A multi-parameter fusion and intelligent perception electrocardio health evaluation method and system
By using a multi-parameter fusion and intelligent sensing method for ECG health assessment, a comprehensive assessment path is dynamically generated and double-layered verification is performed. This solves the adaptability and reliability problems of traditional ECG assessment methods, and achieves efficient and accurate assessment of ECG health status and clear reflection of risk level.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HANGZHOU LUOYU TECH CO LTD
- Filing Date
- 2026-07-07
- Publication Date
- 2026-08-04
AI Technical Summary
Traditional ECG health assessment methods cannot cover various types of ECG imbalances, such as baseline stability, variability response, and trend extrapolation. The assessment process is fixed and difficult to adapt to the ECG functional imbalance characteristics of different assessment subjects, resulting in one-sided assessment results and a lack of bias identification and correction, which reduces the reliability of the assessment results.
A multi-parameter fusion and intelligent sensing electrocardiogram health assessment method is adopted. By dynamically generating a comprehensive electrocardiogram assessment path and combining a two-layer verification mechanism of initial assessment coefficients and feedback assessment coefficients, deviations in the assessment process are identified and corrected, thereby achieving closed-loop process optimization.
It significantly improves the accuracy and reliability of electrocardiogram (ECG) assessment results, ensures that the assessment results match the human physiological state, provides a clear assessment of health risk levels, and provides a reliable basis for subsequent interventions.
Smart Images
Figure CN122511601A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electrocardiogram (ECG) health assessment technology, and more specifically, to an ECG health assessment method and system that integrates multi-parameter fusion and intelligent sensing. Background Technology
[0002] Electrocardiogram (ECG) health assessment methods are core tools for evaluating cardiac electrical activity and assessing cardiovascular health in clinical and routine health monitoring. They rely on the conduction patterns of the heart's own electrophysiological activity to conduct detection and analysis, directly reflecting the rhythm of myocardial contraction, conduction function, and the basic health status of myocardial cells. This type of assessment is based on ECG acquisition, using electrodes strategically attached to specific locations on the body surface to capture the weak electrical signals generated by the heart during each cardiac cycle, transforming electrophysiological changes into visualized waveforms.
[0003] Traditional electrocardiogram (ECG) health assessment methods cannot cover the associated effects of various ECG imbalance types, such as baseline stability, variability response, and trend extrapolation. The assessment process is mostly based on a fixed sequence of preset nodes, which is difficult to adapt to the ECG functional imbalance characteristics of different assessment subjects. It is easy to lead to one-sided conclusions due to bias in assessment focus. At the same time, there is a lack of mechanisms to identify and correct biases in the assessment process, thereby reducing the reliability of the assessment results and even causing misjudgments or omissions. It is difficult to ensure that the results match the actual physiological state of the assessment subjects and cannot provide a reliable basis for health intervention. Summary of the Invention
[0004] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for electrocardiogram health assessment that integrates multi-parameter fusion and intelligent sensing.
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] A multi-parameter fusion and intelligent sensing method for electrocardiogram health assessment includes the following steps:
[0007] Based on the ECG functional imbalance data of the target assessment object, a comprehensive ECG assessment path is dynamically generated that integrates at least two different types of assessment dimensions; wherein, the comprehensive ECG assessment path consists of multiple assessment sub-paths that are cross-sorted according to a preset arrangement rule;
[0008] Collect the actual electrocardiogram (ECG) assessment path of the target assessment subject when performing the assessment task along the comprehensive ECG assessment path;
[0009] The abnormal assessment path segments are obtained by sequence comparison between the actual ECG assessment path and the comprehensive ECG assessment path, and the initial assessment coefficients are obtained according to the type and degree of difference of the abnormal assessment path segments.
[0010] If the initial assessment coefficient indicates a deviation in the ECG health assessment, a feedback assessment path is dynamically generated based on the type of abnormal assessment path segment, and the actual feedback assessment path of the target assessment object when performing the assessment task along the feedback assessment path is collected.
[0011] The feedback evaluation coefficient is obtained by comparing the actual feedback evaluation path with the feedback evaluation path.
[0012] The comprehensive evaluation coefficient is obtained based on the initial evaluation coefficient and the feedback evaluation coefficient. The electrocardiogram health risk level of the target assessment object is then output based on the comprehensive evaluation coefficient.
[0013] Preferably, a comprehensive electrocardiogram (ECG) assessment path that integrates at least two different types of assessment dimensions is dynamically generated based on the ECG functional imbalance data of the target assessment object, specifically including the following steps:
[0014] Obtain electrocardiogram (ECG) functional imbalance data of the target assessment subjects, and determine at least two types of ECG functional imbalance and the degree of imbalance corresponding to each type of ECG functional imbalance based on the ECG functional imbalance data;
[0015] Different assessment parameters for different types of electrocardiogram (ECG) imbalances are set according to the degree of imbalance of different ECG imbalance types.
[0016] For each type of electrocardiogram imbalance, a corresponding preliminary assessment path is generated based on the assessment parameters.
[0017] The assessment sub-paths of different preliminary assessment paths are cross-arranged and integrated according to the preset path arrangement rules to obtain the comprehensive ECG assessment path.
[0018] Preferably, the type of electrocardiographic imbalance includes at least two of the following: baseline stable imbalance, variability response imbalance, and trend extrapolation imbalance.
[0019] Preferably, the comprehensive ECG assessment pathway is obtained by cross-arranging and integrating the assessment sub-paths of different preliminary assessment pathways according to a preset pathway arrangement rule, specifically including the following steps:
[0020] Based on the preset pathway arrangement rules and the various types and degrees of ECG imbalance, the functional attributes corresponding to each preliminary assessment pathway are divided.
[0021] Based on functional attributes, each preliminary evaluation path is broken down into its sub-evaluation paths;
[0022] The assessment sub-paths are arranged alternately and crosswise according to the associated effects of different types of electrocardiogram imbalance.
[0023] The comprehensive ECG assessment pathway is obtained by systematically integrating all the cross-arranged assessment sub-paths.
[0024] Preferably, the comprehensive ECG assessment pathway is obtained by cross-arranging and integrating the assessment sub-paths of different preliminary assessment pathways according to a preset pathway arrangement rule, specifically including the following steps:
[0025] Based on the preset pathway arrangement rules and the various types and degrees of ECG imbalance, the functional attributes corresponding to each preliminary assessment pathway are divided.
[0026] Based on functional attributes, each preliminary evaluation path is broken down into its sub-evaluation paths;
[0027] The assessment sub-paths are arranged alternately and crosswise according to the associated effects of different types of electrocardiogram imbalance.
[0028] The comprehensive ECG assessment pathway is obtained by systematically integrating all the cross-arranged assessment sub-paths.
[0029] Preferably, the method further includes the following steps:
[0030] Obtain the evaluation results corresponding to the initial evaluation coefficients;
[0031] The correlation between anomaly assessment path segments and assessment dimensions is obtained based on the distribution of anomaly assessment path segments;
[0032] Determine the weight of the impact of abnormal assessment path segments on the assessment results based on the correlation relationships;
[0033] Determine whether the initial evaluation coefficients exceed the preset reasonable range based on the influence weights;
[0034] If the initial assessment coefficient exceeds the preset reasonable range, it is determined that there is a deviation in the electrocardiogram health assessment.
