Detection method and system for non-invasive evaluation of degree and position of functional myocardial ischemia

Abstract

The invention provides a detection method and system for non-invasively evaluating the degree and position of functional myocardial ischemia. The method comprises the following steps: acquiring 12-lead high-frequency electrocardiosignal data of a testee; preprocessing the signal, and extracting a QRS wave band signal; performing signal averaging on the qualified QRS wave bands according to leads to generate 12 average high-frequency QRS; extracting a feature combination reflecting the functional myocardial ischemia and having clinical physiological significance from the signals; mapping leads to the heart apex, the front wall, the dividing wall, the lower wall and the side wall and aggregating regional features; obtaining a region score through linear combination; monotonicity is calibrated as regional ischemia percentage by adopting isotonic regression; and outputting the total heart ischemia percentage and the ischemia position in an equal-weight average manner. The method is highly related to the SPECT / CT gold standard, the degree and position of functional myocardial ischemia can be noninvasively and accurately detected, the method is not limited by detection sites, a new method is provided for checking myocardial ischemia, and the incidence rate of malignant heart diseases is reduced.

Description

Technical Field

[0001] This invention relates to the fields of medical engineering and electrocardiogram signal processing, and in particular to a non-invasive method and system for assessing the degree and location of functional myocardial ischemia. Specifically, it is a method and system for quantifying the percentage of functional myocardial ischemia and locating the five zones (anterior wall / septal wall / inferior wall / lateral wall / apex) based on high-frequency electrocardiogram (HF-ECG). Background Technology

[0002] Myocardial ischemia is a functional state of hypoxia caused by insufficient blood supply to the heart. It is commonly seen in coronary artery disease and can easily induce myocardial infarction or sudden death. The three main coronary arteries—the left anterior descending artery, the left circumflex artery, and the right coronary artery—account for about 5% of the heart's blood vessels, and changes in these arteries can cause structural alterations in the heart. Microvessels account for more than 90% of the heart's blood vessels, and changes in these vessels can also cause myocardial ischemia. Early functional myocardial ischemia does not present with structural lesions.

[0003] Currently, the commonly used clinical methods for assessing myocardial ischemia are mainly for structural myocardial ischemia, such as: 1) coronary angiography, which requires catheter insertion and can only provide the anatomical degree of blockage of the three main coronary arteries of the heart. The severity of stenosis obtained by coronary angiography does not completely correspond to the functional changes in hemodynamics; 2) traditional electrocardiography, which is mainly based on low frequency (0-150Hz) and cannot effectively capture the subtle electrophysiological changes in functional myocardial ischemia.

[0004] Early functional myocardial ischemia does not yet show structural changes, and current clinical methods have limitations in quantifying and locating functional myocardial ischemia (especially microcirculatory dysfunction or stress-induced ischemia). Single-photon emission computed tomography (SPECT / CT) in myocardial perfusion imaging can provide functional assessment of myocardial blood flow distribution and is currently the "gold standard" for assessing functional myocardial ischemia, quantitatively evaluating the degree and location of functional myocardial ischemia. However, it is expensive, involves radiation exposure, and is not convenient for continuous bedside monitoring.

[0005] In recent years, high-frequency electrocardiography (150–>1000 Hz) has shown unique advantages in revealing cardiac depolarization abnormalities and is expected to sensitively reflect local conduction changes caused by ischemia. However, existing indicators (such as the High-Frequency Morphology Index (HFMI)) are mostly used for structural ischemia or qualitative judgment, and it is difficult to directly provide quantitative results of "ischemia percentage". Furthermore, it lacks stable mapping and calibration for anatomical regions such as the anterior wall, septum, inferior wall, lateral wall, and apex of the heart.

