Arrhythmia detection method, nonvolatile readable storage medium and wearable device

Abstract

The invention relates to the technical field of electrocardiosignal detection, and discloses an arrhythmia detection method, a nonvolatile readable storage medium and wearable equipment. The method comprises the steps that lead electrocardiosignals and accelerated speed collected by the wearable device are obtained, a cardiac beat area is determined according to the lead electrocardiosignals, a target forgetting factor of the accelerated speed is determined according to cardiac beat time of the cardiac beat area, self-adaptive filtering processing is conducted on the lead electrocardiosignals according to the target forgetting factor and the accelerated speed, and first lead electrocardiosignals are obtained; and performing band-pass filtering processing on the first lead electrocardiosignal to obtain a second lead electrocardiosignal, and generating a heart rhythm detection result according to the second lead electrocardiosignal. The embodiment does not depend on the given current moment to match the forgetting factor for the acceleration, but determines the target forgetting factor of the acceleration according to the cardiac beat time, so that the signal quality and effect of adaptive filtering can be improved, and the detection precision of arrhythmia can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrocardiogram signal detection, and in particular to an arrhythmia detection method, a non-volatile readable storage medium and a wearable device. Background Art

[0002] The ECG signal reflects the electrophysiological process of cardiac activity. Clinically, it is believed that real-time monitoring of the ECG signal can be used to detect arrhythmias and provide emergency treatment in a timely manner. Figure 1 A typical ECG waveform is shown in Figure 1. A normal ECG signal generally consists of a P wave, a QRS complex wave, and a T wave.

[0003] The wearable device provided by the related technology can process the ECG signal in combination with the arrhythmia detection algorithm, so as to detect whether the user is in an arrhythmia state. In the process of processing the ECG signal, the wearable device filters the ECG signal according to the adaptive filtering algorithm. Usually, the adaptive filtering algorithm adopts the recursive least squares algorithm (RLS). The recursive least squares algorithm introduces a forgetting factor. The forgetting factor is used to configure a larger forgetting factor for the error closer to the current moment, and a smaller forgetting factor for the error farther from the current moment. This can improve the convergence speed and achieve the purpose of fast adaptive filtering. However, the ECG signal analysis algorithm of the wearable device is usually performed regularly, and the time of the P wave, QRS complex wave or T wave of the ECG signal may not match the current moment of the forgetting factor. This will cause the forgetting factor corresponding to the P wave, QRS complex wave or T wave to be relatively small, thereby reducing the signal quality and effect of the adaptive filtering, and further reducing the detection accuracy of arrhythmia. Summary of the invention

[0004] An object of an embodiment of the present invention is to provide an arrhythmia detection method, a non-volatile readable storage medium and a wearable device, aiming to solve the technical problem of low detection accuracy in arrhythmia detection in related technologies.

[0005] In a first aspect, an embodiment of the present invention provides an arrhythmia detection method, which is applied to a wearable device, comprising:

[0006] Obtaining lead ECG signals and acceleration collected by the wearable device;

[0007] determining a heartbeat region according to the lead ECG signal;

[0008] Determining a target forgetting factor of the acceleration according to the heartbeat time of the heartbeat area;

[0009] Performing adaptive filtering on the lead ECG signal according to the target forgetting factor and the acceleration to obtain a first lead ECG signal;

[0010] Performing bandpass filtering on the first lead ECG signal to obtain a second lead ECG signal;

[0011] Generate a heart rhythm detection result according to the second-lead ECG signal.

[0012] Optionally, determining the target forgetting factor of the acceleration according to the heartbeat time of the heartbeat area includes:

[0013] determining a signal time of the acceleration;

[0014] The target forgetting factor of the acceleration is determined according to the relative position of the signal time of the acceleration and the heartbeat time of the heartbeat area.

[0015] Optionally, determining the target forgetting factor of the acceleration according to the relative position of the signal time of the acceleration and the heartbeat time of the heartbeat area includes:

[0016] When the signal time of the acceleration is within the heartbeat time of the heartbeat area, determining the target forgetting factor of the acceleration to be a first forgetting factor;

[0017] When the signal time of the acceleration is outside the heartbeat time of the heartbeat area, it is determined that the target forgetting factor of the acceleration is a second forgetting factor, and the second forgetting factor is smaller than the first forgetting factor.

[0018] Optionally, the heartbeat time includes a P wave interval, a QRS interval, and a T wave interval, the first forgetting factor includes a first factor, a second factor, and a third factor, the second factor is greater than the first factor and the third factor, and when the signal time of the acceleration is within the heartbeat time of the heartbeat area, determining that the target forgetting factor of the acceleration is the first forgetting factor includes:

[0019] When the signal time of the acceleration is within the P wave interval of the heartbeat region, determining the target forgetting factor of the acceleration to be the first factor;

[0020] When the signal time of the acceleration is within the QRS interval of the heartbeat region, determining the target forgetting factor of the acceleration to be a second factor;

[0021] When the signal time of the acceleration is within the T wave interval of the heartbeat region, the target forgetting factor of the acceleration is determined to be the third factor.