[0035] Preferably, the feedback evaluation path is dynamically generated based on the type of the anomaly evaluation path segment, specifically including the following steps:
[0036] The abnormal evaluation path segments are classified into categories to obtain the classification results; wherein, the classification results include path segments with missing path execution, path segments with disordered path order, and path segments with parameter perception deviation.
[0037] The direction of evaluation correction is determined based on the inherent characteristics of the division results;
[0038] The perceived assessment content of the comprehensive ECG assessment path is determined based on the weights of the ECG assessment dimensions corresponding to the segmentation results.
[0039] The feedback assessment path is obtained by reorganizing the path elements based on the assessment correction direction and the perceived assessment content.
[0040] Preferably, the feedback evaluation coefficient is obtained by comparing the actual feedback evaluation path with the feedback evaluation path, specifically including the following steps:
[0041] Compare the actual feedback evaluation path with the execution direction and content alignment of the feedback evaluation path, and screen for deviations in the path execution process based on the alignment of execution direction and content.
[0042] The degree of impact of deviation from the content is determined based on the differences in the abnormal assessment path segments;
[0043] The feedback evaluation coefficient is obtained based on the degree of impact.
[0044] Preferably, a comprehensive evaluation coefficient is obtained based on the initial evaluation coefficient and the feedback evaluation coefficient, and the electrocardiogram health risk level of the target assessment object is output based on the comprehensive evaluation coefficient. This specifically includes the following steps:
[0045] The degree of path abnormality difference is determined based on the ECG function fluctuation characteristics reflected by the initial assessment coefficient and the feedback assessment coefficient.
[0046] A comprehensive evaluation coefficient is obtained based on the degree of abnormality in the path.
[0047] The actual manifestation of electrocardiogram dysfunction is determined based on the comprehensive evaluation coefficient and the abnormal distribution characteristics of the pathway.
[0048] Based on the normal fluctuation patterns of electrocardiography, an appropriate grading standard is established, and the electrocardiographic health risk level of the target assessment object is obtained based on the grading standard and the comprehensive evaluation coefficient.
[0049] A multi-parameter fusion and intelligent sensing electrocardiogram health assessment system includes:
[0050] Generation module: Dynamically generates a comprehensive ECG assessment path that integrates at least two different types of assessment dimensions based on the ECG functional imbalance data of the target assessment object; wherein, the comprehensive ECG assessment path consists of multiple assessment sub-paths that are cross-sorted according to a preset arrangement rule;
[0051] Acquisition module: Acquires the actual electrocardiogram (ECG) assessment path of the target assessment object as it performs the assessment task along the comprehensive ECG assessment path;
[0052] Processing module: The actual ECG assessment pathway is compared with the comprehensive ECG assessment pathway to obtain abnormal assessment pathway segments, and the initial assessment coefficients are obtained based on the type and degree of difference of the abnormal assessment pathway segments.
[0053] The generation and acquisition module: If the initial evaluation coefficient indicates that there is a deviation in the ECG health assessment, the feedback evaluation path is dynamically generated based on the type of abnormal evaluation path segment, and the actual feedback evaluation path of the target evaluation object when performing the evaluation task along the feedback evaluation path is collected.
[0054] Comparison module: Obtains feedback evaluation coefficients by comparing the actual feedback evaluation path with the actual feedback evaluation path;
[0055] Output module: Obtains a comprehensive evaluation coefficient based on the initial evaluation coefficient and the feedback evaluation coefficient, and outputs the electrocardiogram health risk level of the target assessment object based on the comprehensive evaluation coefficient.
[0056] Compared with the prior art, the present invention has the following beneficial effects:
[0057] This invention effectively corrects execution deviations in the assessment process through a two-layer verification mechanism of initial assessment coefficients and feedback assessment coefficients. By identifying abnormal segments between the actual path and the preset path through sequence alignment, initial assessment coefficients are quantified and generated. A feedback assessment path is dynamically generated to perform secondary verification for deviations, and the feedback assessment coefficient is calculated, thus forming a closed-loop process of deviation identification, correction, and reassessment, significantly improving the accuracy and reliability of the assessment results. By comprehensively considering assessment coefficients and risk levels, electrocardiogram (ECG) health status is graded, and risk levels are classified according to the normal fluctuation patterns of ECG physiological data. This ensures the adaptability of assessment results to human physiological characteristics and clearly and intuitively reflects the level of health risk, providing a clear basis for subsequent interventions. This scheme upgrades ECG assessment from single-dimensional detection to multi-dimensional dynamic closed-loop verification, combining comprehensiveness, accuracy, and practicality, providing efficient and reliable technical support for ECG health monitoring. Attached Figure Description
[0058] Figure 1 This is a flowchart illustrating a multi-parameter fusion and intelligent sensing electrocardiogram health assessment method provided in an embodiment of the present invention.
[0059] Figure 2 This is a schematic diagram of a multi-parameter fusion and intelligent sensing electrocardiogram health assessment system provided in an embodiment of the present invention. Detailed Implementation
[0060] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0061] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0062] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.
[0063] Reference Figures 1-2 As shown.
[0064] The embodiments further illustrate the electrocardiogram health assessment method and system with multi-parameter fusion and intelligent sensing proposed in this invention.
[0065] A multi-parameter fusion and intelligent sensing method for electrocardiogram health assessment includes the following steps:
[0066] Based on the ECG functional imbalance data of the target assessment object, a comprehensive ECG assessment path is dynamically generated that integrates at least two different types of assessment dimensions; wherein, the comprehensive ECG assessment path consists of multiple assessment sub-paths that are cross-sorted according to a preset arrangement rule;
[0067] Specifically, the following steps are included:
[0068] Obtain electrocardiogram (ECG) functional imbalance data of the target assessment subjects, and determine at least two types of ECG functional imbalance and the degree of imbalance corresponding to each type of ECG functional imbalance based on the ECG functional imbalance data;
[0069] Different assessment parameters for different types of electrocardiogram (ECG) imbalances are set according to the degree of imbalance of different ECG imbalance types.
[0070] For each type of electrocardiogram imbalance, a corresponding preliminary assessment path is generated based on the assessment parameters.
[0071] ECG dysfunction data of the target assessment subjects is collected. ECG dysfunction data is a set of signals in the subject's ECG signal that deviate from normal physiological characteristics, including baseline ECG variability and abnormal rhythm trends. Based on the ECG dysfunction data, at least two different types of ECG dysfunction are identified, and the degree of dysfunction corresponding to each type is quantified. The degree of dysfunction is calculated using a numerical formula. The formula for calculating baseline stable dysfunction is: Degree of dysfunction D = |Actual baseline value - Standard baseline value| / Standard baseline value × 100%.
[0072] Different assessment parameters are set for different ECG imbalance types based on their degree of imbalance. The logic for setting the assessment parameters is directly related to the degree of imbalance, matching a corresponding influence weight coefficient for each ECG imbalance type, and calculating the assessment parameters based on the degree of imbalance. The calculation formula for the assessment parameter P1 of baseline stable imbalance is P1=D1×k1, and the calculation formula for the assessment parameter P2 of variability response imbalance is P2=D2×k2, where k1 and k2 are fixed coefficients pre-set according to the priority of the impact of different ECG imbalance types on ECG health. The higher the degree of imbalance, the larger the value of its assessment parameter.
