Hemocardiogram analysis system and method based on accelerometer data

Abstract

The invention relates to the technical field of medical care informatics, in particular to a seismocardiogram analysis system and method based on accelerometer data, and the system comprises a seismocardiogram detection unit, a motion artifact detection unit and a data processing unit. The heart vibration detection unit is used for collecting sternum vibration signals generated by heart activity. The motion artifact detection unit is used for detecting a sternum posture signal generated by thoracic activity during breathing. In the application, the data processing unit acquires signals of the cardiac vibration detection unit and the motion artifact detection unit, encodes the signals into vector group data, performs digital filtering on the vector group data, preliminarily performs band-pass filtering, only retains low-frequency signals, and performs modal decomposition on sternum vibration signals based on a VMD algorithm. Parameter features of the sternum attitude signals are used as constraint conditions, and SCG signals in the sternum attitude signals are decomposed.

Description

Technical Field

[0001] The present application relates to the technical field of healthcare informatics, and particularly to a ballistocardiogram analysis system and method based on accelerometer data. Background Art

[0002] Currently, the detection of cardiovascular health mainly relies on clinical operations such as electrocardiogram, echocardiogram, and coronary angiography. Among them, both echocardiogram and coronary angiography require professional physicians to operate on the examination table, with high equipment requirements and unable to achieve long-term monitoring. Moreover, the electrocardiogram has insufficient sensitivity in detecting heart diseases, and the characteristics of some arrhythmias caused by diseases are not obvious in the ECG signal.

[0003] The amplitude change of the SCG signal is strongly correlated with the contraction and relaxation of the heart and can reflect the key link information of heart activities. Compared with the electrocardiogram signal, the ballistocardiogram signal (i.e., the SCG signal) has higher sensitivity to abnormal vibrations caused by heart diseases and shows more obvious characteristics in the SCG signal.

[0004] Currently, the close connection between each characteristic point in the SCG signal and heart activities has not been fully understood. To explain the connection between the SCG signal and heart activities, it is necessary to extract the characteristics of the SCG signal, narrow down the recognition range, and thus determine the correlation between the lesion type and signal characteristics.

[0005] In the existing ballistocardiogram signal acquisition system, an accelerometer is mainly fixed in the sternum area of the human body. This acceleration sensor can measure the vibration information generated by the contraction and relaxation of the heart, and then perform signal processing on the vibration information to obtain the ballistocardiogram.

[0006] However, the vibration of the sternum is also caused by human breathing and movement. Accordingly, baseline drift and motion artifacts exist in the vibration information obtained by the accelerometer. Therefore, the vibration information collected by the accelerometer needs to be filtered to obtain an accurate ballistocardiogram.

[0007] The accelerometer data is a non-stationary signal, and its amplitude and phase change frequently, resulting in ineffective processing of the existing band-pass filter.

[0008] To remove the interference of motion artifacts in the SCG signal, the present application provides a ballistocardiogram analysis system and method based on accelerometer data. Summary of the Invention

[0009] To overcome the problems existing in the related art, a first aspect of the present application provides a ballistocardiogram analysis system based on accelerometer data, including: a ballistocardiogram detection unit, a motion artifact detection unit, and a data processing unit;

[0010] The heart vibration detection unit and the motion artifact detection unit are communicatively connected to the data processing unit;

[0011] The heart vibration detection unit is provided with a sternum accelerometer, and the sternum accelerometer is used to collect the sternum vibration signals generated by heart activities;

[0012] The motion artifact detection unit is provided with a patch-type piezoelectric sensor, and the patch-type piezoelectric sensor is adhered to the sternum area for detecting the sternum attitude signals generated by the thoracic cage activities during breathing;

[0013] The data processing unit is used to extract the SCG signals in the sternum vibration signals according to the sternum attitude signals.

[0014] In one embodiment, the patch-type piezoelectric sensor is provided with a dielectric elastomer piezoelectric film, a first electrode and a second electrode;

[0015] The first electrode and the second electrode are respectively arranged on the top surface and the bottom surface of the dielectric elastomer piezoelectric film;

[0016] The first electrode and the second electrode are electrically connected to the data processing unit.

