals / QRS complex data / QRS waveform curves acquired through ECG leads.

[0081] The processing device 200 can be configured to perform the various methods described in the above embodiments.

[0082] Specifically, in one embodiment, a computer-executed method for predicting the risk of coronary artery disease may include: Acquire the exercise electrocardiogram signal of the subject through at least one electrocardiogram lead during the stress exercise electrocardiogram test; Analyze the exercise electrocardiogram (ECG) signals to obtain exercise ECG data; The first point sequence corresponding to each ECG lead is obtained based on the first high-frequency QRS complex data corresponding to each ECG lead obtained from exercise ECG data. The first point sequence includes multiple sampling points arranged in time sequence. For any ECG lead corresponding to the first point sequence, in one iteration, any one of the multiple sampling points in the first point sequence is taken as the current reference sampling point. The sampling points within the first preset time interval starting from the current reference sampling point are traversed to determine whether there are first sampling points and second sampling points that meet preset conditions within the first preset time interval. The first sampling point is earlier than the second sampling point in time, and the sampling points within the first preset time interval include the current reference sampling point. If a first sampling point and a second sampling point that meet the preset conditions are determined, the first waveform feature of the first high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead is determined to indicate coronary artery lesions, thereby determining the possibility of coronary artery lesions. If it is determined that there is no first sampling point or second sampling point that meets the preset conditions, the current traversal loop ends and other sampling points in the first point sequence other than the current reference sampling point are determined as new reference sampling points. For the new reference sampling point, perform the next traversal loop, repeat the traversal loop until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or all the sampling points that need to be traversed in the first point sequence have been traversed.

[0083] In one embodiment, determining the likelihood of coronary artery disease based on exercise electrocardiogram data further includes: If, after all the sampling points that need to be traversed in the first point sequence have been traversed, there are no first sampling points or second sampling points that meet the preset conditions, it is determined that the corresponding first high-frequency QRS complex data does not contain the first waveform feature indicating coronary artery lesions. If the first high-frequency QRS complex data corresponding to all electrocardiogram leads do not show the first waveform characteristics that indicate coronary artery disease, then the possibility of coronary artery disease is determined to be absent.

[0084] In one embodiment, the preset conditions include: The relative decrease in amplitude between the first sampling point and the second sampling point reaches a first preset threshold, and the degree of voltage decrease between the first sampling point and the second sampling point reaches a second preset threshold; or The relative value of the amplitude decrease between the first sampling point and the second sampling point reaches the first preset threshold, or the degree of voltage decrease between the first sampling point and the second sampling point reaches the second preset threshold.

[0085] In one embodiment, the first preset time interval includes a preset duration. For any first point sequence corresponding to an ECG lead, in one iteration, any one of the multiple sampling points in the first point sequence is used as the current reference sampling point. The sampling points within the first preset time interval starting from the current reference sampling point are traversed to determine whether there are first and second sampling points that satisfy preset conditions within the first preset time interval, including: The first sampling point in the first point sequence is taken as the current reference sampling point. The traversal starts from the next sampling point in the time sequence of the current reference sampling point to determine whether the current traversed sampling point and the current reference sampling point meet the preset conditions. If the current traversed sampling point and the current reference sampling point do not meet the preset conditions, determine whether the time interval between the current traversed sampling point and the current reference sampling point exceeds the preset duration; If the time interval between the current traversed sampling point and the current reference sampling point exceeds the preset duration, the traversal loop is terminated.

[0086] In one embodiment, upon determining the existence of a first sampling point and a second sampling point that meet preset conditions, the system determines that the first high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead exhibits a first waveform feature indicative of coronary artery lesions, thereby determining the likelihood of coronary artery lesions, including: If, during the traversal cycle, it is determined that there are current traversal sampling points and current reference sampling points that meet preset conditions, the first high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead is determined to have the first waveform feature indicating coronary artery lesions, thereby determining the possibility of coronary artery lesions.

[0087] In one embodiment, when the current traversal loop ends, other sampling points in the first point sequence besides the current reference sampling point are determined as new reference sampling points, including: The next sampling point in the time sequence of the current reference sampling point is determined as the new reference sampling point.

[0088] In one embodiment, a new traversal loop is performed for the new reference sampling point, and the traversal loop is repeated until it is determined that there are first and second sampling points that meet preset conditions, or all sampling points that need to be traversed in the first point sequence have been traversed, including: The next traversal loop begins from the next sampling point in the time sequence of the new reference sampling point. The traversal loop is repeated until it is determined that there exists a current traversal sampling point and a current reference sampling point that satisfy the preset conditions, or the current traversal sampling point is the last sampling point in the first point sequence.