[0073] For each type of ECG imbalance, a corresponding preliminary assessment path is generated based on the assessment parameters. Each preliminary assessment path is a complete assessment process set up for a single ECG imbalance type, with each node and execution order directly matching the assessment parameters for the corresponding imbalance type. For example, the preliminary assessment path for baseline stable imbalance sequentially sets up a baseline signal acquisition node, a baseline stability analysis node, and a baseline deviation calculation node according to the assessment parameter requirements. The execution standards and data acquisition requirements for each node are determined by the assessment parameters, ensuring that all data related to baseline stable imbalance can be completely collected and judged. The preliminary assessment path for variability response imbalance sequentially sets up an ECG variability signal acquisition node, a variability response rate analysis node, and a variability degree verification node according to the corresponding assessment parameters, ensuring that the relevant characteristics of variability response imbalance can be captured and judged. Each preliminary assessment path serves a single imbalance type, providing a foundation for subsequent multi-path fusion assessment.
[0074] According to the preset path arrangement rules, the assessment sub-paths of different preliminary assessment paths are cross-arranged and integrated to obtain a comprehensive ECG assessment path, which specifically includes the following steps:
[0075] Based on the preset pathway arrangement rules and the various types and degrees of ECG imbalance, the functional attributes corresponding to each preliminary assessment pathway are divided.
[0076] The preset path arrangement rules will pre-define the impact priority and weight calculation method for different types of electrocardiographic imbalance. For example, for baseline stable imbalance (D1) and variability response imbalance (D2), the assessment priority of the two is determined by the weight calculation formulas W1=D1×a and W2=D2×b, where a and b are fixed coefficients set according to clinical electrocardiographic effects. The higher the degree of imbalance, the greater the weight value. Based on the weight results, each preliminary assessment path is divided into core assessment attributes, auxiliary assessment attributes, or correlation verification attributes. For example, a preliminary path for baseline stable imbalance with a weight W1 greater than a preset threshold is classified as a core assessment attribute, and a preliminary path for variability response imbalance with a weight W2 less than a preset threshold is classified as an auxiliary assessment attribute, thereby clarifying the functional positioning of each path in subsequent integration.
[0077] Based on functional attributes, each preliminary evaluation path is broken down into its sub-evaluation paths;
[0078] The granularity of the decomposition is directly related to the functional attributes. The initial path for core evaluation attributes is broken down into more refined sub-paths. For example, the initial path for baseline stability imbalance is broken down into a baseline data acquisition sub-path, a baseline deviation calculation sub-path, and a baseline stability verification sub-path. Each sub-path corresponds to a single evaluation action or data processing step. The initial paths for auxiliary evaluation attributes appropriately merge some steps. For instance, the initial path for variation response imbalance is broken down into a variation signal acquisition sub-path and a variation response feature analysis sub-path, improving process efficiency while ensuring evaluation effectiveness. During the decomposition process, each sub-path is labeled with a corresponding functional attribute tag.
[0079] The assessment sub-paths are arranged alternately and crosswise according to the associated effects of different types of electrocardiogram imbalance.
[0080] There are physiological correlations between different types of electrocardiographic imbalances. For example, fluctuations in baseline stable imbalances directly affect the signal characteristics of variability response imbalances. The correlation between the two types of imbalances can be quantified by the correlation degree calculation formula R=|W1-W2| / (W1+W2). When the correlation degree R is greater than a preset threshold, it is determined that the two types of imbalances are strongly correlated, and their evaluation sub-paths need to be deployed alternately.
[0081] For example, the process first executes the baseline data acquisition sub-path, then the variation signal acquisition sub-path, followed by the baseline deviation calculation sub-path, and finally the variation response characteristic analysis sub-path. This alternating approach captures the interaction between the two types of imbalances in real time during the assessment process, avoiding information lag caused by continuous assessment of a single type and improving the dynamism and accuracy of the assessment process. If trend-based imbalances are involved, the corresponding sub-paths are embedded into cross-sequences based on their correlation with other imbalance types, ensuring that the associated effects of multiple imbalances are covered.
[0082] The comprehensive ECG assessment pathway is obtained by systematically integrating all the cross-arranged assessment sub-paths.
[0083] During the integration process, the execution order of some sub-paths is adjusted according to the priority of their functional attributes, following a cross-deployment sequence. This ensures that sub-paths for core assessment attributes are executed first, while sub-paths for auxiliary assessment attributes are interspersed. For example, the core sub-path for baseline stability imbalance is prioritized, while auxiliary sub-paths for variability response imbalance and verification sub-paths for trend extrapolation imbalance are interspersed between the two core sub-paths, ultimately forming a complete integrated ECG assessment pathway. This pathway includes assessment steps for at least two types of ECG imbalance, and achieves deep integration of assessment processes for different imbalance types through cross-deployment.
[0084] ECG imbalance types include at least two of the following: baseline stability imbalance, variability response imbalance, and trend extrapolation imbalance.
[0085] When the assessment object has two or three types of imbalance, the functional attribute division sub-path decomposition, cross-deployment and integration are completed according to the above process to ensure that the comprehensive ECG assessment path can adapt to the assessment needs of different imbalance combinations.
[0086] Collect the actual electrocardiogram (ECG) assessment path of the target assessment subject when performing the assessment task along the comprehensive ECG assessment path;
[0087] The comprehensive ECG assessment pathway is formed by the interleaving of assessment sub-pathways for various types of ECG imbalances. Each sub-path node includes preset execution rules and data acquisition requirements, such as the node's execution duration, data sampling rate, and stimulation signal parameters. After the target assessment subject initiates the assessment task, the pathway execution process is triggered, and the assessment subject completes the assessment actions corresponding to each sub-path sequentially according to the preset node order.
[0088] The entire execution process of the evaluation object's path is tracked in real time, collecting path execution status data and raw electrocardiographic data. Path execution status data includes the execution sequence, start time, end time, execution duration, completion status, and data acquisition completeness information for each sub-path node. The actual execution duration of a node is calculated using the formula ΔT = T2 - T1, where T1 is the node's start time and T2 is its end time. For example, if the preset execution duration for the baseline acquisition sub-path is 10 seconds, and the actual start time T1 is 0 seconds and the end time T2 is 8 seconds, then the actual execution duration ΔT = 8 - 0 = 8 seconds. Since this is less than the preset duration, the node is marked as incomplete. Raw electrocardiographic data includes ECG waveforms and heart rate. For example, during the execution of the variability response acquisition sub-path, the evaluation object's real-time ECG signal is recorded at a sampling rate of 1000Hz to fully capture the characteristic changes in the variability response.
[0089] The execution status data of the path is bound to the raw ECG data of the corresponding time period. A unique identifier is assigned to each sub-path node, and a correlation is established between the node identifier, execution status, timestamp, and ECG data segment. For example, in the second executed variant response acquisition sub-path node, the node identifier 2, execution status normal, timestamp 11 to 21 seconds, and the corresponding 10-second ECG signal segment are bound and stored to ensure that the execution status and physiological data of each node can be quickly located. The associated data of all nodes are arranged in the execution order to form a sequence structure isomorphic to the preset comprehensive ECG assessment path, thus constituting a complete actual ECG assessment path.