[0006] Therefore, how to accurately, non-invasively, and easily quantitatively assess the degree of functional myocardial ischemia, and achieve early screening and portable and rapid detection of myocardial ischemia, is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0007] To address at least one of the technical problems mentioned in the background above, this invention provides a non-invasive method and system for assessing the degree and location of functional myocardial ischemia. It extracts indicators that accurately reflect the degree of myocardial ischemia from high-frequency electrocardiogram signals and integrates this with a method for monotonic calibration of the ischemia percentage using SPECT / CT, achieving: (1) individual functional ischemia percentage estimation; (2) five-zone localization (anterior wall, septal wall, inferior wall, lateral wall, and apex); and (3) rapid, non-invasive bedside assessment. This addresses the current gap in non-invasive and accurate assessment of functional myocardial ischemia, achieving the invention's objective of rapid, effective, and real-time detection of functional myocardial ischemia in multiple scenarios.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] The first aspect of the present invention provides a non-invasive method for assessing the degree and location of functional myocardial ischemia, comprising the following steps:

[0010] Step 1: Acquire 12-lead high-frequency electrocardiogram (ECG) signal data of the test subject, with a sampling frequency >2000Hz and a longitudinal resolution of the signal amplitude reaching the microvolt level.

[0011] Step 2: Preprocess the signal to remove noise such as baseline drift and power line interference, locate the Q, R, and S wave groups in the high-frequency ECG signal, and extract the QRS band signal.

[0012] Step 3: Remove abnormal QRS bands, average the signals of the QRS bands that meet the conditions by lead, and finally generate 12 average high-frequency QRSs.

[0013] Step 4: Extract feature combinations that reflect functional myocardial ischemia and have clinical physiological significance from the above average high-frequency QRS;

[0014] Step 5: Map the leads to the apex, anterior wall, septum, inferior wall, and lateral wall of the heart and aggregate regional features;

[0015] Step 6: Obtain the regional score using a linear combination. ;

[0016] Step 7, use isotonic regression to... Monotonicity is defined as the percentage of regional ischemia. ;

[0017] Step 8: Output the percentage of total cardiac ischemia using an equal-weighted average. With ischemic sites.

[0018] Furthermore, the functional myocardial ischemia indicators extracted in step 4 include, but are not limited to, the coefficient of variation of the original high-frequency QRS signal. Frequency band energy ratio The number of regions where the amplitude of the envelope decreases Duration Complexity .

[0019] Furthermore, in step 4, the threshold for determining the amplitude reduction region is the amplitude ratio of adjacent peak values. 0.3 and peak spacing 10ms.

[0020] Furthermore, the fixed mappings in step 5 are: V1 / V2 → septum, V3 / V4 → anterior wall, I / aVL → lateral wall, II / III / aVF → inferior wall; V5 / V6 → apex. The regional feature vector is obtained by averaging the multi-lead indices mapped to each region r. .

[0021] Furthermore, the coefficients of the regional scores in step 6 Solve and fix the non-negative least squares with L2 regularization on the training data.

[0022] Furthermore, the monotonic calibration in step 7 The algorithm is implemented using the Pool-Adjacent-Violators (PAV) algorithm, and boundary constant extrapolation is used in the inference phase.

[0023] A second aspect of the present invention provides a non-invasive detection device for assessing the degree and location of functional myocardial ischemia, comprising an acquisition module, a preprocessing module, a signal averaging module, a feature calculation module, a region aggregation module, a scoring module, a monotonic calibration module, and a result output module. The acquisition module acquires high-frequency electrocardiogram signals from the subject; the preprocessing module removes noise such as baseline drift and power line interference from the signal, and also locates the Q / R / S wave groups and extracts the QRS band signal; the signal averaging module averages the qualified QRS bands by lead; the feature calculation module extracts feature combinations that reflect functional myocardial ischemia and have clinical physiological significance, including… , The region aggregation module aggregates features from 12 leads into five regions: apex, anterior wall, septal wall, inferior wall, and lateral wall, and uses averaging to obtain the feature vector for each region. The scoring module uses linear combinations of the region feature vectors to obtain the region score. The monotonic calibration module is used to apply isotonic regression to... The calibrated percentage ischemia area ischemia is used to output the percentage of whole-heart ischemia as an equal-weighted average. With ischemic sites.

