An interventional procedure risk assessment method and system

Abstract

The present application relates to the field of health care information technology, in particular to an interventional operation risk assessment method and system, first, the abnormal period curve is divided in the waveform curve to be analyzed; in the abnormal period curve, according to the change situation of the periodic abnormality metric of periodic fluctuation curve with time sequence, the recovery ability index of abnormal period curve is acquired;According to the overall distribution of the periodic abnormality metric of all periodic fluctuation curves included in the abnormal period curve, the electrocardiogram abnormality index of the abnormal period curve is acquired;In the waveform curve to be analyzed, according to the recovery ability index and electrocardiogram abnormality index of all abnormal period curves, the local risk index of the waveform curve to be analyzed is acquired;According to the local risk index of the electrocardiogram waveform curve corresponding to all leads of the patient, the interventional operation risk of the patient is evaluated. The present application excavates the recovery ability situation and overall risk situation of the heart abnormality in depth to improve the risk assessment accuracy.

Description

Technical Field

[0001] This invention relates to the field of healthcare information technology, specifically to a method and system for risk assessment of interventional procedures. Background Technology

[0002] Coronary artery disease (CAD) is a heart condition caused by atherosclerosis of the coronary arteries, leading to narrowing or blockage of the arteries and subsequently causing myocardial ischemia, hypoxia, or necrosis. For some CAD patients, medication may not effectively improve the narrowing or blockage of the coronary arteries. In such cases, percutaneous coronary intervention (PCI) is one of the preferred treatment options. PCI, a minimally invasive cardiac procedure, uses catheter technology to open narrowed or blocked coronary arteries, restoring myocardial blood flow. Because PCI carries certain risks, risk assessment for CAD interventional procedures can effectively help physicians understand the patient's surgical risks.

[0003] Coronary CT angiography (CCTA) is widely used in clinical practice as the preferred non-invasive imaging examination for assessing coronary artery disease. It reconstructs three-dimensional images of the coronary arteries through intravenous injection of contrast agent and CT scans, assessing vascular stenosis, plaque location, and the degree of calcification. It can quantify the coronary artery calcium score (Agatston score) to aid in risk assessment. However, due to the influence of factors such as coronary artery calcification, CCTA often overestimates the degree of vascular stenosis, leading to a high false positive rate. Furthermore, the degree of coronary artery stenosis does not perfectly match the degree of myocardial ischemia, causing some patients without evidence of myocardial ischemia to undergo unnecessary invasive coronary angiography (ICA).

[0004] Spectroscopic CT, as a quantitative imaging technique, can obtain various quantitative parameter maps after image post-processing, including iodine concentration (IC) maps and effective atomic number (Zeff) maps. IC values ​​have been proven to accurately quantify myocardial iodine concentration, providing a new means for assessing myocardial perfusion. In existing technologies, wearable devices (such as ECG monitoring and early warning systems) can monitor the ECG waveforms of patients with coronary artery disease in real time and determine whether the monitored ECG waveforms exceed preset ranges to achieve real-time assessment of the patient's interventional surgery risk. This risk assessment method can dynamically capture electrophysiological abnormalities (such as asymptomatic myocardial ischemia) that cannot be reflected in real time by CCTA or spectral CT. However, when using real-time monitoring of the ECG waveforms of patients with coronary artery disease to assess the risk of interventional surgery, existing technologies fail to fully consider that the ECG waveforms can reflect the heart's recovery capacity and overall risk in real time, making it difficult to accurately assess the patient's interventional surgery risk. Summary of the Invention

[0005] The present invention addresses the technical problem of accurately assessing the risks of interventional surgeries using existing technologies. The purpose of this invention is to provide a method and system for assessing the risks of interventional surgeries, and the specific technical solution adopted is as follows:

[0006] A method for risk assessment of interventional surgery, the method comprising:

[0007] Acquire a set of electrocardiogram (ECG) data of the patient; the ECG data set includes ECG waveform curves corresponding to each lead;

[0008] Using any one of the aforementioned electrocardiogram waveform curves as the waveform curve to be analyzed; dividing the waveform curve to be analyzed based on the similarity of its local morphological features to obtain each periodic fluctuation curve of the waveform curve to be analyzed; obtaining the periodic abnormality measure of the periodic fluctuation curve based on the morphological feature differences between each periodic fluctuation curve in the waveform curve to be analyzed; and dividing the abnormal time period curve from the waveform curve to be analyzed based on the periodic abnormality measure of the periodic fluctuation curve.

[0009] In the abnormal time period curve, the recovery ability index of the abnormal time period curve is obtained based on the change of the periodic abnormality measure of the periodic fluctuation curve over time; the electrocardiogram abnormality index of the abnormal time period curve is obtained based on the overall distribution of the periodic abnormality measure of all the periodic fluctuation curves included in the abnormal time period curve; in the waveform curve to be analyzed, the local risk index of the waveform curve to be analyzed is obtained based on the recovery ability index of all the abnormal time period curves and the electrocardiogram abnormality index; the interventional surgery risk of the patient is assessed based on the local risk index of the electrocardiogram waveform curve corresponding to all leads of the patient.

