TWI410233B - The Architecture and Method of Removing Shaking Noise Technology of Physiological Signal - Google Patents

Abstract

The present invention relates to a technical framework for reducing artificial motion noises of physiological signal and a method thereof. Specially, it is a structure method that applies an adaptive filter technique to reduce the noises generated by human body's artificial motion coupled with the physiological signal, and uses a tri-axial-accelerameter to record the signal of human body's artificial motion as a reference signal source. Additionally, this method adopts a least-mean square (LMS) algorithm, and this LMS algorithm is applied to the single-chip microcontroller to effectively reduce the artificial motion noises that are easily coupled while measuring the physiological signal. By this design, the present invention can promote the accuracy of an overall physiological monitoring and the testee's use safety.

Description

Structure and method for swaying noise technology for removing physiological signals

The invention belongs to the field of filtering technology for physiological monitoring, and specifically relates to a structure and a method for swaying noise technology capable of removing physiological signals, thereby improving the accuracy and safety of physiological monitoring.

According to the basic life phenomena of the human body, brain waves, muscle movement, heartbeat, blood pressure and respiration, the parameters obtained by measuring these life phenomena are also the most direct indicators for clinically assessing the severity of the disease and the degree of recovery after recovery. The human cardiovascular system is controlled by many rhythmic processes that interact with each other, and the most important is the heart rhythm; and the potential change of the heart itself is reflected to the body surface through the conductive tissue and body fluid around the heart, so-called "Electrocardiogram" is the use of microelectrode technology to record the difference between the internal and external myocardial cells generated by the changes of the tiny electric pulse of the heart. The instrument amplifies the pattern drawn by the electrical activity signal to understand whether the heart is functioning normally. The waveform of the electrocardiogram is often used clinically. The most important tool for diagnosing heart diseases such as arrhythmia, atrioventricular block, and myocardial infarction.

The electrocardiogram is not only used for the diagnosis of heart disease, but also for the evaluation of the effectiveness of medication treatment. It is widely used in respiratory arrest or wear-through physiological monitoring systems; the ECG signal belongs to the low frequency range (0.05~100HZ), and the amplitude It is only 1~10mV, so it needs to be placed on the ECG signal. Large devices and filtering devices, but ECGs often couple various noises, including myoelectric signals and respiratory interference noise caused by the human body itself, and external noise such as AC interference, electromagnetic interference, human body shaking noise, etc. Etc., for the accuracy of measurement, the removal of noise is necessary; the general method of removing noise is to design a fixed filter using the known signal spectrum; but when the spectral distribution of noise and signal overlaps At the same time, the traditional linear filter has no way to produce effective separation; however, in recent years, the portable monitoring system is equipped with a powerful weight adjustment device, so the digital filter can be written into the weight adjustment device. To achieve the performance of real-time calculus. However, the bandwidth of the human body shaking noise is often the same as the bandwidth of the physiological signal, so the human body shaking noise is the most difficult noise to be filtered out.

The adaptive filter can be thought of as a high-pass filter or a filter with a filter. This technology is mainly used for power line interference filtering and fetal electrocardiogram measurement including 60 Hz. Other applications include improved signal-to-noise ratio (SNR) and the use of multiple pilot signals to identify P-waves or to remove interference signals from the chest impedance signal. Thakor and Zhu et al. propose a recursive adaptive filter architecture that uses pulse waves synchronized with QRS waves as input signals. According to these studies, if the measurement system can identify the noise source, the adaptive filter can extract the pure signal source from the measured signal.

However, the adaptive filter is a dynamic filter. When the characteristics of the noise change, it can track the noise and achieve the filtering effect. The filtering is the operation of filtering out the specific band frequency signal in the signal. And one to prevent interference Important measures, in other words, if a method for effectively filtering the human body's swaying noise can be proposed, the accuracy of the overall physiological signal monitoring can be greatly improved, thereby protecting the health and safety of the patient.

In view of this, the inventor of the present invention has intensively discussed the problem that the existing human body can not effectively filter out the swaying noise when using the physiological monitoring device, and actively seeks a solution through years of research and development experience in related industries. Continuous efforts in research and trials have finally succeeded in developing a structure and method for swaying noise technology to remove physiological signals, thereby overcoming the inconvenience caused by the inaccuracy of coupled physiological noise in physiological signals. Troubled.

The object of the present invention is to provide a structure and method for removing the physiological signal swaying noise technology, thereby using the least mean square calculation method to adjust the weight of the adaptive filter, and the adaptive filter can be used over time or Changes in the characteristics of the signal can improve the accuracy of physiological signal monitoring and further promote user safety.