[0024] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the above-described detection method.

[0025] A fourth aspect of the present invention provides a detection system comprising a computer device and an external input device for providing high-frequency electrocardiogram signal data of a test subject to an external device of the computer device. The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-described method for detecting functional myocardial ischemia.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. Unlike current detection methods for structural lesions, this method can detect functional myocardial ischemia non-invasively and accurately.

[0028] 2. , Complementary indicators constitute regional characteristics, and linear scoring is followed by monotonic calibration, taking into account amplitude, duration, morphological complexity and frequency band energy distribution, thereby reducing the risk of missed detections and false alarms.

[0029] 3. Precise Quantification: Directly outputs ischemia percentage score consistent with the gold standard. After monotonic calibration, it is consistent with the percentage scale of SPECT / CT, avoiding information loss from only binary thresholds. It facilitates individual longitudinal comparison, cross-device / cross-department aggregation, and efficacy quantification.

[0030] 4. Precise localization: Fixed lead-anatomical mapping (anterior wall / septal wall / inferior wall / lateral wall / apex) allows each region to output an independent percentage of ischemia, supporting site localization and multifocal assessment, and can be directly aligned with the regional report of the imaging.

[0031] 5. Based on conventional 12-lead electrodes and signal processing, it can be implemented immediately, without radiation and is repeatable; suitable for emergency chest pain triage, bedside assessment in wards, and continuous / dynamic monitoring before / after surgery and during the recovery period.

[0032] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0033] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0034] Figure 1 This is a flowchart of a method for detecting the degree of functional myocardial ischemia according to an embodiment of the present invention.

[0035] Figure 2 This is a comparison chart of high-frequency QRS indicators between healthy individuals and patients according to an embodiment of the present invention.

[0036] Figure 3 This is a schematic diagram of RAZ according to an embodiment of the present invention.

[0037] Figure 4 This is a localization diagram of the APECT / CT17 segment according to an embodiment of the present invention.

[0038] Figure 5 These are the lead diagrams corresponding to the five regions in this embodiment of the invention. Detailed Implementation

[0039] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0040] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, 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.

[0041] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0042] Example 1

[0043] Reference Figure 1 The non-invasive method for assessing the degree and location of functional myocardial ischemia in this embodiment includes the following eight steps:

[0044] Step 1: Collect high-frequency electrocardiogram signals from the subjects.

[0045] Specifically, a dedicated high-frequency ECG acquisition device was used to collect 12-lead ECG data from the test subjects.

[0046] High-frequency electrocardiogram (ECG) signals were collected in a relatively enclosed examination room, where there was no noise, suitable temperature and humidity, and no electromagnetic interference. A dedicated high-frequency ECG acquisition device was used to collect the signals from the test subjects. The device's sampling rate was set to 2kHz, and the longitudinal resolution of the signal amplitude reached the micro-amplitude level. The test subjects lay flat on the experimental bed, and the 10 electrodes for measuring high-frequency ECG were placed according to the standard 12-lead ECG signal positions. After preparation, the test subjects were reminded to remain completely relaxed during the acquisition process. The high-frequency ECG signal in the signal acquisition and processing system software was observed to ensure it was normal. Once the recording conditions were met, the acquisition and recording of high-frequency ECG signals for at least 5 minutes began. If abnormal signals appeared during the acquisition, the acquisition time needed to be appropriately extended to ensure a valid signal duration of 5 minutes.

[0047] To maintain consistency and correspondence, high-frequency electrocardiogram (ECG) and SPECT / CT signals were collected from patients with different degrees of myocardial ischemia, and the obtained data were divided into a test set and a training set.

[0048] Step 2: Signal preprocessing, locating Q, R, and S waves, and extracting the QRS band.