[0017] In one embodiment, the data processing unit includes a signal acquisition module, a pre-filtering module and a modal decomposition module;

[0018] The signal acquisition module is used to acquire the sternum attitude signals and the sternum vibration signals;

[0019] The pre-filtering module is used to perform band-pass filtering on the sternum attitude signals and the sternum vibration signals;

[0020] The modal decomposition module is used to decompose the sternum vibration signals into K modal components and extract the SCG signals from the K modal components.

[0021] In one embodiment, the signal decomposition module performs modal decomposition based on the VMD algorithm, and the VMD algorithm incorporates the parameter characteristics of the sternum attitude signals into the constraint conditions of the objective function when performing variational modal decomposition.

[0022] The second aspect of the present application provides a method for analyzing electrocardiogram based on accelerometer data, which is used for the electrocardiogram analysis system described in the second aspect of the present application, and includes the following steps:

[0023] S1. Acquire the sternum vibration signals and the sternum attitude signals;

[0024] S2. Perform preliminary filtering processing on the sternum vibration signals and the sternum attitude signals;

[0025] S3. Input the sternum vibration signal into an improved VMD algorithm model to obtain K modal components; K is an integer greater than or equal to 2.

[0026] S4. Identify the SCG signal in the k modal components.

[0027] S5. Perform cardiogram analysis based on the SCG signal.

[0028] In one implementation, S2 specifically includes:

[0029] S200. Construct an improved VMD algorithm model.

[0030] S201. Input the filtered sternum vibration signal and the sternum posture signal.

[0031] S203. Iteratively update u k , w k and λ.

[0032] S204. Determine whether the iteration termination condition is satisfied. If so, stop the iteration and output K modal components; if not, execute step S203.

[0033] In one implementation, the mode decomposition function of the VMD algorithm is:

[0034]

[0035] where u k is the k-th modal component, f(t) is the sternum vibration signal, w g (t) is the central frequency of the sternum posture signal, and w 1 (t) is the first modal component of the sternum vibration signal.

[0036] In one implementation, the constraint condition of the VMD algorithm model is:

[0037] Σu k = f(t)

[0038]

[0039] where u k is the k-th modal component, f(t) is the sternum vibration signal, w g (t) is the central frequency of the sternum posture signal, and w 1 (t) is the first modal component of the sternum vibration signal;

[0040] Introduce a penalty factor and a Lagrange multiplier to solve the variational constraint problem, and the resulting augmented Lagrangian expression is:

[0041]

[0042] Among them, L({u k},{w k}, λ) is the augmented Lagrangian function, α is the penalty factor, λ is the Lagrange multiplier, T is the time length, and w g (t) is the center frequency of the sternal posture signal.

[0043] In one embodiment, the termination condition of the VMD algorithm model is

[0044]

[0045] Among them, is the Fourier transform of the u k (t) function in the (n + 1)-th iteration, is the Fourier transform of the u k (t) function in the n-th iteration, ∈ is the discrimination accuracy, and ∈ is greater than 0.

[0046] The technical solution provided by this application may include the following beneficial effects:

[0047] In this application, the data processing unit acquires the signals of the heart shock detection unit and the motion artifact detection unit, encodes them into vector group data, performs digital filtering on the vector group data, initially performs band-pass filtering, only retains low-frequency signals, and then performs modal decomposition on the sternal vibration signal based on the VMD algorithm, and decomposes the SCG signal therein with the parameter characteristics of the sternal posture signal as the constraint condition.

[0048] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit this application. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] By describing the exemplary embodiments of this application in more detail in conjunction with the drawings, the above and other objects, features, and advantages of this application will become more obvious. Among them, in the exemplary embodiments of this application, the same reference numerals generally represent the same components.

[0050] Figure 1 is the layout structure diagram of the heart shock graph analysis system shown in the embodiments of this application;

[0051] Figure 2 is Figure 1 the logical structure diagram of the data analysis unit of the heart shock graph analysis system shown;

[0052] Figure 3 is the flow diagram of the heart shock graph analysis method shown in the embodiments of this application;

[0053] Figure 4 For Figure 3 the schematic flowchart of step S2 in the heart vibration diagram analysis method shown;

[0054] Explanation of the attached figure labels: 1. Heart vibration detection unit; 2. Motion artifact detection unit; 3. Data processing unit. Detailed implementation mode

[0055] The preferred implementation modes of the present application will be described in more detail below with reference to the accompanying drawings. Although the preferred implementation modes of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the implementation modes set forth herein. On the contrary, these implementation modes are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0056] The terms used in the present application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "the" and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0057] It should be understood that although the terms "first", "second", "third", etc. may be used in the present application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, the meaning of "a plurality" is two or more unless otherwise specifically defined.