[0010] Furthermore, the method for obtaining the periodic fluctuation curve includes:

[0011] Based on the fluctuation of the waveform curve to be analyzed, the characteristic fluctuation peaks of the waveform curve to be analyzed are obtained;

[0012] Based on the similarity of morphological features among the various characteristic fluctuation peaks, the similarity parameters of the characteristic fluctuation peaks are obtained;

[0013] Based on the similarity parameters of the characteristic fluctuation peaks, the waveform curve to be analyzed is divided to obtain the periodic fluctuation curves of the waveform curve to be analyzed.

[0014] Furthermore, the method for obtaining the similarity parameters includes:

[0015] The interval between the maximum and minimum abscissa values ​​corresponding to the characteristic fluctuation peak is taken as the total width of the characteristic fluctuation peak; the interval between the maximum and minimum ordinate values ​​corresponding to the characteristic fluctuation peak is taken as the total height of the characteristic fluctuation peak.

[0016] In the waveform curve to be analyzed, one characteristic fluctuation peak is randomly selected as the target fluctuation peak, and all other characteristic fluctuation peaks are used as reference fluctuation peaks. The absolute value of the difference between the total width of the target fluctuation peak and the reference fluctuation peaks is calculated to obtain the local width difference value of the reference fluctuation peak. The mean of the local width difference values ​​of all reference fluctuation peaks is calculated to obtain the overall width difference value of the target fluctuation peak. The absolute value of the difference between the total height of the target fluctuation peak and the reference fluctuation peaks is calculated to obtain the local height difference value of the reference fluctuation peak. The mean of the local height difference values ​​of all reference fluctuation peaks is calculated to obtain the overall height difference value of the target fluctuation peak. The product of the overall width difference value and the overall height difference value is calculated and negatively correlated to obtain the similarity parameter of the target fluctuation peak.

[0017] Furthermore, the method for obtaining the periodic anomaly measurement includes:

[0018] Select one of the periodic fluctuation curves in the waveform curves to be analyzed as the target period, and use all the other periodic fluctuation curves as reference periods.

[0019] Calculate the average duration of all reference periods for the target period to obtain the overall reference duration of the target period; calculate the absolute value of the difference between the duration of the target period and the overall reference duration to obtain the first difference value of the target period.

[0020] The mean of the similarity parameters of all characteristic fluctuation peaks in the target period is calculated and negative correlation mapping is performed to obtain the second difference value of the target period;

[0021] Calculate the product of the first difference value and the second difference value and perform normalization to obtain the periodic anomaly measure of the target period.

[0022] Furthermore, the method for obtaining the abnormal time period curve includes:

[0023] The periodic fluctuation curves whose periodic anomaly measurement is greater than the preset periodic anomaly threshold are taken as abnormal periodic curves; the abnormal periodic curves that are consecutively above a preset statistical number are taken as abnormal time period curves.

[0024] Furthermore, the method for obtaining the recovery capability index includes:

[0025] In the abnormal period curve, the difference between each periodic fluctuation curve and the periodic anomaly measure of the previous periodic fluctuation curve in the time series dimension is calculated and negatively correlated to obtain the local improvement index of the periodic fluctuation curve.

[0026] The recovery capacity index is obtained according to the recovery capacity index formula, which includes:

[0027] ;in, The recovery capability index of the curve during the abnormal period; The periodic anomaly measure is the periodic fluctuation curve of the first periodic fluctuation curve in the time series dimension within the abnormal period curve. The periodic anomaly measure is the periodic fluctuation curve of the last periodic fluctuation curve in the time series dimension within the abnormal period curve. This is the mean of the local improvement indicators of all periodic fluctuation curves in the abnormal period curve; This refers to the total number of all periodic fluctuation curves in the abnormal period curve where the local improvement index is greater than the preset improvement value. This represents the total number of all periodic fluctuation curves within the curve during the abnormal period.

[0028] Furthermore, the method for obtaining the abnormal electrocardiogram indicators includes:

[0029] In the abnormal period curve, the mean of the periodic abnormality measure of all the periodic fluctuation curves is calculated to obtain the electrocardiogram abnormality index of the abnormal period curve.

[0030] Furthermore, the method for obtaining the local risk indicators includes:

[0031] The negative correlation mapping result of the recovery ability index of all abnormal period curves is calculated and multiplied with the electrocardiogram abnormality index to obtain the period risk index of the abnormal period curve.

[0032] By negatively mapping the time interval between the last data point of the abnormal period curve and the current time, the time weight of the abnormal period curve is obtained.

[0033] In the waveform curve to be analyzed, the time weights of all abnormal time period curves are used to perform a weighted summation of the time period risk indicators to obtain the local risk indicators of the waveform curve to be analyzed.

[0034] Furthermore, methods for assessing the risk of interventional procedures in patients include:

[0035] The cumulative value of the local risk index is calculated based on the corresponding ECG waveform curves of all leads of the patient to obtain the interventional surgery risk measure of the patient.

[0036] This invention proposes an interventional surgery risk assessment system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the interventional surgery risk assessment method.