[0090] During the data acquisition process, the execution status and data quality of the evaluation object are continuously monitored. If any abnormalities occur, such as interruption of sub-path node execution, ECG signal interference, or data loss, the abnormal status is immediately recorded and the corresponding node is marked. At the same time, the original data already collected is retained, ensuring that the entire acquisition process is not interrupted. For example, if external interference occurs in the ECG signal when the evaluation object is executing the trend extrapolation analysis sub-path, the data quality of that node is marked as low, and the original interference data is retained. This ensures that the deviation characteristics of that node can be identified during subsequent sequence alignment, and that the entire evaluation process will not fail due to a single abnormality.
[0091] Record all states and physiological data of the evaluation subject when executing the preset path. The sequence structure of the data corresponds completely to the preset path. For example, if the node order of the preset path is 1. Baseline acquisition, 2. Variance acquisition, 3. Baseline analysis, 4. Variance analysis, and the node order of the actual path is 1. Baseline acquisition, 3. Baseline analysis, 2. Variance acquisition, 4. Variance analysis, then record the sequence deviation completely in the actual path, and quickly locate the abnormal segment during subsequent comparison.
[0092] The abnormal assessment path segments are obtained by sequence comparison between the actual ECG assessment path and the comprehensive ECG assessment path, and the initial assessment coefficients are obtained according to the type and degree of difference of the abnormal assessment path segments.
[0093] First, the actual ECG assessment path and the preset comprehensive ECG assessment path are converted into a structured sequence of nodes. Each node contains an execution order identifier, execution parameter requirements, and data acquisition rules to ensure that the two paths have a unified structure that can be directly compared.
[0094] Using the node sequence of the comprehensive ECG assessment pathway as a benchmark, each node is sequentially aligned and matched with the nodes of the actual ECG assessment pathway. Matching dimensions include node execution order, execution time, and data integrity. Abnormal assessment pathway segments are identified by discrepancies between the actual and preset pathways. For example, the node sequence of the comprehensive ECG assessment pathway might be: Node 1 baseline acquisition, Node 2 variation acquisition, Node 3 baseline analysis, Node 4 variation analysis, and Node 5 trend deduction. The node sequence of the actual ECG assessment pathway might be: Node 1 baseline acquisition, Node 3 baseline analysis, Node 2 variation acquisition, Node 4 variation analysis, and Node 5 trend deduction. During the matching of the second node, it is found that the actual node is Node 3, which does not match the preset Node 2. Similarly, during the matching of the third node, it is found that the actual node is Node 2, which does not match the preset Node 3. Therefore, a segment with an disordered order between Node 2 and Node 3 is identified, and this segment is thus an abnormal assessment pathway segment.
[0095] Based on deviation characteristics, fragment types are categorized, with common types including disordered execution fragments, missing execution fragments, and parameter deviation fragments. Disordered execution fragments refer to instances where the execution order of nodes in the actual path does not conform to the preset path arrangement rules, such as swapping the order of nodes 2 and 3. Missing execution fragments refer to instances where some nodes in the actual path were not executed according to the preset path, such as the absence of node 5 in the actual path for trend deduction. Parameter deviation fragments refer to instances where the execution parameters of nodes in the actual path do not conform to the preset requirements, such as the preset execution time for baseline acquisition of node 1 being 10 seconds, while the actual execution time is 6 seconds, exceeding the preset threshold.
[0096] The degree of difference is calculated for each type of abnormal segment. The calculation method varies depending on the type. For segments with disordered order, the degree of difference is calculated by the node position deviation value, with the formula d = |actual node position - preset node position|. For example, if node 2 is at position 2 in the preset path and at position 3 in the actual path, the position deviation value is |3-2| = 1, meaning the degree of difference is 1. For segments with missing execution, the degree of difference is calculated by the ratio of the number of missing nodes to the total number of nodes, with the formula d = number of missing nodes / total number of nodes in the preset path × 100. For example, if the total number of nodes in the preset path is 5 and the actual path is missing 1 node, the degree of difference is 1 / 5 × 100 = 20. For segments with parameter deviation, the degree of difference is calculated by the parameter deviation rate, with the formula d = |actual parameter value - preset parameter value| / preset parameter value × 100. For example, if the preset execution time of node 1 is 10 seconds and the actual execution time is 6 seconds, the parameter deviation rate is |6-10| / 10 × 100 = 40, meaning the degree of difference is 40.
[0097] An impact weight is assigned to each abnormal assessment path segment based on its type. The weight value is set according to the degree of influence of that type of deviation on the ECG assessment result. For example, a segment with disordered sequence has a weight of 2, a segment with missing execution has a weight of 3, and a segment with parameter deviation has a weight of 1. The greater the impact of the deviation on the assessment result, the higher the corresponding weight value. The initial assessment coefficient is calculated using the formula: Initial Assessment Coefficient = Σ(Weight of Abnormal Segment × Degree of Difference of Abnormal Segment). For example, if a comparison identifies three abnormal segments: a segment with disordered sequence (degree of difference 1), a segment with missing execution (degree of difference 20), and a segment with parameter deviation (degree of difference 40), with corresponding weights of 2, 3, and 1 respectively, then the initial assessment coefficient = 2 × 1 + 3 × 20 + 1 × 40 = 102. A higher initial assessment coefficient indicates a greater deviation between the actual path and the preset path, and a higher degree of abnormality in the ECG assessment process.
[0098] It also includes the following steps:
[0099] Obtain the evaluation results corresponding to the initial evaluation coefficients;
[0100] The assessment results include the subject's electrocardiogram health level and imbalance status.
[0101] The correlation between anomaly assessment path segments and assessment dimensions is obtained based on the distribution of anomaly assessment path segments;
[0102] The distribution of abnormal assessment path segments includes the segment's position within the comprehensive ECG assessment path, the number of assessment nodes involved in the sub-pathway of its respective imbalance type, and each assessment dimension corresponding to a type of ECG imbalance, such as baseline stability imbalance, variability response imbalance, trend extrapolation imbalance. The number of abnormal nodes within each assessment dimension is counted, and the correlation degree between the abnormal segment and the corresponding assessment dimension is obtained using the correlation formula: Correlation R = (Number of abnormal nodes in the assessment dimension ÷ Total number of nodes in the assessment dimension) × 100. For example, if the baseline stability imbalance dimension has 4 nodes, and 3 of them are abnormal, the correlation R = 3 ÷ 4 × 100 = 75, indicating a high correlation between this abnormal segment and the baseline stability imbalance dimension, and that the assessment data for this dimension is significantly affected by path abnormalities.
[0103] Determine the weight of the impact of abnormal assessment path segments on the assessment results based on the correlation relationships;
[0104] Pre-assigned base weights to different assessment dimensions. These base weights are determined by the dimension's importance to the ECG health assessment results. For example, baseline stability imbalance, as the core assessment dimension, has a base weight of 3; variability response imbalance, as an auxiliary assessment dimension, has a base weight of 2; and trend extrapolation imbalance, as a validation dimension, has a base weight of 1. The impact weight of each anomalous segment is determined by both the base weight of its respective assessment dimension and its correlation degree. The impact weight W = base weight × correlation degree R ÷ 100. The higher the correlation degree, the greater the impact weight. For example, a baseline stability imbalance anomalous segment with a correlation degree of 75 has an impact weight W = 3 × 75 ÷ 100 = 2.25.