[0022] Optionally, when the signal time of the acceleration is in the P wave interval of the heartbeat area, the closer the signal time of the acceleration is to the QRS interval, the larger the first factor corresponding to the acceleration is, and the farther the signal time of the acceleration is from the QRS interval, the smaller the first factor corresponding to the acceleration is; or,

[0023] When the signal time of the acceleration is in the T wave interval of the heartbeat area, the closer the signal time of the acceleration is to the QRS interval, the larger the third factor corresponding to the acceleration is, and the farther the signal time of the acceleration is from the QRS interval, the smaller the third factor corresponding to the acceleration is.

[0024] Optionally, the heart rhythm detection result includes a ventricular fibrillation detection result. Before generating the heart rhythm detection result, the method further includes:

[0025] Detecting whether a user wearing the wearable device is in a stationary state according to the acceleration;

[0026] If the user is in a stationary state, a ventricular fibrillation detection result is generated according to the second lead electrocardiogram signal.

[0027] Optionally, detecting whether the user wearing the wearable device is in a stationary state according to the acceleration includes:

[0028] Calculating the acceleration value of the acceleration;

[0029] Determining whether the acceleration value is greater than a preset threshold;

[0030] If it is greater than, it is determined that the user wearing the wearable device is in motion;

[0031] If it is less than or equal to, it is determined that the user wearing the wearable device is in a stationary state.

[0032] Optionally, the acceleration is a three-axis acceleration, the three-axis acceleration includes an x-axis acceleration, a y-axis acceleration, and a z-axis acceleration, and the acceleration value of the acceleration is calculated including:

[0033] Calculating the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration respectively;

[0034] Calculating the sum of the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration;

[0035] Taking the square root of the sum, the acceleration values ​​of the three-axis acceleration are obtained.

[0036] In a second aspect, an embodiment of the present invention provides a non-volatile readable storage medium, wherein the non-volatile readable storage medium stores computer executable instructions, and the computer executable instructions are used to enable a controller to execute the above-mentioned arrhythmia detection method.

[0037] In a third aspect, an embodiment of the present invention provides a wearable device, including:

[0038] Heart rate sensor;

[0039] Accelerometer;

[0040] at least one processor, electrically connected to the heart rate sensor and the acceleration sensor respectively; and

[0041] a memory communicatively connected to the at least one processor; wherein,

[0042] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the above-mentioned arrhythmia detection method.

[0043] In the arrhythmia detection method provided in the embodiment of the present invention, the lead ECG signal and acceleration collected by the wearable device are obtained, the heartbeat area is determined according to the lead ECG signal, the target forgetting factor of the acceleration is determined according to the heartbeat time of the heartbeat area, the lead ECG signal is adaptively filtered according to the target forgetting factor and the acceleration to obtain a first lead ECG signal, the first lead ECG signal is bandpass filtered to obtain a second lead ECG signal, and a heart rhythm detection result is generated according to the second lead ECG signal. This embodiment does not rely on the given current moment to match the forgetting factor for acceleration, but fully considers the importance of the heartbeat area and the important influence of the heartbeat area on the heart rhythm detection result, and determines the target forgetting factor of the acceleration according to the heartbeat time, which can improve the signal quality and effect of the adaptive filtering, and thus also improve the detection accuracy of arrhythmia. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0045] Figure 1 A waveform diagram of a typical ECG signal provided for related technologies;

[0046] Figure 2 A waveform diagram of an electrocardiogram signal before the onset of ventricular fibrillation provided for related technologies;

[0047] Figure 3 A waveform diagram of a non-ventricular fibrillation signal under a low signal-to-noise ratio provided for the related art;

[0048] Figure 4 A waveform diagram of an ECG signal when ventricular fibrillation occurs, provided for related technologies;

[0049] Figure 5 A schematic diagram of a circuit structure of a wearable device provided by an embodiment of the present invention;

[0050] Figure 6 A schematic diagram of a flow chart of an arrhythmia detection method provided by an embodiment of the present invention;

[0051] Figure 7a , Figure 7b and Figure 7c are respectively the x-axis acceleration a in the x direction provided by the embodiment of the present invention x , the y-axis acceleration a in the y direction y and the z-axis acceleration a in the z direction z Waveform diagram of

[0052] Figure 8 A schematic diagram of the structure of an arrhythmia detection device provided by an embodiment of the present invention;

[0053] Fig. 9 A schematic diagram of the circuit structure of a controller provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0055] It should be noted that, if there is no conflict, the various features in the embodiments of the present invention can be combined with each other, all within the scope of protection of the present invention. In addition, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a different order from the module division in the device or the flow chart. Furthermore, the words "first", "second", "third", etc. used in the present invention do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0056] Related technologies According to the different sources of ectopic arrhythmia, arrhythmia can be divided into atrial arrhythmia and ventricular arrhythmia. According to the different onset time, arrhythmia can also be divided into non-sustained arrhythmia and sustained arrhythmia. Non-sustained arrhythmia includes ventricular ectopic beats or atrial ectopic beats, etc., and sustained arrhythmia includes ventricular fibrillation or atrial fibrillation, etc.