[0037] The present invention has the following beneficial effects:

[0038] One ECG waveform is randomly selected from multiple waveforms for analysis. Based on the similarity of local morphological features, the entire waveform is divided into several periodic fluctuation curves. Each periodic fluctuation curve represents the cardiac electrophysiological activity of a complete physiological cycle. Since abnormal symptoms can cause abnormal fluctuations in the ECG waveform, after obtaining the periodic fluctuation curves, the morphological differences between these curves can be compared to obtain a periodic anomaly metric, which quantifies the degree of abnormality in the physiological cycle. Abnormal time period curves are then extracted from the waveform to be analyzed, reflecting a complete period of abnormal cardiac physiological activity. Recovery capacity indicators are constructed to reflect the heart's self-regulation and recovery ability after experiencing abnormalities. These indicators are crucial for assessing surgical risk and provide a more comprehensive understanding of the heart's functional state. ECG abnormality indicators quantify the degree of cardiac abnormality, providing important evidence for surgical risk assessment and more accurately determining the presence of potential risk factors. Local risk indicators comprehensively reflect the risk of the recorded cardiac electrical activity across all leads by integrating the risk of each abnormal time period within the waveform to be analyzed. By comprehensively considering local risk indicators from multiple leads of ECG waveform curves, the risk of interventional surgery for patients can be assessed more accurately as a whole. Attached Figure Description

[0039] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 A flowchart illustrating an interventional surgery risk assessment method provided in one embodiment of the present invention;

[0041] Figure 2 A flowchart illustrating a method for obtaining a periodic fluctuation curve according to an embodiment of the present invention;

[0042] Figure 3 This is a structural diagram of an interventional surgery risk assessment system provided in one embodiment of the present invention. Detailed Implementation

[0043] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an interventional surgery risk assessment method and system proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0045] The specific details of the interventional surgery risk assessment method and system provided by the present invention will be described below with reference to the accompanying drawings.

[0046] This invention provides a method and system for risk assessment in interventional surgery. Please refer to [link / reference]. Figure 1 The diagram illustrates a flowchart of an interventional surgery risk assessment method according to an embodiment of the present invention, the method comprising the following steps:

[0047] Step S1: Obtain the patient's electrocardiogram (ECG) data set; the ECG data set includes the ECG waveform curves corresponding to each lead.

[0048] For patients with coronary artery disease, real-time dynamic monitoring of their electrocardiogram (ECG) data is crucial for assessing the risk of interventional procedures. After user authorization, the monitoring system retrieves the ECG waveforms for each lead. The specific acquisition process includes:

[0049] Based on the cardiac triplet method, a portable electrocardiogram (ECG) monitor is used to detect the cardiac electrophysiological activity of patients with coronary heart disease. From the start of monitoring to the current moment, the ECG monitor continuously records the ECG waveform curves of each lead, which visually demonstrate the dynamic changes of the ECG signal. It should be noted that the horizontal axis of the ECG waveform curve represents time, used to accurately mark the temporal characteristics of cardiac electrical activity; the vertical axis represents voltage, used to reflect the amplitude changes of the ECG signal. It should also be noted that the cardiac triplet method is an industry-recognized ECG monitoring technique, and will not be elaborated upon here.

[0050] It should be noted that the data collection in this invention is authorized by the user, does not violate relevant laws and regulations, and does not contravene public order and good morals. For ease of calculation, all indicator data involved in the calculations in this embodiment of the invention have undergone data preprocessing to eliminate the influence of dimensions. The specific methods for removing the influence of dimensions are well-known to those skilled in the art and are not limited here.

[0051] Step S2: Take any ECG waveform curve as the waveform curve to be analyzed; divide the waveform curve to be analyzed according to the similarity of local morphological features, and obtain the periodic fluctuation curves of the waveform curve to be analyzed; obtain the periodic abnormality measure of the periodic fluctuation curves according to the morphological differences between the periodic fluctuation curves in the waveform curve to be analyzed; divide the abnormal period curves from the waveform curve to be analyzed according to the periodic abnormality measure of the periodic fluctuation curves.

[0052] One ECG waveform is randomly selected from multiple waveforms as the target waveform. Based on the similarity of local morphological features, the entire waveform is divided into several periodic fluctuation curves. Each periodic fluctuation curve represents the cardiac electrophysiological activity of a complete physiological cycle. Since abnormal symptoms can cause abnormal fluctuations in the ECG waveform, after obtaining the periodic fluctuation curves of the target waveform, the morphological differences between these periodic fluctuation curves can be further compared to obtain a periodic anomaly measure. This periodic anomaly measure is used to quantify the degree of abnormality in the physiological cycle. Abnormal time period curves are then extracted from the target waveform curve, reflecting a complete period of abnormal cardiac physiological activity.

[0053] Considering the periodicity of normal cardiac physiological activity, the waveform curve is divided into multiple periodic fluctuation curves based on the similarity of local morphological features. Each periodic fluctuation curve initially reflects the physiological activity of a complete physiological cycle. Please refer to [link to relevant documentation]. Figure 2 The diagram illustrates a flowchart of a method for obtaining a periodic fluctuation curve according to an embodiment of the present invention. Preferably, in one embodiment of the present invention, the method for obtaining the periodic fluctuation curve includes:

[0054] Step S201: Based on the fluctuation of the waveform curve to be analyzed, obtain the characteristic fluctuation peak of the waveform curve to be analyzed.

[0055] In order to divide the physiological cycle in the electrocardiogram waveform curve, the characteristic fluctuation peaks of the waveform curve to be analyzed are obtained according to the fluctuation of the waveform curve. The characteristic fluctuation peaks reflect the significant fluctuations in the electrocardiogram waveform curve and represent the main characteristics of the waveform.

