A rapid alarm method for electrocardiogram monitor

Abstract

The present invention discloses a rapid alarm method for an electrocardiogram monitor. After collecting an electrocardiogram signal, the time domain characteristics, amplitude characteristics and waveform characteristics in the electrocardiogram signal graph are obtained, a waveform state display model is constructed, and after the waveform state display value is output, a waveform state display curve within a rated time is obtained, and characteristic analysis is performed on the curve, a waveform state alarm model is constructed, and the waveform state alarm value is obtained. The set threshold is compared to complete the electrocardiogram signal alarm. Compared with the traditional electrocardiogram alarm method, the data processing method in the present invention has stronger applicability, and the accuracy of the obtained alarm is better. The applicability model can be adjusted according to the conditions of different patients, which solves the problem that the existing electrocardiogram monitor alarm system cannot adapt to the personalized needs of different patients and different conditions, and the alarm accuracy needs to be improved.

Description

Technical Field

[0001] The invention relates to the technical field of medical monitoring equipment information processing, and in particular to a rapid alarm method for an electrocardiogram monitor. Background Art

[0002] An ECG monitor is a device used in the medical field to monitor a patient's ECG signals. It can display ECG waveforms in real time and issue an alarm in abnormal situations.

[0003] However, the alarm systems of existing ECG monitors are usually based on fixed thresholds and simple rules, which cannot adapt to the personalized needs of different patients and different situations, and the alarm accuracy needs to be improved. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above problems existing in the existing ECG monitor alarm system, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is to solve the problem that the existing ECG monitor alarm system cannot adapt to the personalized needs of different patients and different situations, and the alarm accuracy needs to be improved.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a rapid alarm method for an ECG monitor, comprising the following steps: S1: ECG signal acquisition: the ECG signal of the patient is acquired through the ECG monitor to form an ECG signal graph; S2: the time domain characteristics, amplitude characteristics and waveform characteristics in the ECG signal graph are acquired; S3: based on the characteristic data acquired in S2, a waveform state display model is constructed, each characteristic value is input, and a waveform state display value is output; S4: the ECG signal graph is acquired at uniform intervals within a rated time, and a waveform state display curve within the rated time is acquired; S5: feature extraction is performed on the waveform state display curve, a waveform state alarm model is constructed, the waveform state alarm value is acquired, and the set threshold is compared to complete the ECG signal alarm.

[0008] As a preferred solution of the rapid alarm method of the ECG monitor described in the present invention, wherein: after collecting the ECG signal in S1, it also includes preprocessing the collected ECG signal; wherein the preprocessing steps specifically include: denoising, normalization, cutting and segmentation, histogram equalization and feature extraction.

[0009] As a preferred solution of the rapid alarm method of the ECG monitor described in the present invention, the time domain characteristics obtained in S2 specifically include: RR interval, QRS complex wave width, P wave width, T wave width and QT interval; the amplitude characteristics obtained in S2 specifically include: R wave amplitude, P wave amplitude and T wave amplitude; the waveform characteristics obtained in S2 specifically include: waveform slope, QRS complex wave area, P wave area and T wave area.

[0010] As a preferred solution of the rapid alarm method of the electrocardiogram monitor of the present invention, the waveform state display model constructed is specifically:

[0011]

[0012] Among them, δ is the waveform status display value, a is the RR interval, c is the QRS complex wave width, f is the QT interval, d is the P wave width, g is the T wave width, b is the R wave amplitude, e is the P wave amplitude, h is the T wave amplitude, k is the waveform slope, S QRS is the area of ​​the QRS complex, S P is the area of ​​the P wave, S T is the area of ​​the T wave, 1.354, 3.91, -1.407, 2.6, -1.39, 2.573 and -1.43 are adjustment constants, and dx is the integral operation.

[0013] As a preferred solution of the rapid alarm method of the electrocardiogram monitor described in the present invention, obtaining the waveform state display curve within the rated time specifically includes the following steps: H1: obtaining each waveform state display value within the rated time; H2: in a two-dimensional coordinate system, with the time sequence as the horizontal coordinate and the waveform state display value as the vertical coordinate, obtaining each reference point; H3: connecting each reference point in sequence with a smooth curve to form the waveform state display curve.