[0058] Embodiment 1

[0059] During the heart vibration detection process, in order to accurately remove the motion artifact signal in the accelerometer data, the embodiment of the present application provides a heart vibration diagram analysis system based on accelerometer data, as Figure 1 shown, including: a heart vibration detection unit 1, a motion artifact detection unit 2, and a data processing unit 3.

[0060] As Figure 1 shown, the heart vibration detection unit 1 is provided with a sternum accelerometer, and the motion artifact detection unit 2 is provided with a patch-type piezoelectric sensor.

[0061] Specifically, the sternum accelerometer is integrated in the MEMS chip, and the MEMS chip is connected to the data collector through a communication line. The patch-type piezoelectric sensor is integrated in a deformable sheet, and the patch-type sensor is connected to the data processing unit 3 through a communication line.

[0062] Furthermore, the patch-type piezoelectric sensor is provided with a dielectric elastomer piezoelectric film, a first electrode, and a second electrode. The first electrode and the second electrode are respectively arranged on the top surface and the bottom surface of the dielectric elastomer piezoelectric film. The first electrode and the second electrode are electrically connected to the data processing unit 3.

[0063] In the embodiment of the present application, the sternum accelerometer is fixed on the top surface of the patch-type piezoelectric sensor, and the bottom surface of the patch-type piezoelectric sensor is adhered to the sternum region of the human body. During heart activity, the sternum accelerometer collects the sternum vibration signals with time sequence changes and sends them to the data processing unit 3. During human breathing, the patch-type piezoelectric sensor collects the sternum posture signals along with the movement of the sternum and sends them to the data processing unit 3.

[0064] The main source of motion artifacts is the expansion and contraction of the sternum. When the sternum moves, the accelerometer on the body surface collects the acceleration changes generated by the sternum movement, resulting in motion artifacts in the sternum vibration signals.

[0065] Furthermore, the data processing unit 3 is used to extract the SCG signal in the sternum vibration signal according to the sternum posture signal.

[0066] In the system shown in the embodiment of the present application, parameters such as the period, phase, and amplitude of the motion artifacts have a strong correlation with the sternum posture. Therefore, in the embodiment of the present application, the sternum posture signal is obtained through the patch-type piezoelectric sensor, and then according to the parameter characteristics of the sternum posture signal, the motion artifacts in the sternum vibration signal are targeted to separate the SCG signal therein.

[0067] Furthermore, as Figure 2 shown, the data processing unit 3 includes a signal acquisition module, a pre-filtering module, and a modal decomposition module.

[0068] The signal acquisition module is used to acquire the sternum posture signal and the sternum vibration signal;

[0069] The pre-filtering module is used to perform preliminary filtering on the sternum posture signal and the mixed-frequency signal;

[0070] The modal decomposition module is used to decompose the mixed-frequency signal into K modal components and extract the SCG signal from the K modal components.

[0071] Further, the signal decomposition module performs modal decomposition processing based on an improved VMD algorithm.

[0072] In the embodiment of the present application, the signal acquisition module acquires the signals of the heart vibration detection unit 1 and the motion artifact detection unit 2 and encodes them into vector group data. The pre-filtering module performs digital filtering on the vector group data, initially performs band-pass filtering, and only retains low-frequency signals. The modal decomposition module performs modal decomposition on the sternum vibration signal based on the VMD algorithm, and uses the parameter characteristics of the sternum posture signal as a constraint condition to decompose the SCG signal therein.