[0056] Preferably, in one embodiment of the present invention, the method for obtaining the characteristic fluctuation peak includes:

[0057] The AMDP (Amplitude Modulation Depth Peak Detection Algorithm) is used to obtain all peak points in the waveform curve to be analyzed. The data point corresponding to the minimum value between adjacent peak points is taken as the valley point. For any peak point in the waveform curve to be analyzed, the valley point adjacent to the peak point is taken as the previous valley point, and the valley point adjacent to the peak point is taken as the next valley point. The waveform curve to be analyzed corresponding to the previous valley point to the next valley point of each peak point is taken as the characteristic fluctuation peak corresponding to the peak point. The characteristic fluctuation peaks corresponding to all peak points in the waveform curve to be analyzed are counted to obtain all characteristic fluctuation peaks of the waveform curve to be analyzed. It should be noted that the AMDP peak detection algorithm is a prior art well known to those skilled in the art and will not be described in detail here. In this invention, the minimum value before the first peak point in the waveform curve to be analyzed is taken as the first valley point; and the minimum value after the last peak point in the waveform curve to be analyzed is taken as the last valley point. It should be noted that the order of the data points here refers to the order in the time dimension.

[0058] Following the steps above, a peak detection algorithm is used to identify all peak points in the ECG waveform curve. Peak points are the local maximum points in the ECG waveform curve, representing the main peaks of the waveform. The data points corresponding to the minimum values ​​between adjacent peak points are identified; these points are designated as valley points. Valley points are the local minimum points in the ECG waveform curve, located between adjacent peak points and used to define the boundaries of each characteristic fluctuation peak. For each peak point, the adjacent valley point preceding it is designated as the previous valley point, and the adjacent valley point following it is designated as the next valley point. Each peak point and its preceding and following valley points define a characteristic fluctuation peak. By traversing the entire ECG waveform curve and counting the characteristic fluctuation peaks corresponding to all peak points, all characteristic fluctuation peaks of the entire ECG waveform curve are obtained.

[0059] Step S202: Based on the similarity of morphological features among the various characteristic fluctuation peaks, obtain the similarity parameters of the characteristic fluctuation peaks.

[0060] Considering that the electrocardiogram waveforms corresponding to different physiological cycles are highly similar under normal circumstances, similarity parameters are constructed to reflect the similarity between characteristic fluctuation peaks, so as to provide a basis for subsequent division of electrocardiogram waveform curves corresponding to physiological cycles.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining similar parameters includes:

[0062] The interval between the maximum and minimum abscissa values ​​corresponding to the characteristic fluctuation peak is taken as the total width of the characteristic fluctuation peak; the interval between the maximum and minimum ordinate values ​​corresponding to the characteristic fluctuation peak is taken as the total height of the characteristic fluctuation peak.

[0063] In the waveform curve to be analyzed, one characteristic fluctuation peak is randomly selected as the target fluctuation peak, and all other characteristic fluctuation peaks are used as reference fluctuation peaks. The absolute value of the difference between the total width of the target fluctuation peak and the reference fluctuation peaks is calculated to obtain the local width difference value of the reference fluctuation peak. The mean of the local width differences of all reference fluctuation peaks is calculated to obtain the overall width difference value of the target fluctuation peak. The absolute value of the difference between the total height of the target fluctuation peak and the reference fluctuation peaks is calculated to obtain the local height difference value of the reference fluctuation peak. The mean of the local height differences of all reference fluctuation peaks is calculated to obtain the overall height difference value of the target fluctuation peak. The product of the overall width difference value and the overall height difference value is calculated and negatively correlated to obtain the similarity parameters of the target fluctuation peak. The negative correlation mapping of this invention can be inversely proportional or negative exponential form, which is not limited here.

[0064] For the steps described above, the total width and total height of the characteristic fluctuation peaks were first defined. Based on these definitions, the differences in width and height between the target fluctuation peak and each reference fluctuation peak were quantified. Specifically, the absolute value of the difference in total width between the target fluctuation peak and each reference fluctuation peak was calculated, i.e., the local width difference value. The local width difference values ​​of all reference fluctuation peaks were then averaged to obtain the overall width difference value of the target fluctuation peak. Similarly, the absolute value of the difference in total height between the target fluctuation peak and each reference fluctuation peak was also calculated, i.e., the local height difference value, to obtain the overall height difference value of the target fluctuation peak. The product of the overall width difference value and the overall height difference value was calculated. This product reflects the comprehensive difference between the target fluctuation peak and other fluctuation peaks in the waveform curve to be analyzed in both width and height dimensions. The larger the product, the greater the difference in width and height between the target fluctuation peak and other fluctuation peaks, and the lower the similarity. A negative correlation mapping was then applied to the above product. Negative correlation mapping means converting the product value into a parameter proportional to the similarity, obtaining the similarity parameter of the target fluctuation peak. The larger the similarity parameter, the higher the similarity between the target fluctuation peak and other fluctuation peaks.

[0065] Step S203: Based on the similarity parameters of the characteristic fluctuation peaks, divide the waveform curve to be analyzed into its respective periodic fluctuation curves to obtain the waveform curve to be analyzed.