[0014] As a preferred solution of the rapid alarm method of the electrocardiogram monitor of the present invention, the features extracted from the waveform state display curve specifically include: the maximum slope of the curve, the minimum slope of the curve, the maximum value of the curve, the minimum value of the curve, the coordinate value of the maximum slope of the curve, and the coordinate value of the minimum slope of the curve;

[0015] The waveform state alarm model constructed is specifically:

[0016]

[0017] Among them, η is the waveform state alarm value, K min is the minimum slope of the curve, K max is the maximum slope of the curve, Y max is the maximum value of the curve, Y minis the minimum value of the curve, Y Kmin Y is the coordinate value of the reference point at the minimum slope of the curve, Kmax is the coordinate value of the reference at the maximum slope of the curve, 1.01 and -0.58 are adjustment constants, and dx is the integral operation.

[0018] As a preferred solution of the rapid alarm method of the ECG monitor described in the present invention, when the waveform state alarm value is higher than the threshold value, it is defined that the current ECG signal is abnormal and an alarm is issued.

[0019] As a preferred solution of the rapid alarm method of the electrocardiogram monitor described in the present invention, the threshold value is optimally set at 0.86 or 0.864.

[0020] Beneficial effects of the present invention: The present invention provides a rapid alarm method for an ECG monitor, which acquires the time domain features, amplitude features and waveform features in the ECG signal graph after collecting the ECG signal, constructs a waveform state display model, outputs the waveform state display value, acquires the waveform state display curve within the rated time, performs feature analysis on the curve, constructs a waveform state alarm model, acquires the waveform state alarm value, compares the set threshold value, and completes the ECG signal alarm. Compared with the traditional ECG alarm method, the data processing method in the present invention is more applicable, the accuracy of the alarm is better, and the applicability model can be adjusted according to the conditions of different patients, which solves the problem that the existing ECG monitor alarm system cannot adapt to the personalized needs of different patients and different conditions, and the alarm accuracy needs to be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0022] Figure 1 The present invention provides an overall method flow chart of the rapid alarm method of the electrocardiogram monitor.

[0023] Figure 2 A flow chart of the method for obtaining the waveform state display curve within a rated time provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0025] The alarm systems of existing ECG monitors are usually based on fixed thresholds and simple rules, which cannot adapt to the personalized needs of different patients and different situations, and the alarm accuracy needs to be improved.

[0026] Therefore, please refer to Figure 1 The present invention provides a rapid alarm method for an electrocardiogram monitor, comprising the following steps:

[0027] S1: ECG signal acquisition: The ECG signal of the patient is collected through an ECG monitor to form an ECG signal graph;

[0028] S2: Obtaining time domain features, amplitude features and waveform features in the ECG signal graph;

[0029] S3: Based on the characteristic data obtained in S2, a waveform state display model is constructed, each characteristic value is input, and a waveform state display value is output;

[0030] S4: collecting ECG signal graphs at even intervals within a rated time, and obtaining a waveform state display curve within the rated time;

[0031] S5: Extract features from the waveform state display curve, build a waveform state alarm model, obtain the waveform state alarm value, compare it with the set threshold, and complete the ECG signal alarm.

[0032] Specifically, after collecting the ECG signal in S1, the collected ECG signal is also preprocessed;

[0033] The preprocessing steps specifically include: denoising, normalization, cutting and segmentation, histogram equalization and feature extraction.

[0034] It should be noted that the preprocessing steps are as follows:

[0035] 1. Denoising:

[0036] Filtering: Use various filters (such as low-pass filtering, high-pass filtering, and band-pass filtering) to remove high-frequency noise and low-frequency interference in ECG signals.

[0037] Median filtering: particularly suitable for removing salt and pepper noise.

[0038] Adaptive filtering: Able to dynamically adjust filtering parameters based on signal characteristics.