[0057] Figure 2 The waveform diagram of the electrocardiogram signal before the onset of ventricular fibrillation provided by the related technology, Figure 3 The waveform diagram of non-ventricular fibrillation signal under low signal-to-noise ratio provided by the related technology, Figure 4 The waveform diagram of the ECG signal when ventricular fibrillation occurs is provided for the related technology. Among arrhythmias, ventricular fibrillation (VF) is a persistent arrhythmia that poses a greater threat to life. When ventricular fibrillation occurs, the heart activity is fast and uncoordinated tremor, and the effective contraction ability is lost. It is equivalent to ventricular arrest in hemodynamics, and its ECG waveform is a series of fast tremor waves with varying amplitudes.

[0058] People can monitor arrhythmias by wearing wearable devices. Wearable devices can measure vital signs such as heart rate and monitor arrhythmias during exercise. However, during exercise, the ECG electrodes of wearable devices often slip on the skin, causing the ECG signals collected by the wearable devices to be submerged by irregular noise signals, which is similar to ventricular fibrillation. For example, Figure 3 and Figure 4 The waveform is shown in the figure. Subsequently, the wearable device uses an algorithm to Figure 4 When the ECG signal shown is used for identification, misjudgment is likely to occur.

[0059] Related technologies When detecting arrhythmias, it is usually necessary to perform adaptive filtering on ECG signals. In current adaptive filters, weight is an important parameter. According to the different weights, adaptive filters can be divided into three main types: minimum mean square (LMS), recursive least squares and affine projection (AP). For the minimum projection algorithm, new data is input into the algorithm for analysis sample by sample point, and the contribution weight of the error signal of each sampling point to the objective function is the same. The advantage of the minimum projection algorithm is fast calculation speed, but slow convergence speed. It is only suitable for stable signals with regular changes. Due to the influence of noise during movement, the ECG signal is in a non-stationary state at this time, which will not only further reduce the convergence speed, but also easily lead to non-convergence.

[0060] The algorithm based on recursive least squares method adds a forgetting factor, and the closer the error signal is to the current moment, the greater the contribution, which improves the convergence speed. However, since the ECG signal analysis algorithm is usually performed regularly, once the heartbeat area is far away from the current moment, it will be assigned a lower forgetting factor, which reduces the signal quality and thus affects the analysis accuracy. Based on this, in the process of realizing the present invention, the inventor found that: determining the target forgetting factor of acceleration in combination with the heartbeat time of the heartbeat area is conducive to improving the adaptive filtering effect, and then to improving the signal quality and detection accuracy. In addition, when the user is in a stationary state to monitor ventricular fibrillation, this detection method is more accurate and reliable.

[0061] As one aspect of an embodiment of the present invention, an embodiment of the present invention provides a wearable device, such as Figure 5 As shown, the wearable device 500 includes a controller 51, a heart rate sensor 52, an acceleration sensor 53, a warning circuit 54 and a wireless communication module 55.

[0062] The controller 51 serves as the control core of the wearable device 500 .

[0063] The heart rate sensor 52 is electrically connected to the controller 51 and is used to collect ECG signals. The heart rate sensor 52 includes a plurality of ECG electrodes. When the user wears the wearable device 500, the plurality of ECG electrodes can be in close contact with the user's skin to collect the user's ECG signals. The user can wear the wearable device 500 on the wrist, elbow, leg, chest, head, etc.

[0064] The acceleration sensor 53 is electrically connected to the controller 51 and is used to collect the acceleration of the user. The acceleration sensor 53 can be a three-axis gyroscope or an inertial detection unit.

[0065] The warning prompt circuit 54 is electrically connected to the controller 51 and is used to generate warning information. The warning prompt circuit 54 can be a voice prompt circuit, a display screen, or an LED light circuit. The voice prompt circuit is used to generate voice warning information, the display screen is used to generate text warning information, and the LED light circuit is used to generate light warning information.

[0066] The wireless communication module 55 is electrically connected to the controller 51, and is used to transmit the heart rhythm detection report transmitted by the controller 51 to a designated terminal. For example, the controller 51 generates a heart rhythm detection report based on the electrocardiogram signal and acceleration, and transmits the heart rhythm detection report to the designated terminal through the wireless communication module 55, wherein the designated terminal may be a designated mobile phone, a designated computer or a designated server, etc. The wireless communication module 55 may be a Bluetooth module, a WIFI module, a 4G module or a 5G module, etc.

[0067] As another aspect of the present invention, the present invention provides a method for detecting arrhythmia, which is applied to a wearable device. Figure 6 , the arrhythmia detection method comprises the following steps:

[0068] S61: Obtain lead ECG signals and acceleration collected by the wearable device.

[0069] In this step, the lead ECG signal is a series of ECG signals collected by the heart rate sensor. In the professional terminology of ECG, the placement of the ECG electrodes on the human body surface and the connection method between the ECG electrodes and the amplifier when recording the ECG are called the leads of the ECG, and a series of ECG signals collected by the leads can be regarded as lead ECG signals. Among them, according to the number of lead channels, the lead ECG signal can be divided into a single-lead ECG signal (i.e., a lead ECG signal collected by one lead) and a multi-lead ECG signal (i.e., a lead ECG signal collected by multiple leads), among which the multi-lead ECG signal can be considered to be composed of multiple single-lead ECG signals, and the more common number of leads of the multi-lead ECG signal is three leads, six leads, twelve leads, and eighteen leads, etc.