[0049] Specifically, the original signal is filtered to remove 50Hz power frequency interference using a band-notch filter and to remove baseline drift using a zero-phase-shift filter; the Pan-Tompkins algorithm is used to locate the Q, R, and S waves of the noise-removed data; and the QRS band signal is extracted.

[0050] Step 3: Remove abnormal QRS bands and generate 12 average high-frequency QRSs after averaging the signal.

[0051] Specifically, QRS bands with obvious abnormalities were removed, and then the QRS bands of each lead were averaged to obtain the average high-frequency QRS signal of the 12 leads.

[0052] Step 4: Extract feature combinations that reflect functional myocardial ischemia and have clinical and physiological significance;

[0053] Specifically, feature combinations that reflect functional myocardial ischemia and have clinical physiological significance are extracted from the average high-frequency QRS obtained in step 3.

[0054] The meanings of each indicator are as follows:

[0055] (4.1) High-frequency QRS raw waveform index: variation index

[0056]

[0057] Where x represents the original high-frequency QRS waveform of 150–250 Hz. and These represent its mean and standard deviation, respectively. For example... Figure 2 As shown, the high-frequency QRS waveforms of patients with myocardial ischemia are more fragmented than those of normal individuals.

[0058] (4.2) High-frequency QRS raw waveform parameters: bandwidth energy ratio

[0059]

[0060] in This represents the power spectral density of the signal after bandpass filtering. This metric is used to measure the energy ratio fluctuation between the high-frequency components of the QRS complex (150–250 Hz) and the baseline ECG signal components (5–40 Hz).

[0061] (4.3) High-frequency QRS envelope waveform parameters: number of RAZs in the amplitude reduction region, duration

[0062] The high-frequency QRS envelope waveform refers to the envelope of the high-frequency QRS in the range of 150–250 Hz, which is obtained using the Hilbert-Huang transform in this embodiment; the reduced amplitude zone (RAZ) is as follows: Figure 3 As shown, this represents the region of amplitude reduction caused by signal vibration due to myocardial ischemia. RAZ is composed of adjacent peaks. , Valley value For a joint definition to be valid, the following conditions must be met simultaneously:

[0063]

[0064]

[0065] in, , These represent the times corresponding to the two maxima of RAZ.

[0066] (a) The number of regions with decreased amplitude is

[0067]

[0068] N represents the number of RAZs that meet the above conditions.

[0069] (b) The duration of the amplitude reduction region is

[0070]

[0071] in, , These represent the start and end times of the k-th RAZ.

[0072] effect: Describe the frequency of occurrence. This reflects the time history of the low amplitude region; together, they quantify the ischemia-related "functional window period".

[0073] (4.4) High-frequency QRS envelope waveform index: envelope complexity

[0074] The complexity of the envelope is represented using multi-scale entropy.

[0075] (4.4.1) Coarsening ( (for scale)

[0076]

[0077] in, This represents the amplitude of the envelope signal at the k-th sampling point on the high-frequency QRS complex.

[0078] (4.4.2) Sample entropy

[0079]

[0080] Where m refers to the embedding dimension, which is usually 2; r refers to the tolerance threshold, which is usually 0.15 to 0.25 times the standard deviation; A and B are the number of matching pairs of length m+1 and m, respectively.

[0081] (4.4.3) Multiscale entropy

[0082]

[0083] Step 5: Map the leads to the apex, anterior wall, septum, inferior wall, and lateral wall of the heart and aggregate regional features.

[0084] Specifically, such as Figure 4 As shown, SPECT / CT divides the whole heart into 17 segments and 5 regions. Therefore, let the set of the five regions be... Establish a mapping matrix from leads to regions. :

[0085] like Figure 5 As shown, the fixed mappings are: V1 / V2 → septum, V3 / V4 → anterior wall, I / aVL → lateral wall, II / III / aVF → inferior wall; V5 / V6 → apex. The regional feature vector is obtained by averaging the multi-lead indices mapped to each region r. .

[0086] Step 6: Obtain the regional score using a linear combination. .