[0089] In one embodiment, the first preset time interval includes the window length of the window function. For any first point sequence corresponding to an ECG lead, in one iteration, any one of the multiple sampling points in the first point sequence is used as the current reference sampling point. The sampling points within the first preset time interval starting from the current reference sampling point are traversed to determine whether there are first and second sampling points that satisfy preset conditions within the first preset time interval, including: Using the earliest sampling point in the first point sequence as the starting point of the window, the sampling points in the first point sequence contained in the window are traversed to determine the first sampling point with the largest root mean square (RMS) voltage and the second sampling point with the smallest RMS voltage after the first sampling point in the current window. Determine whether the first sampling point and the second sampling point meet the preset conditions.

[0090] In one embodiment, upon determining the existence of a first sampling point and a second sampling point that meet preset conditions, the system determines that the first high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead exhibits a first waveform feature indicative of coronary artery lesions, thereby determining the likelihood of coronary artery lesions, including: If the first and second sampling points within the current window meet the preset conditions, the first waveform feature of the first high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead is determined to indicate coronary artery lesions, thereby determining the possibility of coronary artery lesions.

[0091] In one embodiment, if it is determined that there are no first sampling points and second sampling points that satisfy preset conditions, the current traversal loop ends, and other sampling points in the first point sequence other than the current reference sampling point are determined as new reference sampling points, including: If it is determined that there are no first or second sampling points that meet the preset conditions within the current window, the current traversal loop ends, and the window is moved backward by a preset step in time.

[0092] In one embodiment, the next traversal loop is performed for the new reference sampling point until it is determined that there are first and second sampling points that satisfy preset conditions, or all sampling points in the first point sequence that need to be traversed have been traversed, including: The sampling points within the window after the preset step size are moved are traversed again. The traversal is repeated until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or the end point of the current window reaches at least the last sampling point in the first point sequence.

[0093] In one embodiment, the first preset time interval includes the window length of the window function. For any first point sequence corresponding to an ECG lead, in one iteration, any one of the multiple sampling points in the first point sequence is used as the current reference sampling point. The sampling points within the first preset time interval starting from the current reference sampling point are traversed to determine whether there are first and second sampling points that satisfy preset conditions within the first preset time interval, including: Using the earliest sampling point in the first point sequence as the starting point of the window, the sampling points in the first point sequence contained in the window are traversed to determine whether there are first and second sampling points that meet the preset conditions in the current window.

[0094] In one embodiment, upon determining the existence of a first sampling point and a second sampling point that meet preset conditions, the system determines that the first high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead exhibits a first waveform feature indicative of coronary artery lesions, thereby determining the likelihood of coronary artery lesions, including: If the first and second sampling points within the current window meet the preset conditions, the first waveform feature of the first high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead is determined to indicate coronary artery lesions, thereby determining the possibility of coronary artery lesions.

[0095] In one embodiment, if it is determined that there are no first sampling points and second sampling points that satisfy preset conditions, the current traversal loop ends, and other sampling points in the first point sequence other than the current reference sampling point are determined as new reference sampling points, including: If it is determined that there are no first or second sampling points that meet the preset conditions within the current window, the current traversal loop ends, and the window is moved backward by a preset step in time.

[0096] In one embodiment, a new traversal loop is performed for the new reference sampling point, and the traversal loop is repeated until it is determined that there are first and second sampling points that meet preset conditions, or all sampling points that need to be traversed in the first point sequence have been traversed, including: The sampling points within the window after the preset step size are moved are traversed again. The traversal is repeated until it is determined that there is a first sampling point and a second sampling point that meet the preset conditions, or the end point of the current window reaches at least the last sampling point in the first point sequence.

[0097] In one embodiment, determining the likelihood of coronary artery disease based on exercise electrocardiogram data further includes: Given a first sampling point and a second sampling point that satisfy preset conditions, the first-order difference of the root mean square voltage of the sampling points between the first sampling point and the second sampling point in the first point sequence is determined, so as to determine the minimum sampling point corresponding to the minimum value and at least one maximum sampling point corresponding to the maximum value. Calculate the first root mean square voltage difference between the maximum and minimum sampling points; Calculate the second root mean square voltage difference between the first sampling point and the second sampling point; If the ratio of the first root mean square voltage difference to the second root mean square voltage difference is greater than or equal to the third threshold, the first sampling point and the second sampling point will be redefined as not meeting the preset conditions.

[0098] In one embodiment, determining the likelihood of coronary artery disease based on exercise electrocardiogram data further includes: The first high-frequency QRS complex data corresponding to the first lead is used to identify the first waveform feature that indicates the presence of coronary artery lesions. When the positive indicator in the first lead is positive or borderline, the type of coronary artery disease is determined to be coronary artery stenosis; If the positive result in the first lead indicates a negative result, the type of coronary artery disease is determined to be coronary atherosclerosis.