[0066] By quantifying the similarity between characteristic fluctuation peaks, fluctuation peaks with similar shapes can be identified, thereby helping to divide the period of the waveform curve to be analyzed and obtain the fluctuation curves of each period of the waveform curve to be analyzed.

[0067] Preferably, in one embodiment of the present invention, the method for obtaining the periodic fluctuation curve includes:

[0068] Each fluctuation peak with a similarity parameter greater than a preset similarity threshold is used as a dividing fluctuation peak; the first data point of each dividing fluctuation peak is used as a dividing point, dividing the waveform curve to be analyzed into various periodic fluctuation curves. It should be noted that each dividing point is located between corresponding adjacent periodic fluctuation curves, and the periodic fluctuation curve to which each dividing point is divided is not limited here. In one embodiment of the present invention, the preset similarity threshold is 0.81, and the implementer can set it according to the implementation scenario.

[0069] Following the steps described above, characteristic fluctuation peaks with similarity parameters greater than a preset similarity threshold are used as dividing fluctuation peaks. These dividing fluctuation peaks have high similarity and can serve as reference points for dividing the period. The first data point of each dividing fluctuation peak is used as the dividing point; based on the dividing points, the waveform curve to be analyzed is divided into multiple continuous periodic fluctuation curves. Each periodic fluctuation curve contains waveform data from one dividing point to the next, reflecting a complete cardiac activity cycle.

[0070] To quantify the degree of abnormality in the menstrual cycle, preferably, in one embodiment of the present invention, the method for obtaining the measurement of cycle abnormality includes:

[0071] In the waveform curves to be analyzed, one periodic fluctuation curve is randomly selected as the target period, and all other periodic fluctuation curves are used as reference periods. The average duration of all reference periods of the target period is calculated to obtain the overall reference duration of the target period. The absolute value of the difference between the duration of the target period and the overall reference duration is calculated to obtain the first difference value of the target period. The average value of the similarity parameters of all characteristic fluctuation peaks in the target period is calculated and negative correlation mapping is performed to obtain the second difference value of the target period. The product of the first difference value and the second difference value is calculated and normalized to obtain the periodic anomaly measure of the target period. It should be noted that negative correlation mapping and normalization are techniques well known to those skilled in the art. Negative correlation mapping can be inversely proportional or negative exponential form, and normalization can be linear normalization, etc., without limitation.

[0072] Regarding the above steps, considering that cardiac abnormalities mainly manifest as differences in duration and waveform characteristics—for example, arrhythmias often cause the duration of the periodic fluctuation curve to differ from that of the periodic fluctuation curve corresponding to normal physiological activity, and myocardial infarction can cause significant changes in the peak shape of the periodic fluctuation curve—one periodic fluctuation curve is randomly selected as the target period. All remaining periodic fluctuation curves are considered as reference periods for comparison with the target period. The average duration of all reference periods is calculated to obtain the overall reference duration of the target period. The overall reference duration represents the normal baseline. The absolute value of the difference between the duration of the target period and the overall reference duration is calculated to obtain the first difference value of the target period. The first difference value reflects the degree of deviation of the target period from the average duration. To analyze the differences in waveform characteristics, a negative correlation mapping is needed on the mean of similar parameters to ensure that the greater the difference in waveform characteristics, the larger the second difference value. The product of the first and second difference values ​​is calculated and normalized to obtain the periodic anomaly measure of the target period. The period anomaly metric reflects the degree of comprehensive difference between the target period and other periods in terms of duration and waveform characteristics. The larger the period anomaly metric, the greater the difference between the target period and other periods in terms of duration and waveform characteristics, and therefore the greater the possibility of waveform anomalies.

[0073] To determine the complete period of abnormal cardiac physiological activity, preferably, in one embodiment of the present invention, the method for obtaining the abnormal period curve includes:

[0074] Periodic fluctuation curves with a periodic anomaly measurement greater than a preset periodic anomaly threshold are designated as abnormal periodic curves; consecutive abnormal periodic curves exceeding a preset statistical number are designated as abnormal time period curves. In one embodiment of the invention, the preset periodic anomaly threshold is 0.63, which is set based on clinical experience and actual needs to distinguish between normal and abnormal periods. In one embodiment of the invention, the preset statistical number is set to 2, meaning that the appearance of two consecutive abnormal periodic curves can be considered as a period of abnormal cardiac physiological activity. The implementer can set this value according to the implementation scenario.

[0075] Regarding the above steps, the periodic anomaly measurement reflects the degree of abnormality in the physiological cycle. In the waveform curves to be analyzed, all periodic fluctuation curves with a periodic anomaly measurement greater than a preset periodic anomaly threshold are selected; these curves are considered abnormal periodic curves. Based on a preset statistical number, time periods in which consecutive abnormal periodic curves appear are divided from the waveform curves to be analyzed; curves within these time periods are merged into abnormal time period curves. Abnormal time period curves reflect a complete period of abnormal cardiac physiological activity. Abnormal time period curves are curves that reflect periods of abnormal cardiac physiological activity, segmented from the waveform curves to be analyzed. Through in-depth analysis of abnormal time period curves, the dynamic changes in the condition can be captured more accurately, thereby leading to a more accurate assessment of surgical risks.