[0087] Specifically, such as Figure 4 As shown, all 17 segments are scored from 0 to 4. The ratio of the total score for each region to the total score represents the percentage of ischemia in that region. Therefore, for each region...

[0088]

[0089] Among them, coefficient The above feature indicators can be learned by using L2-regularized non-negative least squares on the training data, combining them with the labeled ischemia percentage of SPECT / CT scans using a constrained regression method.

[0090] Step 7, This monotonicity is defined as the percentage of regional ischemia. ;

[0091] Specifically, in order to obtain the ischemia percentage consistent with the gold standard, isotonic regression was used to compare the results. Monotonically increasing calibration is performed. This invention constructs a monotonically increasing calibration function for each region. :

[0092]

[0093] The implementation employs the Pooled Adjacent Violation (PAV) algorithm.

[0094] When making inferences:

[0095]

[0096] Extrapolation strategy valley positioning boundary constant extension.

[0097] Step 8: Output the percentage of total cardiac ischemia using an equal-weighted average. With ischemic sites.

[0098] Specifically, output the percentage of functional ischemia in the whole heart:

[0099]

[0100] Based on the ischemic percentage of the region obtained in step 7, determine the ischemic location and output it.

[0101] This embodiment applies 12-lead high-frequency electrocardiogram signals to the screening or clinical auxiliary diagnosis of functional myocardial ischemia. It can achieve accurate ischemia percentage and localization of ischemia in five zones. It is non-invasive and precise, and helps to greatly reduce the incidence of malignant heart disease.

[0102] Example 2

[0103] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the following steps:

[0104] Collect high-frequency electrocardiogram signals from the subjects;

[0105] Signal preprocessing, locating Q, R, and S waves, and extracting the QRS band;

[0106] Abnormal QRS bands are removed, and the signal is averaged to generate 12 average high-frequency QRS bands;

[0107] Extract characteristic combinations that reflect functional myocardial ischemia and have clinical physiological significance;

[0108] Map leads to the apex, anterior wall, septum, inferior wall, and lateral wall, and aggregate regional features;

[0109] Regional scores are obtained through linear combination. ;

[0110] Will Monotonicity is defined as the percentage of regional ischemia. ;

[0111] Output the percentage of total cardiac ischemia using equal weighted average. With ischemic sites.

[0112] The steps implemented by the processor in this embodiment when executing the program are the same as the specific implementation process of each module of the non-invasive assessment method for detecting the degree and location of functional myocardial ischemia in Embodiment 1, and will not be repeated here.

[0113] Example 3

[0114] This embodiment provides a non-invasive detection system for assessing the degree and location of functional myocardial ischemia, comprising a computer device and an external input device that provides high-frequency electrocardiogram (ECG) signal data of the test subject to external devices of the computer device. The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps:

[0115] Collect high-frequency electrocardiogram signals from the subjects;

[0116] Signal preprocessing, locating Q, R, and S waves, and extracting the QRS band;

[0117] Abnormal QRS bands are removed, and the signal is averaged to generate 12 average high-frequency QRS bands;

[0118] Extract characteristic combinations that reflect functional myocardial ischemia and have clinical physiological significance;

[0119] Map leads to the apex, anterior wall, septum, inferior wall, and lateral wall, and aggregate regional features;

[0120] Regional scores are obtained through linear combination. ;

[0121] Will Monotonicity is defined as the percentage of regional ischemia. ;

[0122] Output the percentage of total cardiac ischemia using equal weighted average. With ischemic sites.

[0123] The steps implemented by the processor in this embodiment when executing the program are the same as the specific implementation process of each module of the non-invasive assessment method for detecting the degree and location of functional myocardial ischemia in Embodiment 1, and will not be repeated here.

[0124] The steps implemented by the processor in this embodiment when executing the program are the same as the specific implementation process of each module of the non-invasive assessment method for detecting the degree and location of functional myocardial ischemia in Embodiment 1, and will not be repeated here.