[0105] Determine whether the initial evaluation coefficients exceed the preset reasonable range based on the influence weights;
[0106] A baseline reasonable range for the initial evaluation coefficient is preset, such as 0 to 100. The upper limit of this reasonable range is dynamically adjusted based on the sum of the influence weights of all abnormal segments. The adjusted reasonable upper limit = baseline reasonable upper limit - ΣW × adjustment coefficient, where the adjustment coefficient is a preset constant, for example, set to 10, and ΣW is the sum of the influence weights of all abnormal segments. The higher the weight, the lower the adjusted reasonable upper limit, meaning a smaller allowable deviation in the initial evaluation coefficient, because abnormal segments with high weights have a greater impact on the evaluation result, resulting in a lower tolerance for error. For example, if the baseline reasonable upper limit is 100, the sum of the influence weights of all abnormal segments ΣW = 2.25 + 0.8 = 3.05, and the adjustment coefficient is 10, then the adjusted reasonable upper limit = 100 - 3.05 × 10 = 69.5. The initial evaluation coefficient is compared with the adjusted reasonable upper limit. If the initial evaluation coefficient is greater than 69.5, it is determined to exceed the preset reasonable range.
[0107] If the initial assessment coefficient exceeds the preset reasonable range, it is determined that there is a deviation in the electrocardiogram health assessment.
[0108] If the initial assessment coefficients exceed the preset reasonable range, the ECG health assessment is deemed to have a bias. This means that the assessment results obtained based on the initial assessment coefficients are greatly affected by abnormal pathway segments, lack reliability, and cannot accurately reflect the ECG health status of the assessed individual.
[0109] If the initial assessment coefficients indicate a bias in the ECG health assessment, a feedback assessment path is dynamically generated based on the type of abnormal assessment path segment. This includes the following steps:
[0110] The abnormal evaluation path segments are classified into categories to obtain the classification results; the classification results include path segments with missing path execution, path segments with disordered path order, and path segments with parameter perception deviation.
[0111] The classification results include path segments with missing execution, path segments with disordered order, and path segments with parameter perception deviation. Classification is achieved by matching the deviation characteristics of abnormal segments with three preset abnormal patterns. Path segments with missing execution are characterized by nodes in the actual path that were not executed within the preset path; for example, a preset comprehensive ECG assessment path may contain 5 nodes, but the actual path only executes 4, missing the 3rd node. Path segments with disordered order are characterized by a node execution order that does not conform to the preset arrangement rules; for example, the preset order is node 1, node 2, node 3, but the actual execution order is node 1, node 3, node 2. Path segments with parameter perception deviation are characterized by deviations between the node execution parameters and preset requirements; for example, the preset execution time for a node is 10 seconds, but the actual execution time is only 6 seconds, or the preset ECG signal sampling rate is 1000Hz, but the actual sampling rate is only 500Hz. The matching degree is determined by the following formula: matching degree = number of overlap points between abnormal fragment features and category features ÷ total number of category features × 100. When the matching degree is greater than the preset threshold of 80, the abnormal fragment is determined to belong to the corresponding category. After classifying all fragments, a complete classification result is obtained.
[0112] The direction of evaluation correction is determined based on the inherent characteristics of the division results;
[0113] Each type of abnormal segment has a corresponding inherent characteristic, and based on these characteristics, the core correction direction for feedback evaluation is clearly defined. The inherent characteristic of a missing path execution segment is incomplete node execution, indicating that some key evaluation steps were not executed in the initial evaluation. The correction direction is to complete the missing node execution steps to ensure that all key nodes related to the evaluation results are fully executed.
[0114] The inherent characteristic of path order disordered segments is the disordered execution order of nodes, indicating that there is an error in the cross arrangement of assessment sub-paths for different ECG imbalance types in the initial assessment. The correction direction is to restore the preset node execution order or adjust the node order to reduce the correlation between different imbalance types.
[0115] The inherent characteristic of parameter-aware deviation segments is that the execution parameters do not conform to the preset standards, indicating a deviation in the parameter settings of data collection during the initial evaluation. The correction direction is to calibrate the execution parameters of the nodes, such as adjusting the execution duration and sampling rate, to ensure that the parameters are consistent with the preset requirements. For example, if there are path execution missing segments in the partitioning results, completing the missing nodes will be the core correction direction; if there are parameter-aware deviation segments, then calibrating the execution parameters will be the correction direction, ensuring that the secondary evaluation can specifically address the initial deviation.
[0116] The perceived assessment content of the comprehensive ECG assessment path is determined based on the weights of the ECG assessment dimensions corresponding to the segmentation results.
[0117] Fixed weights are pre-assigned to different ECG assessment dimensions. The weight value is determined by the importance of that dimension to the ECG health assessment result. For example, the weight of the baseline stability imbalance dimension is 3, the weight of the variability response imbalance dimension is 2, and the weight of the trend inference imbalance dimension is 1. Each abnormal segment corresponds to an assessment dimension. The priority and coverage of the perception assessment content are determined according to the weight of that dimension. The higher the weight of the dimension, the more detailed and comprehensive the perception assessment content.
[0118] The number of nodes N = dimension weight × preset number of basic nodes. The preset number of basic nodes is 2. For example, the weight of the baseline stability imbalance dimension is 3 and the number of basic nodes is 2, so the number of nodes N = 3 × 2 = 6. That is, the perception assessment content corresponding to this dimension includes 6 nodes, including baseline signal acquisition, baseline stability analysis, and baseline deviation verification. On the other hand, the weight of the trend inference imbalance dimension is 1 and the number of nodes N = 1 × 2 = 2. The corresponding perception assessment content only includes trend signal acquisition and trend feature analysis, thus ensuring that the assessment content of high-weight dimensions is more comprehensive.
[0119] The feedback assessment path is obtained by reorganizing the path elements based on the assessment correction direction and the perceived assessment content.
[0120] The feedback evaluation path is obtained by recombining path elements based on the evaluation correction direction and perceived evaluation content. Path elements include node execution order, execution parameters, and evaluation stages. These elements are recombined with the correction direction and perceived evaluation content to construct a completely new evaluation path. If the correction direction is to complete missing segments of the path execution, all nodes corresponding to the perceived evaluation content are arranged in a preset order to supplement the nodes missing in the initial evaluation, while setting reasonable execution parameters to ensure that each node can be executed completely. If the correction direction is to restore the path order, the nodes of the perceived evaluation content are arranged according to a preset cross-arrangement rule to avoid the problem of disordered order recurrence. If the correction direction is to calibrate parameters, the execution parameters of the nodes are reset, such as adjusting the execution duration to a preset 10 seconds and the sampling rate to 1000Hz to ensure that the parameters meet the requirements. The recombined feedback evaluation path corrects the deviations in the initial evaluation. For example, if the initial evaluation lacks a baseline analysis node, the feedback evaluation path supplements this node; if the node order in the initial evaluation is disordered, the feedback evaluation path arranges them in the correct order; if the parameters in the initial evaluation deviate, the feedback evaluation path calibrates the parameters.