[0073] Embodiment 2

[0074] Based on Embodiment 1, the embodiment of the present application provides a heart vibration diagram analysis method based on accelerometer data, which is applied to the heart vibration diagram analysis system described in Embodiment 1, as Figure 3 shown, and includes the following steps:

[0075] S1. Acquire the sternum vibration signal and the sternum posture signal;

[0076] S2. Perform preliminary filtering processing on the sternum vibration signal and the sternum posture signal;

[0077] S3. Input the sternum vibration signal into an improved VMD algorithm model to obtain K modal components;

[0078] S4. Identify the SCG signal in the K modal components;

[0079] S5. Perform heart vibration diagram analysis and detection according to the SCG signal.

[0080] Specifically, as Figure 4 shown, S2 specifically includes:

[0081] S200. Construct a VMD algorithm model;

[0082] Specifically, the modal decomposition function of the VMD algorithm model is:

[0083]

[0084] where {u k} is a combination of k modal components, {w k} is the corresponding center frequency of the k modal components, δ(t) is the unit impulse function, j is the imaginary unit, * is the convolution operator, is the partial derivative function, and t is time.

[0085] The constraint condition of the VMD algorithm model is:

[0086]

[0087] w g w(t) = 1 (t)

[0088] where u k is the k-th modal component, f(t) is the sternum vibration signal, and w g (t) is the central frequency of the sternum posture signal, and w 1 (t) is the first modal component of the sternum vibration signal.

[0089] Exemplarily, an embodiment of the present application uses the frequency of the sternum posture signal as one of the constraint conditions. In other embodiments, the period or amplitude parameter of the sternum posture signal can also be used as the constraint condition.

[0090] In the embodiment of the present application, the posture change during human breathing is collected by an accelerometer, thereby generating motion artifacts. By virtue of the strong correlation between the posture change signal and the motion artifacts, when processing the signal through the VMD algorithm, setting constraint conditions based on the parameter characteristics of the posture change signal can decompose the motion artifacts generated by breathing from the sternum vibration signal, thereby obtaining a more accurate SCG signal.

[0091] Introduce a penalty factor and a Lagrange multiplier to solve the variational constraint problem, and the obtained augmented Lagrangian expression is:

[0092]

[0093] where L({u k},{w k}, λ) is the augmented Lagrangian function, α is the penalty factor, λ is the Lagrange multiplier, T is the time length, and w g (t) is the central frequency of the sternum posture signal.

[0094] S201. Input the filtered sternum vibration signal and the sternum posture signal;

[0095] S202. Initialize {u k}, {w k}, λ, and n;

[0096] where {u k} is the k modal components, {w k} is the central frequencies of the k modal components, λ is the Lagrange operator, and n is the number of iterations.

[0097] S203. Iteratively update {u k}, {w k}, and λ;

[0098] S204. Determine whether the iteration termination condition is met. If so, stop the iteration and output k modal components; if not, execute step S203.

[0099] Specifically, the iteration termination condition is

[0100]

[0101] where is the Fourier transform of the u k (t) function in the (n + 1)-th iteration, is the Fourier transform of the u k (t) function in the n-th iteration, and ∈ is the discrimination accuracy, where ∈ > 0.

[0102] In the embodiments of the present application, the phase and frequency characteristics of the sternum posture signal are used as the constraint conditions during modal decomposition, so that the phase characteristics and frequency characteristics of a decomposed modal component approach the sternum posture signal.

[0103] Regarding the device in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method. Therefore, detailed descriptions will not be provided here again.

[0104] The solutions of the present application have been described in detail above with reference to the accompanying drawings. In the above embodiments, the descriptions of each embodiment have their own emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. Those skilled in the art should also be aware that the actions and modules involved in the specification are not necessarily essential to the present application.

[0105] In addition, it can be understood that the steps in the method embodiments of the present application can be adjusted, combined, and deleted according to actual needs, and the modules in the device embodiments of the present application can be combined, divided, and deleted according to actual needs.

[0106] In addition, the method according to the present application can also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0107] Alternatively, the present application may also be implemented as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) having executable code (or computer program, or computer instruction code) stored thereon, and when the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or an electronic device, a server, etc.), the processor is caused to execute some or all of the steps of the above method according to the present application.

[0108] Those skilled in the art will also understand that the various exemplary logical blocks, modules, circuits, and algorithm steps described in connection with the application herein can be implemented as electronic hardware, computer software, or a combination of both.