The present invention mainly implements the foregoing objectives and effects through the following technical means; the method includes: a first signal receiver for monitoring a physiological signal of a human body, a measurable physiological signal, and an eye movement a second signal receiver for receiving a three-axis shaking signal of the human body, and an electrocardiogram, an electromyogram, an electroencephalogram, an electroencephalogram, and a light volume change tracing; The second signal receiver comprises an accelerometer, a gyroscope and a strain gauge three-axis accelerator; an amplifier module, the input end of which is connected to the first signal receiver, and the amplifier module comprises an amplifier with positive and negative input a signal amplifier and an output filter; an analog-to-digital converter having an input terminal connected to the aforementioned amplifier module and a second signal receiver; a multiplier connected to an output of the analog-to-digital converter outputted by the amplifier module and An output of the weight adjustment device, wherein the multiplier receives the signal of the shaking of the second signal receiver as a reference signal; an adder connected to the output of the multiplier; and a subtractor connected to the first signal receiver An analog-to-digital converter and an output of the multiplier; a weight adjustment device coupled to the output of the subtractor, the weight adjustment device having a built-in minimum mean square algorithm. Used for weight adjustment.

The method steps include: using a signal for obtaining and recording body shaking; using the shaking signal as a filtering reference signal; and using a minimum mean square calculation to adjust the weight of the filtering.

Therefore, through the foregoing technical means of the present invention, the present invention can use the three-axis accelerator to record the signal of the body shaking as the reference signal source, and adopt the least mean square algorithm, which can effectively apply the algorithm to the weight. Adjustment In the device, the sloshing noise in the physiological signal is effectively filtered, and the accuracy of the overall physiological monitoring can be improved, thereby improving the safety of the object to be tested.

The following is a description of the preferred embodiments of the present invention, and is described in detail with reference to the drawings, and the .

The invention relates to a structure and a method for swaying noise technology for removing physiological signals, and a specific embodiment of the structure of the swaying noise technology for removing physiological signals of the present invention and components thereof are illustrated with reference to the accompanying drawings, all related to front and rear, left References to the right, top and bottom, upper and lower, and horizontal and vertical are for convenience of description only, and are not intended to limit the invention, nor to limit its components to any position or spatial orientation. The drawings and the dimensions specified in the specification may be varied in accordance with the design and needs of the specific embodiments of the present invention without departing from the scope of the invention.

Referring to FIG. 1 , the architecture and method of the swaying noise technology for removing physiological signals provided by the present invention have a first signal receiver A and a second signal receiver B for receiving human body, wherein the first signal is received. The device A can be an eye movement for monitoring, an electrocardiogram, an electromyogram, an electroencephalogram, and a photoplethysmograph, while the second signal receiver B includes an accelerometer, a gyroscope, and a strain. The triaxial accelerator of the specification further has an amplifier module 1 for receiving the signal of the first signal receiver A, and the amplifier module 1 includes the input amplifier 11 of the positive and negative electrodes, the signal amplifier 12 and the output filter. And an analog-to-digital converter 2 is connected to the output of the amplifier module 1 for converting the analog signal received by the first signal receiver A into a digital signal output, and the input of the analog converter 2 is Connecting the foregoing second signal receiver B for converting the analog signal of the second signal receiver B received by the second signal receiver B into a digital signal output; and the analog digital converter 2 The output terminal is connected with a multiplier 3 and a subtractor 5, wherein the output of the multiplier 3 is connected with a subtractor 5, and the output of the subtractor 5 is adjusted by a minimum mean square algorithm (LMS). The device 4 adjusts the weight W and connects the multiplier 3 to generate the body shaking noise contained in the physiological signal, and passes through the subtractor 5, and finally outputs an effective physiological monitoring chart 7 (for example, an eye movement, an electrocardiogram, and a muscle). And the first signal receiver A and the second signal receiver B can receive analog signals of human physiological and physical shaking, and are converted into digital numbers by the analog digital converter 2 After the signal is passed through the multiplier 3, the weight adjustment device 4 and the subtractor 5 with the least mean square algorithm (LMS), the coupling sway noise in the human physiological signal is removed, thereby forming the sway in the physiological signal removal of the present invention. The architect of the noise.

The implementation steps of the method for removing the physiological signal swaying noise technique of the present invention include a signal for obtaining and recording body sway, a signal for filtering the swaying signal, and a minimum mean square calculation for tuning. The weight of the entire filter.

For the actual operation, please still refer to the one shown in Figure 1 and Figure 2. First, the three-axis accelerator of the second signal receiver B is used to acquire and record the mobility signal when the user body is shaking; then, when the shaking is performed The mobility signal is used as a reference signal of the multiplier 3; thereafter, the calculation is performed by the weight adjustment device 4 having the least mean square algorithm (LMS) for adjusting the weight, and the weight is given to the multiplier 3, so that The matrix is multiplied by the input reference signal of the multiplier 3 to generate the body shaking noise contained in the physiological signal, and the body shaking noise in the physiological signal recorded by the first signal receiver A is removed through the subtractor 5. .

The actual implementation of the present invention is as follows: first, the physiological signal containing the sloshing noise is transmitted through the first signal receiver A to the amplifier module 1, and then the reference signal and the deviation of the three-axis accelerator passing through the second signal receiver B are input. Signal; X ( n )=[1 , Acc _x ( n ) , Acc _y ( n ), Acc _z ( n )]................. [B signal formula]

The multiplicand of the multiplier 3 is W = [ w 0, w 1, w 2, w 3], and the multiplier of the multiplier 3 is the deviation signal X of the three-axis accelerator of the second signal receiver B, the multiplier 3 The output is y = XW ^T .