[0121] Collect the actual feedback evaluation path of the target evaluation object when it performs the evaluation task along the feedback evaluation path;
[0122] The feedback evaluation path is customized to address deviations in the initial evaluation. It corrects issues such as missing execution, disordered sequence, or parameter deviations in the initial path. Each node has clearly defined preset requirements for execution duration, data sampling rate, execution order, and data acquisition rules. For example, to address the issue of missing baseline analysis nodes in the initial evaluation, the feedback evaluation path supplements this node and sets an execution duration of 10 seconds and a sampling rate of 1000Hz.
[0123] The entire execution process of the target evaluation object is tracked in real time, collecting path execution status data, including the actual execution order, start and end times, effective execution duration, and execution completion status of each node. The effective execution duration of a node is expressed by the formula: ΔT 有效 =T6-T5-T 停 Where T5 is the actual start time of the node, T6 is the actual end time of the node, and T... 停 The effective execution time is the pause time during execution. For example, if the preset execution time of a node in the feedback evaluation path is 10 seconds, the actual start time T5 is 0 seconds, the actual end time T6 is 15 seconds, and there is a 3-second pause during execution, then the effective execution time is ΔT. 有效 =15-0-3=12 seconds. Compare the effective execution time of this node with the preset time and mark whether the execution status meets the requirements. Record the identifier of each node according to the actual execution order. For example, the preset node order of the feedback path is node 1, node 2, node 3, and the actual execution order is node 1, node 3, node 2. Record the actual order completely to provide a basis for subsequent sequence comparison.
[0124] According to the preset sampling rate and data acquisition rules of each node in the feedback evaluation path, the electrocardiographic (ECG) data of the evaluation subject, including ECG waveforms and heart rate, are collected synchronously. The amount of sampled data, K, is calculated as: Sampling rate × Effective execution time. For example, if the preset sampling rate for a node is 1000Hz and the effective execution time is 10 seconds, then the number of ECG data points to be collected is K = 1000 × 10 = 10000. The actual number of valid data points collected is recorded according to this standard to determine the completeness of the data acquisition. For example, if only 9000 valid data points are actually collected, the data completeness of that node is marked as 90%.
[0125] The collected path execution status data is bound to the raw ECG data for the corresponding time period. A unique identifier is assigned to each node, and a correlation is established between the node identifier, actual execution order, effective execution duration, timestamp, ECG data segment, and data integrity. For example, for node 3, which is the second node in the actual execution order, the node identifier 3, actual execution order 2, effective execution duration of 8 seconds, timestamps from 5 to 13 seconds, an 8-second ECG signal segment for the corresponding time period, and 100% data integrity are bound and stored to ensure that the execution status and physiological data corresponding to each node can be quickly located. The associated data of all nodes are arranged according to the actual execution order, forming a sequence structure isomorphic to the feedback evaluation path, thus constituting a complete actual feedback evaluation path.
[0126] During the data acquisition process, the execution status and data quality of the evaluation object are continuously monitored. If any abnormalities occur, such as node execution interruption, ECG signal interference, or data loss, the abnormal status is immediately recorded and the corresponding node is marked. Simultaneously, the original data already collected is retained, ensuring the entire acquisition process is not interrupted. For example, if external interference occurs to the ECG signal when the evaluation object is executing node 2, the data quality of that node is marked as low, and the original interference data is retained. This ensures that the deviation characteristics of that node can be identified during subsequent comparisons, preventing the secondary evaluation process from failing due to a single abnormality. The actual feedback evaluation path completely records all states and physiological data of the target evaluation object during the execution of the feedback evaluation path, and its sequence structure corresponds perfectly to the preset feedback path.
[0127] The feedback evaluation coefficient is obtained by comparing the actual feedback evaluation path with the feedback evaluation path, specifically including the following steps:
[0128] Compare the actual feedback evaluation path with the execution direction and content alignment of the feedback evaluation path, and screen for deviations in the path execution process based on the alignment of execution direction and content.
[0129] The degree of impact of deviation from the content is determined based on the differences in the abnormal assessment path segments;
[0130] The feedback evaluation coefficient is obtained based on the degree of impact.
[0131] The actual feedback evaluation path and the feedback evaluation path are both decomposed into structured node sequences. Each node includes an execution order identifier, preset execution parameters, and data collection rules. The execution direction refers to the arrangement order of the nodes, and the content fit refers to the degree of matching between the node execution parameters and the data collection results and preset requirements. The comparison process matches the execution direction node by node in sequence. For example, if the preset feedback evaluation path's node order is Node 1 baseline collection, Node 2 baseline analysis, Node 3 variation collection, and Node 4 variation analysis, and the actual feedback evaluation path's node order is Node 1 baseline collection, Node 3 variation collection, Node 2 baseline analysis, and Node 4 variation analysis, then the sequence deviation between Node 2 and Node 3 is identified.
[0132] Content fit is calculated using a node fit formula: Node Fit C = 1 - (absolute value of the difference between the actual parameter and the preset parameter) ÷ (result of the preset parameter) × 100. For example, if the preset execution time of node 1 is 10 seconds and the actual effective execution time is 8 seconds, the node fit C = 80. If the node fit is lower than the preset threshold of 90, then the node is considered to have content deviation. All node segments whose execution direction does not match the content fit are marked as deviating content.
[0133] The impact of deviations is determined based on the differential characteristics of abnormal assessment path segments. Different types of deviations correspond to different differential characteristics of abnormal assessment path segments. For example, the differential characteristic of path sequence disorder deviations is node position deviation, the differential characteristic of parameter perception deviations is parameter deviation rate, and the differential characteristic of execution missing deviations is the importance level of the missing node. Basic weights are pre-set for different deviation types. For example, the weight for path execution missing deviations is 4, the weight for path sequence disorder deviations is 3, and the weight for parameter perception deviations is 2. The weight value is determined by the degree of influence of this type of deviation on the ECG assessment result; the greater the influence, the higher the weight. Simultaneously, the standardized deviation degree is calculated: Deviation degree P = Deviation feature value ÷ Maximum value of the corresponding dimension × 100. For example, if the node position deviation is 1 and the preset total number of nodes is 4, the deviation degree P = 1 ÷ 4 × 100 = 25; if the parameter deviation rate is 20, the deviation degree P equals 20; and if the missing node is a core assessment node with an importance level of 1, the deviation degree P equals 100. The degree of influence I = deviation type weight × deviation degree P ÷ 100. For example, if the weight of the path order disorder deviation is 3 and the deviation degree D is 25, then the degree of influence I = 3 × 25 ÷ 100 = 0.75; if the weight of the parameter perception deviation deviation is 2 and the deviation degree P is 20, then the degree of influence I = 2 × 20 ÷ 100 = 0.4.