[0109] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems and methods according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of code, or a portion thereof that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or by a combination of dedicated hardware and computer instructions.

[0110] The various embodiments of the present application have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, the practical application, or the improvement of technologies in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A seismogram analysis system based on accelerometer data, characterized in that: include: A heart tremor detection unit (1), a motion artifact detection unit (2) and a data processing unit (3); The cardiac tremor detection unit 1 and the motion artifact detection unit (2) are communicatively connected to the data processing unit (3); The cardiac tremor detection unit (1) is provided with a sternum accelerometer, and the sternum accelerometer is used to collect sternum vibration signals generated by cardiac activity; The motion artifact detection unit (2) is provided with a patch-type piezoelectric sensor, which is adhered to the sternum area and is used to detect a sternum posture signal generated by chest cage movement during breathing; The data processing unit (3) is used to extract the SCG signal from the sternum vibration signal according to the sternum posture signal.

2. The seismogram analysis system based on accelerometer data according to claim 1, characterized in that: The patch type piezoelectric sensor is provided with a dielectric elastomer piezoelectric film, a first electrode and a second electrode; The first electrode and the second electrode are respectively disposed on the top surface and the bottom surface of the dielectric elastomer piezoelectric film; The first electrode and the second electrode are electrically connected to the data processing unit (3).

3. The seismogram analysis system based on accelerometer data according to claim 1, characterized in that: The data processing unit (3) comprises a signal acquisition module, a pre-filtering module and a modal decomposition module; The signal acquisition module is used to obtain the sternum posture signal and the sternum vibration signal; The pre-filtering module is used to perform band-pass filtering on the sternum posture signal and the sternum vibration signal; The modal decomposition module is used to decompose the sternum vibration signal into K modal components and extract the SCG signals from the K modal components.

4. The seismogram analysis system based on accelerometer data according to claim 3, characterized in that: The signal decomposition module performs modal decomposition based on the VMD algorithm, and the VMD algorithm incorporates parameter features of the sternum posture signal into the constraint conditions of the objective function when performing variational modal decomposition.

5. A seismogram analysis method based on accelerometer data, characterized in that: The seismogram analysis system used in any one of claims 1 to 4 comprises the following steps: S1, obtaining the sternum vibration signal and the sternum posture signal; S2, performing preliminary filtering processing on the sternum vibration signal and the sternum posture signal; S3, inputting the sternum vibration signal into the improved VMD algorithm model to obtain K modal components, where K is an integer greater than or equal to 2; S4, identifying SCG signals in k of the modal components; S5. Perform seismocardiogram analysis based on the SCG signal.

6. A seismogram analysis method based on accelerometer data according to claim 5, characterized in that: S2 specifically includes: S200, constructing an improved VMD algorithm model; S201, inputting the filtered sternum vibration signal and the sternum posture signal; S203, iterative update u k 、w k and λ; S204, judging whether the iteration termination condition is met, if so, stopping the iteration and outputting K modal components; if not, executing step S203.

7. A seismogram analysis method based on accelerometer data according to claim 6, characterized in that: The mode decomposition function of the VMD algorithm is: Among them, {u k } is a combination of K modal components, {w k } is the corresponding center frequency of the K modal components, δ(t) is a single * is a convolution operator, is the partial derivative function, and t is the time.

8. The method for analyzing seismogram based on accelerometer data according to claim 7, characterized in that: The constraints of the VMD algorithm model are: w g (t)=w1(t) Among them, u k is the kth modal component, f(t) is the sternum vibration signal, w g (t) is the sternum; The penalty factor and Lagrange multiplier are introduced to solve the variational constraint problem, and the obtained augmented Lagrangian expression is: Among them, L({u k },{w k },λ) is the augmented Lagrangian function, α is the penalty factor, λ is the Lagrangian multiplier, T is the time length, w g (t) is the center frequency of the sternum posture signal.

9. A seismogram analysis method based on accelerometer data according to claim 8, characterized in that: The termination condition of the VMD algorithm model is in, is the u of the n+1th iteration k (t) is the Fourier transform of the function, is the u of the nth iteration k (t) is the Fourier transform of the function, ∈ is the discrimination accuracy, ∈ is greater than 0.

Images