[0070] A lead ECG signal refers to a single lead ECG signal, which can be a single lead ECG signal directly collected, or a single lead ECG signal selected from multiple lead ECG signals. Exemplarily, the wearable device can obtain at least one lead ECG signal collected by at least one lead, wherein the number of lead channels, sampling duration and sampling frequency can be set according to actual conditions. When multiple lead ECG signals are obtained, each lead ECG signal is an ECG signal collected at the same time. The lead ECG signal is composed of sample points obtained by sampling, wherein each sample point represents the ECG signal collected at the corresponding sampling moment. For the lead ECG signal, the horizontal axis is the time axis, which is used to record the sampling moment of the sample point in units of time. At this time, the position of the sample point in the lead ECG signal refers to the sampling moment of the sample point, and the vertical axis is the intensity of the ECG signal, which is often characterized by voltage. Optionally, the characteristic waves in the lead ECG signal include at least P wave, QRS wave and T wave, and the number of each characteristic wave is not limited in the embodiment. Among them, an adjacent and continuous P wave, a QRS wave and a T wave constitute a heart beat, and multiple heart beats constitute a lead ECG signal.

[0071] The acceleration is collected by the acceleration sensor. When the acceleration is three-axis acceleration, the three-axis acceleration (a x , a y , a z ) includes the x-axis acceleration a x , y-axis acceleration a y and z-axis acceleration a z Please also refer to Figure 7a , Figure 7b and Figure 7c When a user wears a wearable device to exercise, the acceleration can be decomposed into the x-axis acceleration a in the x direction. x , the y-axis acceleration a in the y direction y and the z-axis acceleration a in the z direction z .

[0072] S62: Determine the heart beat area according to the lead ECG signal.

[0073] In this step, the heartbeat region is the region of the electrocardiographic characteristic wave. In some embodiments, the heartbeat region may be the QRS wave region, that is, the region between the starting point of the QRS wave and the end point of the QRS wave. In some embodiments, the heartbeat region may be a region composed of a P wave region, a QRS wave region, and a T wave region in sequence, that is, the region between the starting point of the P wave and the end point of the T wave.

[0074] Determining the heartbeat region according to the lead ECG signal includes: inputting the lead ECG signal into a preset heartbeat feature model to obtain the heartbeat region. The heartbeat feature model can be obtained by pre-training by the designer.

[0075] S63: Determine a target forgetting factor of acceleration according to the heartbeat time of the heartbeat area.

[0076] In this step, the heartbeat time is the time corresponding to the heartbeat area. In some embodiments, when the heartbeat area is a QRS wave area, the heartbeat time is from the start of the QRS wave to the end of the QRS wave. In some embodiments, when the heartbeat area is an area composed of a P wave area, a QRS wave area, and a T wave area in sequence, the heartbeat time is from the start of the P wave to the end of the T wave.

[0077] The target forgetting factor is a forgetting factor determined by the heartbeat time to the acceleration. In the subsequent adaptive filtering process, this embodiment needs to combine the forgetting factor of the acceleration with the acceleration to perform adaptive filtering on the lead ECG signal.

[0078] S64: Adaptively filter the lead ECG signal according to the target forgetting factor and the acceleration to obtain a first lead ECG signal.

[0079] In this step, the first lead ECG signal is the lead ECG signal processed by adaptive filtering. Let the lead ECG signal of a single channel be y, and the acceleration be (a x , a y , a z ), wherein in each analysis process of this embodiment, the length of the lead ECG signal and acceleration obtained each time is n. If the sampling rate of the lead ECG signal and acceleration is not 250Hz, the signal is resampled to 250Hz before analysis.

[0080] In the adaptive filtering algorithm, the weight of acceleration is set to w, and the dimension of the weight of acceleration is equal to the number of acceleration channels. e is the error signal of adaptive filtering, μ adap is the step size (μ) of adaptive filter weight update adap =0.01), ε is a very small positive number (ε = 1e-3). When processing the i = 1, ..., nth sample points of the lead ECG signal and acceleration, the error, weight, and target optimization function of the adaptive filter are shown in Formula 1, Formula 2, and Formula 3:

[0081] e(i)=y bp (i)-w T (i)a(i) Formula 1

[0082] w(i+1)=w(i)+2μ adap / (ε+a T (i)a(i))a(i)e(i) Formula 2

[0083]

[0084] Where λ is the forgetting factor, 0<λ≤1, J n (w) is the target optimization function. In this embodiment, the target optimization function can be optimized according to the minimum mean square error method to output an adaptive filtering result.

[0085] S65: Perform bandpass filtering on the first lead ECG signal to obtain a second lead ECG signal.

[0086] In this step, the bandpass filtering process is an operation of retaining the first lead ECG signal of the specified frequency band, and the second lead ECG signal is the first lead ECG signal processed by the bandpass filtering process. In this embodiment, a second-order Butterworth filter with a frequency range of 0.5 Hz to 35 Hz is selected as the bandpass filter to perform bandpass filtering on the first lead ECG signal.

[0087] S66: Generate a heart rhythm detection result according to the second lead ECG signal.