[0134] The feedback evaluation coefficient is derived based on the degree of impact of all deviations. This coefficient quantifies the overall deviation level of the secondary evaluation. The feedback evaluation coefficient F = baseline score - sum of the degree of impact of all deviations × adjustment coefficient. The adjustment coefficient is a preset fixed constant, for example, 10, used to amplify the impact of deviations and facilitate differentiation between different evaluation scenarios. For instance, if a screening identifies two deviations with impact levels of 0.75 and 0.4 respectively, their sum is 1.15, then the feedback evaluation coefficient F = 100 - 1.15 × 10 = 88.5. A lower feedback evaluation coefficient indicates a greater deviation between the actual feedback path and the preset feedback path, and a weaker correction effect of the secondary evaluation on the initial deviation; a higher coefficient indicates a better correction effect and higher reliability of the secondary evaluation data.
[0135] A comprehensive evaluation coefficient is obtained based on the initial evaluation coefficient and the feedback evaluation coefficient. The electrocardiogram health risk level of the target assessment subject is then output based on the comprehensive evaluation coefficient. The specific steps include:
[0136] The degree of path abnormality difference is determined based on the ECG function fluctuation characteristics reflected by the initial assessment coefficient and the feedback assessment coefficient.
[0137] The initial assessment coefficient is a quantitative result of the pathway deviation in the first ECG assessment; a higher value indicates a greater deviation in the initial assessment. The feedback assessment coefficient is a quantitative result of the pathway deviation in the second feedback assessment; a higher value indicates a smaller deviation in the second assessment and a better correction effect. The degree of pathway abnormality difference is calculated using the formula: Pathway Abnormality Difference M = Absolute value of the difference between the initial assessment coefficient A and the feedback assessment coefficient B ÷ Initial Assessment Coefficient A × 100. For example, if the initial assessment coefficient A is 100 and the feedback assessment coefficient B is 80, the pathway abnormality difference M = |100 - 80| ÷ 100 × 100 = 20, indicating a 20% difference in deviation between the two assessments, and the ECG function fluctuation amplitude is at a moderate level.
[0138] A comprehensive evaluation coefficient is obtained based on the degree of abnormality in the path.
[0139] Different weights are pre-assigned to the initial evaluation coefficient and the feedback evaluation coefficient. The weight of the initial evaluation coefficient N1 is 0.4, and the weight of the feedback evaluation coefficient N2 is 0.6. The comprehensive evaluation coefficient X = A × N1 + B × N2 × (1 - G ÷ 100), where G is the degree of path anomaly. For example, if A is 100, B is 80, N1 is 0.4, N2 is 0.6, and G is 20, then the comprehensive evaluation coefficient X is equal to 70.4.
[0140] The actual manifestation of electrocardiogram dysfunction is determined based on the comprehensive evaluation coefficient and the abnormal distribution characteristics of the pathway.
[0141] The abnormal distribution characteristics of the pathway refer to the proportion of all abnormal assessment pathway segments distributed across different ECG assessment dimensions, including baseline stability imbalance, variability response imbalance, and trend extrapolation imbalance. The percentage of abnormal segments within each dimension is statistically analyzed; for example, 60% for baseline stability imbalance, 30% for variability response imbalance, and 10% for trend extrapolation imbalance. The main manifestations of imbalance are then determined by combining these findings with the comprehensive assessment coefficient.
[0142] When the comprehensive assessment coefficient is 70.4, based on the fact that 60% of the abnormal segments are distributed in the baseline stability imbalance dimension, it can be determined that the actual manifestation of ECG functional imbalance is mainly baseline fluctuation abnormality, accompanied by unstable variability response. If the comprehensive assessment coefficient is low and the abnormal segments are evenly distributed across multiple dimensions, it indicates that the ECG functional imbalance belongs to a multidimensional comprehensive abnormal state.
[0143] Based on the normal fluctuation patterns of electrocardiography, an appropriate grading standard is established, and the electrocardiographic health risk level of the target assessment object is obtained based on the grading standard and the comprehensive evaluation coefficient.
[0144] Based on the normal fluctuation data of electrocardiogram (ECG), risk levels were divided into five ranges, with the comprehensive assessment coefficient divided into five levels: 90 to 100 represents low risk, indicating that ECG function is within the normal fluctuation range and there is no obvious imbalance; 70 to 89 represents low to medium risk, indicating slight ECG function fluctuations, mainly in a single dimension; 50 to 69 represents medium risk, indicating moderate ECG function imbalance with abnormalities in multiple dimensions; 30 to 49 represents medium to high risk, indicating relatively obvious ECG function imbalance with persistent abnormal fluctuations; and 0 to 29 represents high risk, indicating severe ECG function imbalance and significant health risks.
[0145] For example, a comprehensive assessment coefficient of 70.4 falls within the range of 70 to 89, corresponding to a low to medium risk level. This indicates that the overall electrocardiographic health of the assessed individual is acceptable, but abnormal baseline fluctuations require attention. If the comprehensive assessment coefficient is 40, it falls within the range of 30 to 49, corresponding to a medium to high risk level, suggesting the need for medical examination and intervention.
[0146] A multi-parameter fusion and intelligent sensing electrocardiogram health assessment system includes:
[0147] Generation module: Based on the ECG functional imbalance data of the target assessment object, dynamically generate a comprehensive ECG assessment path that integrates at least two different types of assessment dimensions; wherein, the comprehensive ECG assessment path consists of multiple assessment sub-paths that are cross-sorted according to a preset arrangement rule;
[0148] Acquisition module: Acquires the actual electrocardiogram (ECG) assessment path of the target assessment object as it performs the assessment task along the comprehensive ECG assessment path;
[0149] Processing module: The actual ECG assessment pathway is compared with the comprehensive ECG assessment pathway to obtain abnormal assessment pathway segments, and the initial assessment coefficients are obtained based on the type and degree of difference of the abnormal assessment pathway segments.
[0150] The generation and acquisition module: If the initial evaluation coefficient indicates that there is a deviation in the ECG health assessment, the feedback evaluation path is dynamically generated based on the type of abnormal evaluation path segment, and the actual feedback evaluation path of the target evaluation object when performing the evaluation task along the feedback evaluation path is collected.
[0151] Comparison module: Obtains feedback evaluation coefficients by comparing the actual feedback evaluation path with the actual feedback evaluation path;
[0152] Output module: Obtains a comprehensive evaluation coefficient based on the initial evaluation coefficient and the feedback evaluation coefficient, and outputs the electrocardiogram health risk level of the target assessment object based on the comprehensive evaluation coefficient.
[0153] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0154] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A multi-parameter fusion and intelligent sensing method for electrocardiogram health assessment, characterized in that, Includes the following steps: Based on the ECG functional imbalance data of the target assessment object, a comprehensive ECG assessment path is dynamically generated that integrates at least two different types of assessment dimensions; wherein, the comprehensive ECG assessment path consists of multiple assessment sub-paths that are cross-sorted according to a preset arrangement rule; Collect the actual electrocardiogram (ECG) assessment path of the target assessment subject when performing the assessment task along the comprehensive ECG assessment path; The abnormal assessment path segments are obtained by sequence comparison between the actual ECG assessment path and the comprehensive ECG assessment path, and the initial assessment coefficients are obtained according to the type and degree of difference of the abnormal assessment path segments. If the initial assessment coefficient indicates a deviation in the ECG health assessment, a feedback assessment path is dynamically generated based on the type of abnormal assessment path segment, and the actual feedback assessment path of the target assessment object when performing the assessment task along the feedback assessment path is collected. The feedback evaluation coefficient is obtained by comparing the actual feedback evaluation path with the feedback evaluation path. The comprehensive evaluation coefficient is obtained based on the initial evaluation coefficient and the feedback evaluation coefficient. The electrocardiogram health risk level of the target assessment object is then output based on the comprehensive evaluation coefficient.