[0039] Specific:

[0040] Low pass filter: H low (f) = 1 / (1+jf / f c );

[0041] High pass filter: H high (f) = (jf / f c ) / (1+jf / f c );

[0042] Bandpass filtering: combines low-pass and high-pass filtering to allow only signals within a specific frequency range to pass;

[0043] 2. Normalization:

[0044] The amplitude of the ECG signal is normalized to a certain fixed range to eliminate the differences between different individuals.

[0045] Specific:

[0046] Linear normalization: x norm =(xx min ) / (x max -x min ), where x min and x max are the minimum and maximum values ​​of the signal respectively;

[0047] 3. Cutting and segmentation:

[0048] The ECG signal is cut into single heart cycles to facilitate subsequent feature extraction and analysis.

[0049] Usually, the starting point of the cardiac cycle is determined based on R wave detection, and then the signal is cut according to the RR interval;

[0050] 4. Histogram equalization:

[0051] By adjusting the histogram distribution of the ECG signal, the contrast of the signal is enhanced to make the features more prominent, such as using the cumulative distribution function (CDF) to adjust the histogram;

[0052] 5. Feature extraction:

[0053] Extract key features from the preprocessed ECG signal, such as R wave position, QRS complex wave width, T wave width, etc.

[0054] R wave detection: detects the position of the R wave in the ECG signal.

[0055] QRS width calculation: The width of the QRS complex is calculated, usually from the beginning of the R wave to the lowest point of the S wave.

[0056] T wave detection: detect the starting point and end point of the T wave and calculate the width of the T wave.

[0057] Furthermore, the time domain features obtained in S2 specifically include: RR interval, QRS complex wave width, P wave width, T wave width and QT interval;

[0058] The amplitude characteristics obtained in S2 specifically include: R wave amplitude, P wave amplitude and T wave amplitude;

[0059] The waveform features obtained in S2 specifically include: waveform slope, area of ​​QRS complex wave, area of ​​P wave and area of ​​T wave.

[0060] Specific:

[0061] Time domain characteristics:

[0062] RR interval: The time interval between two consecutive R waves, reflecting the rhythm of the heartbeat.

[0063] QRS complex width: The time interval from the beginning of the Q wave to the end of the S wave, reflecting the speed of ventricular depolarization.

[0064] P wave width: The duration of the P wave, which reflects the speed of atrial depolarization.

[0065] T wave width: The duration of the T wave, which reflects the speed of ventricular repolarization.

[0066] QT interval: The time interval from the beginning of the QRS complex to the end of the T wave, which is the total time for ventricular depolarization and repolarization.

[0067] Amplitude characteristics:

[0068] R wave amplitude: The maximum amplitude of the R wave.

[0069] P wave amplitude: The maximum amplitude of the P wave.

[0070] T wave amplitude: the maximum amplitude of the T wave.

[0071] Waveform characteristics:

[0072] Waveform slope: The slope of the waveform rising or falling can reflect the dynamic changes of the ECG signal.

[0073] Waveform area: The area of ​​the QRS complex wave, P wave and T wave can reflect the energy of the ECG signal.

[0074] Furthermore, the waveform state display model constructed is specifically as follows:

[0075]

[0076] Among them, δ is the waveform status display value, a is the RR interval, c is the QRS complex wave width, f is the QT interval, d is the P wave width, g is the T wave width, b is the R wave amplitude, e is the P wave amplitude, h is the T wave amplitude, k is the waveform slope, S QRS is the area of ​​the QRS complex, S P is the area of ​​the P wave, S T is the area of ​​the T wave, 1.354, 3.91, -1.407, 2.6, -1.39, 2.573 and -1.43 are adjustment constants, and dx is the integral operation.

[0077] For further information, see Figure 2 , obtaining the waveform status display curve within the rated time specifically includes the following steps:

[0078] H1: Get the waveform status display value of each waveform within the rated time;

[0079] H2: In the two-dimensional coordinate system, the time sequence is used as the horizontal coordinate and the waveform state display value is used as the vertical coordinate to obtain each reference point;

[0080] H3: Connect each reference point in sequence with a smooth curve to form a waveform status display curve.