[0076] Step S3: In the abnormal time period curve, obtain the recovery ability index of the abnormal time period curve based on the change of the periodic abnormality measure of the periodic fluctuation curve over time; obtain the electrocardiogram abnormality index of the abnormal time period curve based on the overall distribution of the periodic abnormality measure of all periodic fluctuation curves included in the abnormal time period curve; in the waveform curve to be analyzed, obtain the local risk index of the waveform curve to be analyzed based on the recovery ability index and electrocardiogram abnormality index of all abnormal time period curves; assess the patient's interventional surgery risk based on the local risk index of the electrocardiogram waveform curve corresponding to all leads of the patient.

[0077] Recovery capacity indicators are constructed to reflect the heart's self-regulation and recovery capabilities after experiencing abnormalities. These indicators are crucial for assessing surgical risk and provide a more comprehensive understanding of the heart's functional state. Electrocardiogram (ECG) abnormality indicators quantify the degree of cardiac abnormalities, providing important evidence for surgical risk assessment and allowing for a more accurate identification of potential cardiac risk factors. Local risk indicators reflect the risk of cardiac electrical activity recorded in each lead by comprehensively analyzing the risk of abnormal periods in the waveform curve. By considering local risk indicators from multiple leads of ECG waveform curves, a more accurate overall assessment of the patient's interventional surgical risk can be achieved.

[0078] To assess the heart's recovery capacity during periods of abnormal cardiac physiological activity, preferably, in one embodiment of the present invention, the method for obtaining recovery capacity indicators includes:

[0079] In the abnormal period curve, the difference between each periodic fluctuation curve and the periodic anomaly measure of the previous periodic fluctuation curve in the time series dimension is calculated and negatively correlated to obtain the local improvement index of the periodic fluctuation curve.

[0080] Following the steps outlined above, in the abnormal period curves, the difference between the periodic anomaly measurement of each periodic fluctuation curve and the periodic anomaly measurement of the previous periodic fluctuation curve in the time series dimension is calculated. A negative correlation mapping is then applied to the calculated differences to obtain a local improvement index. The core concept of this negative correlation mapping is that when the difference is negative, it indicates that the anomaly measurement in the current period has decreased compared to the previous period, i.e., showing an improving trend, and the local improvement index will be assigned a positive value. Conversely, if the difference is positive, it indicates that the anomaly measurement in the current period has increased compared to the previous period, i.e., the situation has worsened, and the local improvement index will show a negative value.

[0081] The recovery capacity index is obtained based on the recovery capacity index formula, which includes:

[0082] ;in, This is an indicator of the recovery capability of the curve during abnormal periods; This is a measure of the periodic anomaly of the first periodic fluctuation curve in the time series dimension within the curve during an abnormal period. This is a measure of the periodic anomaly of the last periodic fluctuation curve in the time series dimension within the abnormal period curve. This represents the average of the local improvement indicators for all cyclical fluctuation curves within the abnormal period curve. This refers to the total number of all periodic fluctuation curves in the abnormal period curve where the local improvement index is greater than the preset improvement value; This represents the total number of periodic fluctuation curves within the abnormal period curve. In one embodiment of the present invention, the preset improvement value is 0, which can be set by the implementer according to the implementation scenario.

[0083] Regarding the above formula, The larger the value, the more significant the heart's recovery during the abnormal period, meaning the greater the degree to which the heart's activity transitions from an abnormal state to a normal state, indicating a clear trend of recovery or improvement in heart activity. It reflects the average recovery trend of cardiac activity in each cycle throughout the entire abnormal period. A higher level indicates a clear trend of recovery in cardiac activity. A higher value indicates that a larger proportion of cycles are in a state of recovery or improvement during the abnormal period, suggesting a significant trend of recovery or improvement in cardiac activity. The final recovery capacity index is obtained, which comprehensively assesses the heart's ability to recover during periods of abnormal physiological activity; a higher recovery capacity index indicates a lower risk of surgery.

[0084] To assess the overall degree of cardiac abnormality during periods of abnormal cardiac physiological activity, preferably, in one embodiment of the present invention, the method for obtaining electrocardiogram abnormality indicators includes:

[0085] In the abnormal period curve, the mean of the periodic abnormality measure of all periodic fluctuation curves is calculated to obtain the electrocardiogram abnormality index of the abnormal period curve.

[0086] Following the steps above, the mean value of the periodic abnormality measure of all periodic fluctuation curves in the abnormal period curve is calculated to obtain the electrocardiogram (ECG) abnormality index of the abnormal period curve. A higher ECG abnormality index indicates a more severe condition during periods of abnormal cardiac physiological activity, and a greater risk of surgery.

[0087] Preferably, in one embodiment of the present invention, the method for obtaining local risk indicators includes:

[0088] The negative correlation mapping result of the recovery ability index of all abnormal period curves is calculated and multiplied with the electrocardiogram abnormality index to obtain the period risk index of the abnormal period curve.

[0089] By negatively mapping the time interval between the last data point of the abnormal period curve and the current time, the time weight of the abnormal period curve is obtained.

[0090] In the waveform curve to be analyzed, the time weights of all abnormal period curves are used to perform a weighted summation of the period risk indicators to obtain the local risk indicators of the waveform curve to be analyzed. The negative correlation mapping in this invention can be inversely proportional or negative exponential form; this is not limited here.