2. The electrocardiogram health assessment method based on multi-parameter fusion and intelligent sensing according to claim 1, characterized in that, Based on the ECG functional imbalance data of the target assessment object, a comprehensive ECG assessment path that integrates at least two different types of assessment dimensions is dynamically generated, specifically including the following steps: Obtain electrocardiogram (ECG) functional imbalance data of the target assessment subjects, and determine at least two types of ECG functional imbalance and the degree of imbalance corresponding to each type of ECG functional imbalance based on the ECG functional imbalance data; Different assessment parameters for different types of electrocardiogram (ECG) imbalances are set according to the degree of imbalance of different ECG imbalance types. For each type of electrocardiogram imbalance, a corresponding preliminary assessment path is generated based on the assessment parameters. The assessment sub-paths of different preliminary assessment paths are cross-arranged and integrated according to the preset path arrangement rules to obtain the comprehensive ECG assessment path.
3. The ECG health assessment method and system based on multi-parameter fusion and intelligent sensing according to claim 2, characterized in that the ECG imbalance type includes at least two of baseline stable imbalance, variability response imbalance, and trend extrapolation imbalance.
4. The electrocardiogram health assessment method based on multi-parameter fusion and intelligent sensing according to claim 2, characterized in that, According to the preset path arrangement rules, the assessment sub-paths of different preliminary assessment paths are cross-arranged and integrated to obtain a comprehensive ECG assessment path, which specifically includes the following steps: Based on the preset pathway arrangement rules and the various types and degrees of ECG imbalance, the functional attributes corresponding to each preliminary assessment pathway are divided. Based on functional attributes, each preliminary evaluation path is broken down into its sub-evaluation paths; The assessment sub-paths are arranged alternately and crosswise according to the associated effects of different types of electrocardiogram imbalance. The comprehensive ECG assessment pathway is obtained by systematically integrating all the cross-arranged assessment sub-paths.
5. The electrocardiogram health assessment method based on multi-parameter fusion and intelligent sensing according to claim 1, characterized in that, The abnormal assessment pathway fragments are obtained by sequence alignment between the actual ECG assessment pathway and the comprehensive ECG assessment pathway. The specific steps include: Extract assessment behavior units from the comprehensive ECG assessment pathway and the actual ECG assessment pathway; Based on the execution arrangement of the assessment behavior units, identify abnormal sections where there is misalignment or content deviation between the comprehensive ECG assessment path and the actual ECG assessment path. Anomaly assessment path segments are obtained by performing boundary regularization and continuous integration on the abnormal sections.
6. The electrocardiogram health assessment method and system based on multi-parameter fusion and intelligent sensing according to claim 1, characterized in that, It also includes the following steps: Obtain the evaluation results corresponding to the initial evaluation coefficients; The correlation between anomaly assessment path segments and assessment dimensions is obtained based on the distribution of anomaly assessment path segments; Determine the weight of the impact of abnormal assessment path segments on the assessment results based on the correlation relationships; Determine whether the initial evaluation coefficients exceed the preset reasonable range based on the influence weights; If the initial assessment coefficient exceeds the preset reasonable range, it is determined that there is a deviation in the electrocardiogram health assessment.
7. The electrocardiogram health assessment method based on multi-parameter fusion and intelligent sensing according to claim 1, characterized in that, The feedback evaluation path is dynamically generated based on the type of the anomaly evaluation path fragment, specifically including the following steps: The abnormal evaluation path segments are classified into categories to obtain the classification results; wherein, the classification results include path segments with missing path execution, path segments with disordered path order, and path segments with parameter perception deviation. The direction of evaluation correction is determined based on the inherent characteristics of the division results; The perceived assessment content of the comprehensive ECG assessment path is determined based on the weights of the ECG assessment dimensions corresponding to the segmentation results. The feedback assessment path is obtained by reorganizing the path elements based on the assessment correction direction and the perceived assessment content.
8. The electrocardiogram health assessment method based on multi-parameter fusion and intelligent sensing according to claim 1, characterized in that, The feedback evaluation coefficient is obtained by comparing the actual feedback evaluation path with the feedback evaluation path, specifically including the following steps: Compare the actual feedback evaluation path with the execution direction and content alignment of the feedback evaluation path, and screen for deviations in the path execution process based on the alignment of execution direction and content. The degree of impact of deviation from the content is determined based on the differences in the abnormal assessment path segments; The feedback evaluation coefficient is obtained based on the degree of impact.
9. The electrocardiogram health assessment method based on multi-parameter fusion and intelligent sensing according to claim 8, characterized in that, A comprehensive evaluation coefficient is obtained based on the initial evaluation coefficient and the feedback evaluation coefficient. The electrocardiogram health risk level of the target assessment subject is then output based on the comprehensive evaluation coefficient. The specific steps include: The degree of path abnormality difference is determined based on the ECG function fluctuation characteristics reflected by the initial assessment coefficient and the feedback assessment coefficient. A comprehensive evaluation coefficient is obtained based on the degree of abnormality in the path. The actual manifestation of electrocardiogram dysfunction is determined based on the comprehensive evaluation coefficient and the abnormal distribution characteristics of the pathway. Based on the normal fluctuation patterns of electrocardiography, an appropriate grading standard is established, and the electrocardiographic health risk level of the target assessment object is obtained based on the grading standard and the comprehensive evaluation coefficient.
10. A multi-parameter fusion and intelligent sensing electrocardiogram health assessment system, applied to the multi-parameter fusion and intelligent sensing electrocardiogram health assessment method according to any one of claims 1 to 9, characterized in that, include: Generation module: Dynamically generates a comprehensive ECG assessment path that integrates at least two different types of assessment dimensions based on the ECG functional imbalance data of the target assessment object; wherein, the comprehensive ECG assessment path consists of multiple assessment sub-paths that are cross-sorted according to a preset arrangement rule; Acquisition module: Acquires the actual electrocardiogram (ECG) assessment path of the target assessment object as it performs the assessment task along the comprehensive ECG assessment path; Processing module: The actual ECG assessment pathway is compared with the comprehensive ECG assessment pathway to obtain abnormal assessment pathway segments, and the initial assessment coefficients are obtained based on the type and degree of difference of the abnormal assessment pathway segments. The generation and acquisition module: If the initial evaluation coefficient indicates that there is a deviation in the ECG health assessment, the feedback evaluation path is dynamically generated based on the type of abnormal evaluation path segment, and the actual feedback evaluation path of the target evaluation object when performing the evaluation task along the feedback evaluation path is collected. Comparison module: Obtains feedback evaluation coefficients by comparing the actual feedback evaluation path with the actual feedback evaluation path; Output module: Obtains a comprehensive evaluation coefficient based on the initial evaluation coefficient and the feedback evaluation coefficient, and outputs the electrocardiogram health risk level of the target assessment object based on the comprehensive evaluation coefficient.