[0081] Specifically, the features extracted from the waveform state display curve include: the maximum slope of the curve, the minimum slope of the curve, the maximum value of the curve, the minimum value of the curve, the coordinate value of the maximum slope of the curve, and the coordinate value of the minimum slope of the curve;

[0082] The waveform state alarm model constructed is specifically as follows:

[0083]

[0084] Among them, η is the waveform state alarm value, K min is the minimum slope of the curve, K max is the maximum slope of the curve, Y max is the maximum value of the curve, Y min is the minimum value of the curve, Y Kmin Y is the coordinate value of the reference point at the minimum slope of the curve, Kmax is the coordinate value of the reference at the maximum slope of the curve, 1.01 and -0.58 are adjustment constants, and dx is the integral operation.

[0085] Specifically, when the waveform state alarm value is higher than the threshold, it is defined that the current ECG signal is abnormal and an alarm is issued.

[0086] Furthermore, the threshold is optimally set at 0.86 or 0.864.

[0087] In order to verify the technical effect of the present invention, the following simulation test is now carried out:

[0088] 1. Experimental Preparation

[0089] Select a certain number of ECG monitors as test equipment.

[0090] Collect real ECG signal data from patients of different ages and health conditions.

[0091] Determine the standards for normal and abnormal ECG signals for subsequent data comparison.

[0092] 2. Data Collection

[0093] Use an ECG monitor to collect the ECG signals of each patient in a quiet state to ensure the accuracy and consistency of the data.

[0094] Record each patient's personal information, ECG signal characteristics, and known health conditions.

[0095] 3. Data processing

[0096] According to the method of the present invention, features are extracted from the collected electrocardiographic signals, including time domain features, amplitude features and waveform features.

[0097] Use the constructed waveform state display model to output the waveform state display value.

[0098] Get the waveform status display curve within a certain period of time and perform feature analysis.

[0099] 4. Alarm model construction

[0100] According to the characteristic analysis results, a waveform state alarm model is constructed and the waveform state alarm value is output.

[0101] Set appropriate alarm thresholds.

[0102] 5. Data Analysis

[0103] Compare the alarm value output by the alarm model with the actual health status to evaluate the alarm accuracy.

[0104] Analyze false positives and false negatives.

[0105] 6. Recording of results

[0106] Record test results, including alarm accuracy under normal and abnormal conditions.

[0107] Data Validation Form:

[0108] Patient ID age gender Health status ECG signal characteristic value Alarm threshold Actual situation Alarm model output value Alarm results False positives / false negatives Remark 1 45 male normal RR interval: 800ms, QRS width: 90ms, ... 0.86 normal 0.376 No alarm No false negatives / 2 30 female Coronary heart disease RR interval: 600ms, QRS width: 110ms, ... 0.86 abnormal 0.910 Call the police No false negatives / 3 55 female Arrhythmias RR interval: 620ms, QRS width: 100ms, ... 0.86 abnormal 1.65 Call the police No false negatives / 4 70 male hypertension RR interval: 720ms, QRS width: 95ms, ... 0.86 abnormal 1.033 Call the police No false negatives /

[0109] Among them, RR interval (ms), QRS width (ms), P wave width (ms), T wave width (ms), QT interval (ms), R wave amplitude (mV), P wave amplitude (mV), T wave amplitude (mV), waveform slope, QRS complex wave area (mV·ms), P wave area (mV·ms), T wave area (mV·ms).

[0110] Notes: This column can include any additional observations or special notes, such as signal quality, changes in patient position, effects of medications, etc.

[0111] illustrate:

[0112] The “ECG signal characteristic value” column in the table contains the key characteristic data of the ECG signal of each patient.

[0113] The "alarm model output value" is the alarm value processed by the method of the present invention, and is used to compare with the "alarm threshold" to determine whether to trigger an alarm.

[0114] The “Alarm Result” column indicates whether the ECG monitor correctly triggered the alarm based on the model output value.

[0115] The “False Positive / False Negative” column records the false positives or false negatives that occurred in the experiment, which is crucial for evaluating the accuracy of the system.