[0088] In this step, the heart rhythm detection results include ventricular beat detection results, atrial beat detection results, ventricular fibrillation detection results and atrial fibrillation detection results. The ventricular beat detection results include normal ventricular beat results and ventricular ectopic beat results. The atrial beat detection results include normal atrial beat results and atrial ectopic beat results. The ventricular fibrillation detection results include normal ventricular results and ventricular fibrillation results. The atrial fibrillation detection results include normal atrial results and atrial fibrillation results.

[0089] In general, this embodiment does not rely on the given current moment to match the forgetting factor for acceleration, but fully considers the importance of the heartbeat area and the important influence of the heartbeat area on the heart rhythm detection results, and determines the target forgetting factor of acceleration according to the heartbeat time. This can improve the signal quality and effect of adaptive filtering, and thus also improve the detection accuracy of arrhythmia.

[0090] In some embodiments, determining the target forgetting factor of acceleration according to the heartbeat time of the heartbeat area includes the following steps: determining the signal time of the acceleration, and determining the target forgetting factor of the acceleration according to the relative position of the signal time of the acceleration and the heartbeat time of the heartbeat area. This embodiment can match the corresponding target forgetting factor for the acceleration according to the relative position of the signal time and the heartbeat time, and fully integrate the importance of the heartbeat area and the important influence of the heartbeat area on the heart rhythm detection result into the selection of the forgetting factor, which is conducive to improving the adaptive filtering effect, thereby helping to improve the detection accuracy.

[0091] In some embodiments, determining a target forgetting factor of acceleration based on the relative position of the signal time of acceleration and the heartbeat time of a heartbeat area includes: when the signal time of acceleration is within the heartbeat time of the heartbeat area, determining the target forgetting factor of acceleration to be a first forgetting factor; when the signal time of acceleration is outside the heartbeat time of the heartbeat area, determining the target forgetting factor of acceleration to be a second forgetting factor, and the second forgetting factor is smaller than the first forgetting factor.

[0092] For example, when the signal time of the acceleration is within the heartbeat time of the heartbeat region, the target forgetting factor of the acceleration that appears within the heartbeat time is set to the first forgetting factor. When the signal time of the acceleration is outside the heartbeat time of the heartbeat region, the target forgetting factor of the acceleration that appears outside the heartbeat time is set to the second forgetting factor.

[0093] It is understandable that the first forgetting factor and the second forgetting factor can be customized by the designer according to engineering experience, for example, the first forgetting factor is 0.8 and the second forgetting factor is 0.2.

[0094] It can also be understood that, since the second forgetting factor is smaller than the first forgetting factor, the acceleration signal in the heart beat region at the signal time will be given a larger forgetting factor, that is, the acceleration signal in the heart beat region at the signal time will be forgotten less, and the acceleration signal outside the heart beat region will be given a smaller forgetting factor, that is, the acceleration signal in the heart beat region at the signal time will be forgotten more, which is conducive to rapid convergence and improving the adaptive filtering effect.

[0095] In some embodiments, the heartbeat time includes a P wave interval, a QRS interval, and a T wave interval, the first forgetting factor includes a first factor, a second factor, and a third factor, and the second factor is greater than the first factor and the third factor. When the signal time of the acceleration is within the heartbeat time of the heartbeat area, determining the target forgetting factor of the acceleration as the first forgetting factor includes the following steps: when the signal time of the acceleration is in the P wave interval of the heartbeat area, determining the target forgetting factor of the acceleration as the first factor, when the signal time of the acceleration is in the QRS interval of the heartbeat area, determining the target forgetting factor of the acceleration as the second factor, and when the signal time of the acceleration is in the T wave interval of the heartbeat area, determining the target forgetting factor of the acceleration as the third factor.

[0096] The QRS wave is an important characteristic wave of the ECG signal. Similarly, the QRS interval is an important interval of the heart beat time. Therefore, a larger forgetting factor can be matched to the QRS interval, while a relatively smaller forgetting factor can be matched to the P wave interval or the T wave interval. This is conducive to improving the adaptive filtering effect, retaining important waveform features, and helping the subsequent steps to perform heart rhythm detection reliably and accurately.

[0097] In some embodiments, when the signal time of acceleration is in the P wave interval of the heart beat area, the closer the signal time of acceleration is to the QRS interval, the larger the first factor corresponding to the acceleration is, and the farther the signal time of acceleration is from the QRS interval, the smaller the first factor corresponding to the acceleration is. As mentioned above, the QRS wave is an important characteristic wave of the ECG signal. The closer the ECG signal of the P wave is to the QRS wave, the more obvious the waveform characteristics it carries will be, and the farther away from the QRS wave, the less obvious the waveform characteristics it carries will be. Therefore, this embodiment can monotonically increase or decrease the first factor according to the distance between the ECG signal of the P wave and the QRS wave, so as to improve the adaptive filtering effect and retain important waveform characteristics.

[0098] In some embodiments, when the signal time of acceleration is in the T wave interval of the heart beat area, the closer the signal time of acceleration is to the QRS interval, the larger the third factor corresponding to the acceleration is, and the farther the signal time of acceleration is from the QRS interval, the smaller the third factor corresponding to the acceleration is. As mentioned above, the QRS wave is an important characteristic wave of the ECG signal. The closer the ECG signal of the T wave is to the QRS wave, the more obvious the waveform characteristics it carries will be, and the farther away from the QRS wave, the less obvious the waveform characteristics it carries will be. Therefore, this embodiment can monotonically increase or decrease the third factor according to the distance between the ECG signal of the T wave and the QRS wave, so as to improve the adaptive filtering effect and retain important waveform characteristics.