[0099] In one embodiment, determining the likelihood of myocardial bridging based on exercise electrocardiogram data includes: Candidate first high-frequency QRS complex data are determined from the first high-frequency QRS complex data corresponding to each electrocardiogram lead. The candidate first high-frequency QRS complex data does not have the first waveform characteristics that indicate coronary artery lesions. The second point sequence is obtained based on the candidate first high-frequency QRS group data. This second point sequence includes multiple sampling points arranged in time sequence. For any candidate first high-frequency QRS group data corresponding to the second point sequence, in one traversal loop, any one of the multiple sampling points in the second point sequence is taken as the current reference sampling point, and the sampling points in the second preset time interval starting from the current reference sampling point are traversed to determine whether there are a third sampling point and a fourth sampling point that satisfy another preset condition in the second preset time interval, wherein the third sampling point is earlier than the fourth sampling point in time. If it is determined that there are third and fourth sampling points that meet another preset condition, the second waveform feature of the arbitrary candidate first high-frequency QRS complex data that indicates myocardial bridging is determined, thereby determining the possibility of myocardial bridging. If it is determined that there are no third or fourth sampling points that satisfy another preset condition, the current traversal loop ends, and other sampling points in the second point sequence other than the current reference sampling point are determined as new reference sampling points; For the new reference sampling point, perform the next traversal loop until it is determined that there are a third and fourth sampling points that satisfy another preset condition, or all the sampling points that need to be traversed in the second point sequence have been traversed.

[0100] The technical solution provided in this application, compared to focusing only on the steep drop wave in the first three minutes of exercise, determines the presence of a steep drop wave in the high-frequency QRS complex data throughout the entire load exercise detection process. This can improve the accuracy of the qualitative analysis of coronary artery lesions, thereby improving the sensitivity of coronary artery lesion identification without significantly reducing or even reducing specificity.

[0101] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0103] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0104] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0105] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0106] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0107] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0108] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0109] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method of predicting the risk of coronary artery disease, characterized by, The method comprises: obtaining motion electrocardiogram data output by at least one electrocardiogram lead; obtaining a point sequence corresponding to each electrocardiogram lead from high-frequency QRS complex data corresponding to each electrocardiogram lead obtained from the motion electrocardiogram data, the point sequence comprising a plurality of sampling points arranged in time sequence; for the point sequence corresponding to any electrocardiogram lead, in an iteration loop, taking any sampling point in the plurality of sampling points in the point sequence as a current reference sampling point, and traversing the sampling points in a preset time interval from the current reference sampling point to determine whether there are first and second sampling points satisfying a preset condition in the preset time interval, wherein the first sampling point is earlier in time sequence than the second sampling point; in the case where it is determined that there are first and second sampling points satisfying the preset condition, it is determined that the high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead has waveform characteristics indicating coronary artery lesion, thereby determining the possibility of coronary artery lesion; in the case where it is determined that there are no first and second sampling points satisfying the preset condition, the current iteration loop ends, and other sampling points in the point sequence except the current reference sampling point are determined as new reference sampling points; the next iteration loop is performed for the new reference sampling point, and the iteration loop is repeated until it is determined that there are first and second sampling points satisfying the preset condition, or all the sampling points to be traversed in the point sequence have been traversed.

2. The coronary artery lesion risk prediction method according to claim 1, characterized by, The method further comprises: in the case where there are no first and second sampling points satisfying the preset condition after all the sampling points to be traversed in the point sequence have been traversed, it is determined that the high-frequency QRS complex data corresponding to the arbitrary electrocardiogram lead has no waveform characteristics indicating coronary artery lesion; in the case where all the high-frequency QRS complex data corresponding to the electrocardiogram leads has no waveform characteristics indicating coronary artery lesion, it is determined that there is no possibility of coronary artery lesion.

3. The coronary artery lesion risk prediction method according to claim 1, characterized by, The preset condition comprises: the amplitude drop relative value between the first sampling point and the second sampling point reaches a first preset threshold value, and the voltage drop degree between the first sampling point and the second sampling point reaches a second preset threshold value; or the amplitude drop relative value between the first sampling point and the second sampling point reaches the first preset threshold value, or the voltage drop degree between the first sampling point and the second sampling point reaches the second preset threshold value.

4. The coronary artery lesion risk prediction method according to claim 1, characterized by, The preset time interval comprises a preset time length, and for the point sequence corresponding to any electrocardiogram lead, in an iteration loop, taking any sampling point in the plurality of sampling points in the point sequence as a current reference sampling point, and traversing the sampling points in the preset time interval from the current reference sampling point to determine whether there are first and second sampling points satisfying a preset condition in the preset time interval, comprising: taking the first sampling point earliest in time sequence in the point sequence as the current reference sampling point, and starting to traverse from the next sampling point in time sequence of the current reference sampling point to determine whether the current traversal sampling point and the current reference sampling point satisfy the preset condition; in the case where the current traversal sampling point and the current reference sampling point do not satisfy the preset condition, it is determined whether the time interval between the current traversal sampling point and the current reference sampling point exceeds the preset time length; In a case where it is determined that a time interval between the current traversal sampling point and the current reference sampling point exceeds a preset time length, it is determined that the traversal loop ends.