[0116] The present invention provides a rapid alarm method for an electrocardiogram monitor. After collecting an electrocardiogram signal, the time domain characteristics, amplitude characteristics and waveform characteristics in the electrocardiogram signal graph are obtained, a waveform state display model is constructed, and after the waveform state display value is output, a waveform state display curve within a rated time is obtained, and characteristic analysis is performed on the curve, a waveform state alarm model is constructed, and a waveform state alarm value is obtained. The set threshold is compared to complete the electrocardiogram signal alarm. Compared with the traditional electrocardiogram alarm method, the data processing method in the present invention has stronger applicability, and the accuracy of the obtained alarm is better. The applicability model can be adjusted according to the conditions of different patients, which solves the problem that the existing electrocardiogram monitor alarm system cannot adapt to the personalized needs of different patients and different conditions, and the alarm accuracy needs to be improved.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A rapid alarm method for an electrocardiogram monitor, characterized in that: The steps include: S1: ECG signal acquisition: The ECG signal of the patient is collected through an ECG monitor to form an ECG signal graph; S2: Obtaining time domain features, amplitude features and waveform features in the ECG signal graph; S3: Based on the characteristic data obtained in S2, a waveform state display model is constructed, each characteristic value is input, and a waveform state display value is output; S4: collecting ECG signal graphs at even intervals within a rated time, and obtaining a waveform state display curve within the rated time; S5: extracting features from the waveform state display curve, constructing a waveform state alarm model, obtaining a waveform state alarm value, comparing it with a set threshold, and completing an ECG signal alarm; Among them, the time domain features obtained in S2 specifically include: RR interval, QRS complex wave width, P wave width, T wave width and QT interval; The amplitude characteristics obtained in S2 specifically include: R wave amplitude, P wave amplitude and T wave amplitude; The waveform features obtained in S2 specifically include: waveform slope, area of ​​QRS complex wave, area of ​​P wave and area of ​​T wave; The waveform state display model constructed is specifically: Among them, δ is the waveform status display value, a is the RR interval, c is the QRS complex wave width, f is the QT interval, d is the P wave width, g is the T wave width, b is the R wave amplitude, e is the P wave amplitude, h is the T wave amplitude, k is the waveform slope, S QRS is the area of ​​the QRS complex, S P is the area of ​​the P wave, S T is the area of ​​the T wave, 1.354, 3.91, -1.407, 2.6, -1.39, 2.573 and -1.43 are adjustment constants, and dx is the integral operation; The features extracted from the waveform state display curve specifically include: the maximum slope of the curve, the minimum slope of the curve, the maximum value of the curve, the minimum value of the curve, the coordinate value of the maximum slope of the curve, and the coordinate value of the minimum slope of the curve; The waveform state alarm model constructed is specifically: Among them, η is the waveform state alarm value, K min is the minimum slope of the curve, K max is the maximum slope of the curve, Y max is the maximum value of the curve, Y min is the minimum value of the curve, Y Kmin is the coordinate value of the reference at the minimum slope of the curve, Y Kmax is the coordinate value of the reference at the maximum slope of the curve, 1.01 and -0.58 are adjustment constants, and dx is the integral operation.

2. The rapid alarm method of the electrocardiogram monitor according to claim 1, characterized in that: After collecting the ECG signal in S1, it also includes preprocessing the collected ECG signal; The preprocessing steps specifically include: denoising, normalization, cutting and segmentation, histogram equalization and feature extraction.

3. The rapid alarm method of the electrocardiogram monitor according to claim 2, characterized in that: Obtaining the waveform state display curve within the rated time specifically includes the following steps: H1: Get the waveform status display value of each waveform within the rated time; H2: In a two-dimensional coordinate system, the time sequence is used as the horizontal coordinate and the waveform state display value is used as the vertical coordinate to obtain each reference point; H3: Connect each reference point in sequence with a smooth curve to form the waveform state display curve.

4. The rapid alarm method of the electrocardiogram monitor according to claim 3, characterized in that: When the waveform state alarm value is higher than the threshold value, it is defined that the current ECG signal is abnormal and an alarm is issued.

5. The rapid alarm method of the electrocardiogram monitor according to claim 4, characterized in that: The threshold was set at 0.86 or 0.864.

Images