[0099] In some embodiments, the heart rhythm detection result includes a ventricular fibrillation result. Before generating the heart rhythm detection result, the method also includes the following steps: detecting whether the user wearing the wearable device is in a stationary state based on acceleration; if the user is in a stationary state, generating a ventricular fibrillation detection result based on the second lead ECG signal; if the user is in motion, generating a ventricular beat detection result, an atrial beat detection result or an atrial fibrillation detection result based on the second lead ECG signal.

[0100] Among the various major arrhythmia states, the ECG performance of ectopic beats and atrial fibrillation is regular and obviously abnormal compared to the performance of noise, while in ventricular fibrillation, the ECG performance is irregular, resulting in partial similarity with noise. When ventricular fibrillation occurs, due to severe hemodynamic disorders, the patient will experience symptoms such as loss of consciousness and syncope and stop exercising. Therefore, it can be considered that ventricular fibrillation usually does not occur when the user exercises normally. It is easy to misidentify ventricular fibrillation when the user is in motion, for example, ventricular fibrillation may be misidentified as non-ventricular fibrillation.

[0101] The acceleration provided by the related technology is often used for fall detection or is only used in the filtering and noise reduction process. However, this embodiment not only applies the acceleration to the adaptive filtering process, but also adds it to the arrhythmia recognition process, thereby reducing the misidentification of ventricular fibrillation.

[0102] In some embodiments, detecting whether a user wearing a wearable device is in a stationary state according to acceleration includes the following steps: calculating the acceleration value of the acceleration, determining whether the acceleration value is greater than a preset threshold, if greater than, determining that the user wearing the wearable device is in motion, if less than or equal to, determining that the user wearing the wearable device is in a stationary state. The preset threshold is customized by the designer based on engineering experience.

[0103] In some embodiments, the acceleration is a three-axis acceleration, which includes an x-axis acceleration, a y-axis acceleration, and a z-axis acceleration. Calculating the acceleration value of the acceleration includes the following steps: respectively calculating the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration, calculating the sum of the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration, taking the square root of the sum to obtain the acceleration value of the three-axis acceleration.

[0104] For example, the present embodiment calculates the acceleration value of the three-axis acceleration according to the following formula, which is as follows: Where a is the acceleration value of the three-axis acceleration.

[0105] This embodiment can synthesize the acceleration values ​​of multiple channels into three-axis acceleration, which is helpful to reliably determine whether the user is in a state of motion, and further reliably detect whether ventricular fibrillation occurs.

[0106] In some embodiments, when there are multiple lead ECG signals, generating a heart rhythm detection result based on the second lead ECG signal includes: determining the number of second lead ECG signals corresponding to the same type of heart rhythm detection results, and the total number of lead ECG signals, calculating the ratio of each type of number to the total number, determining the maximum ratio among multiple ratios, and judging whether the maximum ratio is greater than a preset ratio threshold; if so, reporting the heart rhythm detection result corresponding to the maximum ratio to the background; if less than or equal to, re-performing the arrhythmia detection operation.

[0107] For example, the wearable device can simultaneously detect 4-channel lead ECG signals, and the preset ratio threshold is 50%. If the heart rhythm detection results corresponding to the lead ECG signals of the first channel to the third channel are all normal atrial beat results, and the heart rhythm detection result corresponding to the lead ECG signal of the fourth channel is an atrial fibrillation result, since the number of lead ECG signals with the same normal atrial beat results is 3, the total number is 4, the ratio is 3 / 4, and the number of lead ECG signals with atrial fibrillation results is 1, the ratio is 1 / 4, therefore, the maximum ratio is 3 / 4, and 3 / 4 is greater than 50%. Therefore, this embodiment reports the normal atrial beat results to the background, which can improve the detection robustness and avoid some noise factors that cause the corresponding channels to be misidentified, resulting in the final detection results being inaccurate and unreliable.

[0108] It should be noted that, in each of the above-mentioned embodiments, there is not necessarily a certain order between the above-mentioned steps. A person skilled in the art can understand, based on the description of the embodiments of the present invention, that in different embodiments, the above-mentioned steps may have different execution orders, that is, they may be executed in parallel, may be executed interchangeably, and so on.

[0109] As another aspect of the present invention, the present invention provides an arrhythmia detection device, which is applied to a wearable device. The arrhythmia detection device can be a software module, which includes a number of instructions stored in a memory, and a processor can access the memory and call the instructions for execution to complete the arrhythmia detection method described in each of the above embodiments.

[0110] In some embodiments, the arrhythmia detection device can also be constructed by hardware devices, for example, the arrhythmia detection device can be constructed by one or more chips, and the chips can work in coordination with each other to complete the arrhythmia detection methods described in the above embodiments. For another example, the arrhythmia detection device can also be constructed by various logic devices, such as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISC Machine) or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination of these components.