[0091] Following the steps above, the negative correlation mapping result of the recovery ability index for all abnormal time period curves is calculated, along with the product of the ECG abnormality index. The negative correlation mapping result reflects the inverse relationship between recovery ability and risk; that is, the stronger the recovery ability, the lower the risk. The time interval between the last data point of the abnormal time period curve and the current time is negatively correlated to obtain the time weight. This reflects the impact of the distance between the abnormal time period and the current time on the current surgical risk. The time weights of all abnormal time period curves are used to weighted sum the time period risk indicators to obtain the local risk index of the waveform curve to be analyzed. The local risk index reflects the risk of the ECG activity recorded in the leads by comprehensively considering the risk situation of each abnormal time period in the waveform curve to be analyzed.

[0092] To assess the risks of interventional procedures from a holistic perspective, preferably, in one embodiment of the present invention, the method for assessing the risks of a patient's interventional procedures includes:

[0093] The cumulative value of local risk indicators for the patient's electrocardiogram waveforms across all leads is calculated to obtain a measure of the patient's interventional surgical risk.

[0094] Regarding the above steps, considering that different leads can capture electrical activity information from different parts of the heart, the local risk index of the ECG waveform curves of all leads is combined. That is, the cumulative value of the local risk index of the ECG waveform curves of all leads of the patient is calculated. This cumulative value reflects the patient's current cardiac risk level for surgery. The higher the interventional surgery risk measure, the greater the risk of performing the current interventional surgery.

[0095] This invention also proposes an interventional surgery risk assessment system; please refer to [link / reference]. Figure 3 The diagram illustrates a structural diagram of an interventional surgery risk assessment system provided by an embodiment of the present invention. The system includes: a data acquisition module 101, an abnormal time period curve division module 102, and a risk assessment module 103.

[0096] The data acquisition module 101 is used to acquire the patient's electrocardiogram (ECG) data set; the ECG data set includes the ECG waveform curves corresponding to each lead.

[0097] The abnormal period curve segmentation module 102 is used to take any ECG waveform curve as the waveform curve to be analyzed; divide the waveform curve to be analyzed according to the similarity of local morphological features, and obtain the periodic fluctuation curves of the waveform curve to be analyzed; obtain the periodic abnormality measure of the periodic fluctuation curves according to the morphological feature differences between the periodic fluctuation curves in the waveform curve to be analyzed; and divide the abnormal period curves from the waveform curve to be analyzed according to the periodic abnormality measure of the periodic fluctuation curves.

[0098] The risk assessment module 103 is used to obtain the recovery capability index of the abnormal period curve based on the change of the periodic abnormality measure of the periodic fluctuation curve over time; to obtain the electrocardiogram abnormality index of the abnormal period curve based on the overall distribution of the periodic abnormality measure of all periodic fluctuation curves included in the abnormal period curve; to obtain the local risk index of the waveform curve to be analyzed based on the recovery capability index and electrocardiogram abnormality index of all abnormal period curves; and to assess the patient's interventional surgery risk based on the local risk index of the electrocardiogram waveform curve corresponding to all leads of the patient.

[0099] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the interventional surgery risk assessment system and the interventional surgery risk assessment method embodiment provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiment, which will not be repeated here.

[0100] In summary, this invention provides a method and system for assessing interventional surgical risks. First, in an electrocardiogram (ECG), characteristic fluctuation peaks of the ECG curve are obtained based on its fluctuation pattern. Then, similarity parameters of the fluctuation peaks are obtained based on their morphological distribution characteristics. Based on these similarity parameters, the ECG curve is divided into periodic fluctuation curves. A periodic anomaly measure is obtained based on the similarity parameters of the fluctuation peaks and the duration of the periodic fluctuation curves. Abnormal time periods are identified from the ECG curve based on these periodic anomaly measures. Within these abnormal time periods, a recovery capacity index is obtained based on the temporal variation of the periodic anomaly measure. Abnormal ECG time periods are further defined based on the distribution of abnormal time periods across all ECG curves. Finally, a severity measure of the abnormal ECG time period is obtained based on the recovery capacity index and duration of the abnormal ECG time period. This method and system assess the patient's interventional surgical risk. This invention improves the accuracy of risk assessment by reasonably measuring the severity of the patient's condition.