The physiological signal error is defined as the difference between the output of the first signal receiver A through the analog-to-digital converter 2 and the output of the multiplier 3, e = s + n - XW ^T , (1)

e ² = (s + n) ² -2 (s + n)XW ^T + WX ^T XW ^T . (2)

e ² is the physiological signal error energy, and the Eq(2) is minimized by the minimum mean square calculation method, that is, the body shaking signal contained in the physiological signal can be removed.

The least mean square is defined as: E [ e ² ]= E [( s + n) ² ]+2 E [( s + n ) X ] W ^T + WE [ X ^T X ] W ^T , (3)

The minimum mean square calculus is to minimize Eq(3).

According to this method, the next weight W _t+1 is equal to the current weight W _t and a change proportional to the negative slope: W _{t +1} = W _t - α ▽ _t , (4)

The parameter α is an adjustment ratio of the adjustment weight W. The slope of the error energy is defined as:

Therefore the estimated slope is defined as:

Instead of using the estimated slope on the true slope of Eq(5). The Hough minimum mean square algorithm is generated by the formula (4): W _{t +1} = W _t +2 αeX ............(7) [weight adjustment device 4 formula]

This weight adjustment equation is embedded in the weight adjustment means 4 of the weight adjustment algorithm.

During the experiment, Figure 3(a) is the actual ECG signal used during the journey. Figure 3 (c) shows the three-axis signal measured by the three-axis accelerator of the second signal receiver B, which includes the left and right, front and rear and up and down motions, and the third (b) shows the filtered ECG. signal.

Through the design of the invention, the adaptive filtering method is used to remove the moving human noise, and the three-axis accelerator is used to record the body shaking signal as the reference signal source, and the least mean square algorithm is used, which can effectively apply the algorithm. In the weight adjustment device 4 with the weight adjustment algorithm, the noise generated by the human body shaking in the physiological signal can be effectively filtered, and the accuracy of the overall physiological monitoring can be improved, thereby improving the use of the object to be tested. safety.

In summary, this case is not only innovative in terms of space type, but also can enhance the above-mentioned multiple functions compared with the customary items. It should fully comply with the statutory invention patent requirements of novelty and progressiveness, and apply for it according to law. This invention patent application, in order to invent invention, to the sense of virtue.

1‧‧‧Amplifier Module

11‧‧‧Input amplifier

12‧‧‧Signal Amplifier

13‧‧‧Output filter

2‧‧‧ analog digital converter

3‧‧‧Multiplier

4‧‧‧ weight adjustment device

5‧‧‧Subtractor

6‧‧‧effective physiological surveillance map

A‧‧‧First Signal Receiver

B‧‧‧second signal receiver

FIG. 1 is a schematic structural diagram of a sloshing noise technology for removing a physiological signal according to the present invention; FIG. 2 is a schematic diagram of a sloshing noise technical architecture for removing a physiological signal according to the present invention. Schematic diagram of detailed monitoring of the electrocardiogram; Figure 3 (a) shows the waveform diagram of the actual ECG signal between the users; Figure 3 (b) shows the waveform diagram of the filtered ECG signal; and Figure 3 (c) shows A schematic diagram of the waveform of a three-axis signal measured by a three-axis accelerometer.

1‧‧‧Amplifier Module

11‧‧‧Input amplifier

12‧‧‧Signal Amplifier

13‧‧‧Output filter

2‧‧‧ analog digital converter

3‧‧‧Multiplier

4‧‧‧ weight adjustment device

5‧‧‧Subtractor

6‧‧‧Adder

7‧‧‧ Effective physiological surveillance map

A‧‧‧First Signal Receiver

B‧‧‧second signal receiver

Claims (1)

An architecture for swaying noise technology for removing physiological signals, comprising: a first signal receiver for monitoring a physiological signal of a human body, a measurable physiological signal, and including an eye movement, an electrocardiogram, an electromyogram, Brain wave diagram and light volume change trace, etc.; a second signal receiver for receiving a three-axis shaking signal of the human body, and the second signal receiver includes a three-axis accelerator with an accelerometer, a gyroscope and a strain gauge; an amplifier a module, wherein the input end is connected to the first signal receiver, and the amplifier module comprises an input amplifier, a signal amplifier and an output filter of positive and negative electrodes; and an analog converter, the input end of which is connected to the foregoing amplifier module And a second signal receiver; a multiplier connected to the output of the analog-to-digital converter outputted by the amplifier module and the output of the weight adjustment device, and the multiplier receives the signal of the shaking of the second signal receiver as a reference signal; an adder connected to the output of the aforementioned multiplier; a subtractor connected to the first signal receiver, the analog digital converter and The output of the adder; a weight adjusting means, which is connected the output terminal of the subtracter, the weight adjustment device with a built for calculating a minimum mean square algorithm for adjusting the weights; Thereby, the group constitutes a structure and method for the swaying noise technology for removing the physiological signal.