[0111] See also Figure 8 The arrhythmia detection device 800 includes a signal acquisition module 81, a heartbeat determination module 82, a weight adjustment module 83, an adaptive filtering module 84, a bandpass filtering module 85 and a result generation module 86.

[0112] The signal acquisition module 81 is used to acquire the lead ECG signal and acceleration collected by the wearable device, the heartbeat determination module 82 is used to determine the heartbeat area according to the lead ECG signal, the weight adjustment module 83 is used to determine the target forgetting factor of the acceleration according to the heartbeat time of the heartbeat area, the adaptive filtering module 84 is used to perform adaptive filtering processing on the lead ECG signal according to the target forgetting factor and acceleration to obtain a first lead ECG signal, the bandpass filtering module 85 is used to perform bandpass filtering processing on the first lead ECG signal to obtain a second lead ECG signal, and the result generation module 86 is used to generate a heart rhythm detection result according to the second lead ECG signal.

[0113] This embodiment does not rely on the given current moment as the acceleration matching forgetting factor, but fully considers the importance of the heartbeat area and the important influence of the heartbeat area on the heart rhythm detection results, and determines the target forgetting factor of acceleration according to the heartbeat time. This can improve the signal quality and effect of adaptive filtering, and thus also improve the detection accuracy of arrhythmia.

[0114] In some embodiments, the weight adjustment module 83 is specifically used to: determine the signal time of the acceleration, and determine the target forgetting factor of the acceleration according to the relative position of the signal time of the acceleration and the heartbeat time of the heartbeat area.

[0115] In some embodiments, the weight adjustment module 83 is also specifically used to: when the signal time of acceleration is within the heart beat time of the heart beat area, determine the target forgetting factor of acceleration as the first forgetting factor; when the signal time of acceleration is outside the heart beat time of the heart beat area, determine the target forgetting factor of acceleration as the second forgetting factor, and the second forgetting factor is smaller than the first forgetting factor.

[0116] In some embodiments, the heart beat time includes a P wave interval, a QRS interval and a T wave interval, the first forgetting factor includes a first factor, a second factor and a third factor, the second factor is greater than the first factor and the third factor, and in some embodiments, the weight adjustment module 83 is also specifically used for: when the signal time of acceleration is in the P wave interval of the heart beat area, determining the target forgetting factor of acceleration as the first factor, when the signal time of acceleration is in the QRS interval of the heart beat area, determining the target forgetting factor of acceleration as the second factor, and when the signal time of acceleration is in the T wave interval of the heart beat area, determining the target forgetting factor of acceleration as the third factor.

[0117] In some embodiments, when the signal time of acceleration is in the P wave interval of the heartbeat area, the closer the signal time of acceleration is to the QRS interval, the larger the first factor corresponding to the acceleration is, and the farther the signal time of acceleration is from the QRS interval, the smaller the first factor corresponding to the acceleration is.

[0118] In some embodiments, when the signal time of acceleration is in the T wave interval of the cardiac beat area, the closer the signal time of acceleration is to the QRS interval, the larger the third factor corresponding to the acceleration is, and the farther the signal time of acceleration is from the QRS interval, the smaller the third factor corresponding to the acceleration is.

[0119] In some embodiments, the heart rhythm detection result includes a ventricular fibrillation detection result. Before generating the heart rhythm detection result, the result generation module 86 is specifically used to: detect whether the user wearing the wearable device is in a stationary state based on acceleration; if the user is in a stationary state, generate a ventricular fibrillation detection result based on the second lead ECG signal.

[0120] In some embodiments, the result generation module 86 is also specifically used to: calculate the acceleration value of the acceleration, determine whether the acceleration value is greater than a preset threshold, if greater than, determine that the user wearing the wearable device is in motion, if less than or equal to, determine that the user wearing the wearable device is in a stationary state.

[0121] In some embodiments, the acceleration is a three-axis acceleration, which includes an x-axis acceleration, a y-axis acceleration, and a z-axis acceleration. The result generation module 86 is also specifically used to: calculate the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration, respectively, calculate the sum of the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration, take the square root of the sum, and obtain the acceleration value of the three-axis acceleration.

[0122] It should be noted that the above-mentioned arrhythmia detection device can execute the arrhythmia detection method provided in the embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For technical details not fully described in the embodiment of the arrhythmia detection device, please refer to the arrhythmia detection method provided in the embodiment of the present invention.

[0123] See also Fig. 9 , Fig. 9 A schematic diagram of the circuit structure of a controller provided by an embodiment of the present invention. Fig. 9 As shown, the controller 900 includes one or more processors 91 and a memory 92. Fig. 9 A processor 91 is taken as an example.

[0124] The processor 91 and the memory 92 may be connected via a bus or other means. Fig. 9 The example of connecting through bus is taken in the following.

[0125] The memory 92 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the arrhythmia detection method in the embodiment of the present invention. The processor 91 executes various functional applications and data processing of the arrhythmia detection device by running the non-volatile software programs, instructions and modules stored in the memory 92, that is, realizes the arrhythmia detection method provided in the above method embodiment and the functions of each module or unit in the above device embodiment.