[0101] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0102] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for risk assessment in interventional surgery, characterized in that, The method includes: Acquire the patient's electrocardiogram (ECG) data set; the ECG data set includes the ECG waveform curves corresponding to each lead; Take any ECG waveform curve as the waveform curve to be analyzed; divide the waveform curve to be analyzed based on the similarity of local morphological features, and obtain the periodic fluctuation curves of the waveform curve to be analyzed; obtain the periodic abnormality measure of the periodic fluctuation curve based on the morphological differences between the periodic fluctuation curves in the waveform curve to be analyzed; and divide the abnormal period curves from the waveform curve to be analyzed based on the periodic abnormality measure of the periodic fluctuation curve. In the abnormal period curve, the recovery ability index of the abnormal period curve is obtained based on the change of the periodic abnormality measure of the periodic fluctuation curve over time; the electrocardiogram abnormality index of the abnormal period curve is obtained based on the overall distribution of the periodic abnormality measure of all periodic fluctuation curves included in the abnormal period curve; in the waveform curve to be analyzed, the local risk index of the waveform curve to be analyzed is obtained based on the recovery ability index and electrocardiogram abnormality index of all abnormal period curves; the interventional surgery risk of the patient is assessed based on the local risk index of the electrocardiogram waveform curve corresponding to all leads of the patient. The methods for obtaining recovery capability indicators include: calculating the difference between the periodic anomaly measure of each periodic fluctuation curve and the periodic anomaly measure of the previous periodic fluctuation curve in the time series dimension, and performing a negative correlation mapping to obtain a local improvement indicator of the periodic fluctuation curve; obtaining the recovery capability indicator according to the recovery capability indicator formula, which includes: ;in, This is an indicator of the recovery capability of the curve during abnormal periods; This is a measure of the periodic anomaly of the first periodic fluctuation curve in the time series dimension within the curve during an abnormal period. This is a measure of the periodic anomaly of the last periodic fluctuation curve in the time series dimension within the abnormal period curve. This represents the average of the local improvement indicators for all cyclical fluctuation curves within the abnormal period curve. This refers to the total number of all periodic fluctuation curves in the abnormal period curve where the local improvement index is greater than the preset improvement value; This represents the total number of all periodic fluctuation curves within the abnormal period curve; The method for obtaining local risk indicators includes: calculating the product of the negative correlation mapping result of the recovery ability indicator of all abnormal time period curves and the ECG abnormality indicator to obtain the time period risk indicator of the abnormal time period curve; performing a negative correlation mapping between the time interval corresponding to the last data point of the abnormal time period curve and the current time to obtain the time weight of the abnormal time period curve; and using the time weight of all abnormal time period curves to perform a weighted summation of the time period risk indicator in the waveform curve to be analyzed to obtain the local risk indicator of the waveform curve to be analyzed.

2. The interventional surgery risk assessment method according to claim 1, characterized in that, Methods for obtaining periodic fluctuation curves include: Based on the fluctuation of the waveform curve to be analyzed, obtain the characteristic fluctuation peaks of the waveform curve to be analyzed; Based on the similarity of morphological characteristics among the various characteristic fluctuation peaks, the similarity parameters of the characteristic fluctuation peaks are obtained. Based on the similarity parameters of the characteristic fluctuation peaks, the waveform curve to be analyzed is divided to obtain the periodic fluctuation curves of the waveform curve to be analyzed.

3. The interventional surgery risk assessment method according to claim 2, characterized in that, Methods for obtaining similarity parameters include: The interval between the maximum and minimum abscissa values ​​corresponding to the characteristic fluctuation peak is taken as the total width of the characteristic fluctuation peak; the interval between the maximum and minimum ordinate values ​​corresponding to the characteristic fluctuation peak is taken as the total height of the characteristic fluctuation peak. In the waveform curve to be analyzed, select any one characteristic fluctuation peak as the target fluctuation peak, and use all other characteristic fluctuation peaks as reference fluctuation peaks. Calculate the absolute value of the difference between the total width of the target fluctuation peak and the reference fluctuation peaks to obtain the local width difference value of the reference fluctuation peak. Calculate the mean of the local width differences of all reference fluctuation peaks to obtain the overall width difference value of the target fluctuation peak. Calculate the absolute value of the difference between the total height of the target fluctuation peak and the reference fluctuation peaks to obtain the local height difference value of the reference fluctuation peak. Calculate the mean of the local height differences of all reference fluctuation peaks to obtain the overall height difference value of the target fluctuation peak. Calculate the product of the overall width difference value and the overall height difference value and perform a negative correlation mapping to obtain the similarity parameters of the target fluctuation peak.

4. The interventional surgery risk assessment method according to claim 2, characterized in that, Methods for obtaining periodic anomaly metrics include: Select one periodic fluctuation curve from the waveform curves to be analyzed as the target period, and use all other periodic fluctuation curves as reference periods. Calculate the average duration of all reference periods for the target period to obtain the overall reference duration of the target period; calculate the absolute value of the difference between the duration of the target period and the overall reference duration to obtain the first difference value of the target period. The mean of similarity parameters of all characteristic fluctuation peaks in the target period is calculated and negative correlation mapping is performed to obtain the second difference value of the target period; Calculate the product of the first and second difference values ​​and normalize it to obtain the periodic anomaly measure of the target period.

5. The interventional surgery risk assessment method according to claim 1, characterized in that, Methods for obtaining curves during abnormal periods include: Periodic fluctuation curves with periodic anomaly measurements exceeding a preset periodic anomaly threshold are designated as abnormal periodic curves; abnormal periodic curves exceeding a preset statistical number of consecutive occurrences are designated as abnormal time period curves.

6. The interventional surgery risk assessment method according to claim 1, characterized in that, Methods for obtaining abnormal electrocardiogram (ECG) indicators include: In the abnormal period curve, the mean of the periodic abnormality measure of all periodic fluctuation curves is calculated to obtain the electrocardiogram abnormality index of the abnormal period curve.

7. The interventional surgery risk assessment method according to claim 1, characterized in that, Methods for assessing the risk of interventional procedures in patients include: The cumulative value of local risk indicators for the patient's electrocardiogram waveforms across all leads is calculated to obtain a measure of the patient's interventional surgical risk.

8. An interventional surgery risk assessment system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the interventional surgery risk assessment method as described in any one of claims 1 to 7.

Images