[0125] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0126] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0127] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0128] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0129] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0130] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A non-invasive method for assessing the degree and location of functional myocardial ischemia, characterized in that, Includes the following steps: Step 1: Acquire 12-lead high-frequency electrocardiogram (ECG) signal data of the test subject, with a sampling frequency >2000Hz and a longitudinal resolution of the signal amplitude reaching the microvolt level. Step 2: Preprocess the signal to remove noise from baseline drift and power line interference, locate the Q, R, and S wave groups in the high-frequency ECG signal, and extract the QRS band signal. Step 3: Remove abnormal QRS bands, average the signals of the QRS bands that meet the conditions by lead, and finally generate 12 average high-frequency QRSs. Step 4: Extract feature combinations that reflect functional myocardial ischemia and have clinical physiological significance from the above average high-frequency QRS; Step 5: Map the leads to the apex, anterior wall, septum, inferior wall, and lateral wall of the heart and aggregate regional features; Step 6: Obtain the regional score using a linear combination. ; Step 7, use isotonic regression to... Monotonicity is defined as the percentage of regional ischemia. ; Step 8: Output the percentage of total cardiac ischemia using an equal-weighted average. With ischemic sites.

2. The non-invasive method for assessing the degree and location of functional myocardial ischemia according to claim 1, characterized in that, The functional myocardial ischemia indicators extracted in step 4 include, but are not limited to, the coefficient of variation of the original high-frequency QRS signal. Frequency band energy ratio The number of regions where the amplitude of the envelope decreases Duration Complexity .

3. The non-invasive method for assessing the degree and location of functional myocardial ischemia according to claim 2, characterized in that, In step 4, the threshold for determining the amplitude reduction region is the amplitude ratio of adjacent peak values. 0.3 and peak spacing 10ms.

4. The non-invasive method for assessing the degree and location of functional myocardial ischemia according to claim 1, characterized in that, In step 5, the fixed mappings are: V1 / V2 → septum, V3 / V4 → anterior wall, I / aVL → lateral wall, II / III / aVF → inferior wall; V5 / V6 → apex of the heart. The regional feature vector is obtained by averaging the multi-lead indices mapped to each region r. .

5. The non-invasive method for assessing the degree and location of functional myocardial ischemia according to claim 1, characterized in that, The coefficient of the regional score in step 6 Solve and fix the non-negative least squares with L2 regularization on the training data.

6. The non-invasive method for assessing the degree and location of functional myocardial ischemia according to claim 1, characterized in that, Monotonic calibration in step 7 The adjacent violator merging algorithm (PAV) is used, and boundary constant extension extrapolation is employed during the inference phase.

7. A system for implementing the method of any one of claims 1 to 6, comprising: The system comprises an acquisition module, a preprocessing module, a signal averaging module, a feature calculation module, a region aggregation module, a scoring module, a monotonic calibration module, and a result output module. The acquisition module acquires high-frequency electrocardiogram (ECG) signals from the subjects. The preprocessing module removes noise such as baseline drift and power line interference, locates Q, R, and S groups, and extracts the QRS complex signal. The signal averaging module averages eligible QRS complexes by lead. The feature calculation module extracts clinically significant feature combinations reflecting functional myocardial ischemia, including... , The region aggregation module aggregates features from 12 leads into five regions: apex, anterior wall, septal wall, inferior wall, and lateral wall, and uses averaging to obtain the feature vector for each region. The scoring module uses linear combinations of the region feature vectors to obtain the region score. The monotonic calibration module is used to apply isotonic regression to... The calibrated percentage ischemia area ischemia is used to output the percentage of whole-heart ischemia as an equal-weighted average. With ischemic sites.

8. A computer device comprising a processor and a memory, wherein the memory stores a program, and the processor executes the program to implement the method of any one of claims 1 to 6.

9. A computer-readable storage medium having a program stored thereon, the program being executed by a processor to implement the method of any one of claims 1 to 6.

Images