[0126] The memory 92 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 92 may optionally include a memory remotely arranged relative to the processor 91, and these remote memories may be connected to the processor 91 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0127] The program instructions / modules are stored in the memory 92, and when executed by the one or more processors 91, the arrhythmia detection method in any of the above method embodiments is executed.

[0128] An embodiment of the present invention further provides a non-volatile computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are executed by one or more processors, such as Fig. 9 A processor 91 in the embodiment may enable the one or more processors to execute the arrhythmia detection method in any of the above method embodiments.

[0129] An embodiment of the present invention further provides a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a controller, the controller executes any one of the arrhythmia detection methods.

[0130] The above described device or equipment embodiments are merely illustrative, wherein the unit modules described as separate components may or may not be physically separated, and the components displayed as module units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple network module units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0131] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may also be combined, the steps may be implemented in any order, and there are many other changes in different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they may still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting arrhythmia, applied to a wearable device, characterized in that: include: Obtaining lead ECG signals and acceleration collected by the wearable device; determining a heartbeat region according to the lead ECG signal; Determining a target forgetting factor of the acceleration according to the heartbeat time of the heartbeat area; Performing adaptive filtering on the lead ECG signal according to the target forgetting factor and the acceleration to obtain a first lead ECG signal; Performing bandpass filtering on the first lead ECG signal to obtain a second lead ECG signal; Generate a heart rhythm detection result according to the second-lead ECG signal.

2. The method according to claim 1, characterized in that The step of determining the target forgetting factor of the acceleration according to the heartbeat time of the heartbeat area comprises: determining a signal time of the acceleration; The target forgetting factor of the acceleration is determined according to the relative position of the signal time of the acceleration and the heartbeat time of the heartbeat area.

3. The method according to claim 2, characterized in that Determining the target forgetting factor of the acceleration according to the relative position between the signal time of the acceleration and the heartbeat time of the heartbeat area comprises: When the signal time of the acceleration is within the heartbeat time of the heartbeat area, determining the target forgetting factor of the acceleration to be a first forgetting factor; When the signal time of the acceleration is outside the heartbeat time of the heartbeat area, it is determined that the target forgetting factor of the acceleration is a second forgetting factor, and the second forgetting factor is smaller than the first forgetting factor.

4. The method according to claim 3, characterized in that The heartbeat time includes a P wave interval, a QRS interval and a T wave interval, the first forgetting factor includes a first factor, a second factor and a third factor, the second factor is greater than the first factor and the third factor, and when the signal time of the acceleration is within the heartbeat time of the heartbeat area, determining that the target forgetting factor of the acceleration is the first forgetting factor includes: When the signal time of the acceleration is within the P wave interval of the heartbeat region, determining the target forgetting factor of the acceleration to be the first factor; When the signal time of the acceleration is within the QRS interval of the heartbeat region, determining the target forgetting factor of the acceleration to be a second factor; When the signal time of the acceleration is within the T wave interval of the heartbeat region, the target forgetting factor of the acceleration is determined to be the third factor.

5. The method according to claim 4, characterized in that When the signal time of the acceleration is in the P wave interval of the heartbeat area, the closer the signal time of the acceleration is to the QRS interval, the larger the first factor corresponding to the acceleration is, and the farther the signal time of the acceleration is from the QRS interval, the smaller the first factor corresponding to the acceleration is; or, When the signal time of the acceleration is in the T wave interval of the heartbeat area, the closer the signal time of the acceleration is to the QRS interval, the larger the third factor corresponding to the acceleration is, and the farther the signal time of the acceleration is from the QRS interval, the smaller the third factor corresponding to the acceleration is.

6. The method according to any one of claims 1 to 5, characterized in that: The heart rhythm detection result includes a ventricular fibrillation detection result. Before generating the heart rhythm detection result, the method further includes: Detecting whether a user wearing the wearable device is in a stationary state according to the acceleration; If the user is in a stationary state, a ventricular fibrillation detection result is generated according to the second lead electrocardiogram signal.

7. The method according to claim 6, characterized in that The detecting, according to the acceleration, whether the user wearing the wearable device is in a stationary state comprises: Calculating the acceleration value of the acceleration; Determining whether the acceleration value is greater than a preset threshold; If it is greater than, it is determined that the user wearing the wearable device is in motion; If it is less than or equal to, it is determined that the user wearing the wearable device is in a stationary state.

8. The method according to claim 7, characterized in that The acceleration is a three-axis acceleration, which includes an x-axis acceleration, a y-axis acceleration, and a z-axis acceleration. The acceleration value of the acceleration is calculated as follows: Calculating the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration respectively; Calculating the sum of the square of the x-axis acceleration, the square of the y-axis acceleration, and the square of the z-axis acceleration; Taking the square root of the sum, the acceleration values ​​of the three-axis acceleration are obtained.

9. A non-volatile readable storage medium, characterized in that: The non-volatile readable storage medium stores computer executable instructions, and the computer executable instructions are used to enable a controller to execute the arrhythmia detection method according to any one of claims 1 to 8.

10. A wearable device, characterized in that: include: Heart rate sensor; Accelerometer; at least one processor, electrically connected to the heart rate sensor and the acceleration sensor respectively; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the arrhythmia detection method according to any one of claims 1 to 8.

Images