5. The coronary artery lesion risk prediction method according to claim 4, characterized by, A next traversal loop is performed for a new reference sampling point, and the traversal loop is repeated until it is determined that there are the first sampling point and the second sampling point satisfying the preset condition, or the sampling points that need to be traversed in the point sequence are all traversed, including: A next traversal loop is performed from a next sampling point of the new reference sampling point in time sequence, and the traversal loop is repeated until it is determined that there are the current traversal sampling point and the current reference sampling point satisfying the preset condition, or the current traversal sampling point is the last sampling point in the point sequence.

6. The coronary artery lesion risk prediction method of claim 1, wherein, The preset time interval includes a window length of a window function, and for a point sequence corresponding to any electrocardiogram lead, in a traversal loop, any sampling point of a plurality of sampling points in the point sequence is taken as a current reference sampling point, sampling points in a preset time interval from the current reference sampling point are traversed to determine whether there are the first sampling point and the second sampling point satisfying the preset condition in the preset time interval, including: The first sampling point and the second sampling point are determined whether to satisfy the preset condition. A next traversal loop is performed for a new reference sampling point, and the traversal loop is repeated until it is determined that there are the first sampling point and the second sampling point satisfying the preset condition, or the sampling points that need to be traversed in the point sequence are all traversed, including:

7. The coronary artery lesion risk prediction method of claim 1, wherein, A next traversal loop is performed for a new reference sampling point, and the traversal loop is repeated until it is determined that there are the first sampling point and the second sampling point satisfying the preset condition, or the sampling points that need to be traversed in the point sequence are all traversed, including: The preset time interval includes a window length of a window function, and for a point sequence corresponding to any electrocardiogram lead, in a traversal loop, any sampling point of a plurality of sampling points in the point sequence is taken as a current reference sampling point, sampling points in a preset time interval from the current reference sampling point are traversed to determine whether there are the first sampling point and the second sampling point satisfying the preset condition in the preset time interval, including:

8. The coronary artery lesion risk prediction method of claim 1, wherein, The first sampling point and the second sampling point are determined whether to satisfy the preset condition. A next traversal loop is performed for a new reference sampling point, and the traversal loop is repeated until it is determined that there are the first sampling point and the second sampling point satisfying the preset condition, or the sampling points that need to be traversed in the point sequence are all traversed, including:

9. The coronary artery lesion risk prediction method of claim 8, wherein, ​ The next traversal loop is performed on the sampling points in the window after the preset step length is moved, and the traversal loop is repeated until it is determined that there are first sampling points and second sampling points that meet the preset condition, or the end point of the current window reaches at least the last sampling point in the point sequence.

10. The coronary artery lesion risk prediction method of claim 1, wherein, Further comprising: In a case where the first sampling point and the second sampling point that meet the preset condition are determined, a first-order difference of mean square error voltages of the sampling points between the first sampling point and the second sampling point in the point sequence is determined to determine a minimum value sampling point corresponding to a minimum value and at least one maximum value sampling point corresponding to a maximum value; A first root mean square voltage difference value between the maximum value sampling point and the minimum value sampling point is calculated; A second root mean square voltage difference value between the first sampling point and the second sampling point is calculated; In a case where a ratio of the first root mean square voltage difference value to the second root mean square voltage difference value is greater than or equal to a third threshold value, the first sampling point and the second sampling point are re-determined as not meeting the preset condition.

11. The coronary artery lesion risk prediction method of claim 1, wherein, Further comprising: A lead positive indicator corresponding to high-frequency QRS complex data of a waveform feature indicating a coronary artery lesion is determined to exist; In a case where the lead positive indicator indicates positive or critical, the type of the coronary artery lesion is determined to be coronary stenosis; In a case where the lead positive indicator indicates negative, the type of the coronary artery lesion is determined to be coronary sclerosis.

12. A processing device, characterized by A processor and a memory are included, and the memory stores instructions, and the processor is configured to call and execute the instructions from the memory to implement the coronary artery lesion risk prediction method according to any one of claims 1 to 11.

13. A cardiac health status assessment system, characterized by, Comprising: An electrocardiosignal acquisition device including at least one electrode forming at least one electrocardiogram lead for acquiring an electrocardiosignal of a subject; And The processing device according to claim 12.

14. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing a machine to perform the coronary artery lesion risk prediction method according to any one of claims 1 to 11.

Images