CN103222864A - Self-adaption electrocardiograph (ECG) detection method and monitoring system thereof - Google Patents

Abstract

The invention discloses an adaptive electrocardiogram detection method and a monitoring system thereof. The detection method comprises the following steps: performing band-pass filtering, absolute value calculation, and low-pass filtering on the output ECG signal of the acquisition module; Threshold to detect the R wave in the low-pass filtered signal. The monitoring system includes: ECG acquisition module, intelligent equipment and remote terminal. The ECG acquisition module includes a sequentially connected ECG preamplifier circuit, 50Hz notch filter, secondary amplifier circuit, anti-aliasing filter circuit, A/D A conversion circuit, a microcontroller and a bluetooth module, the microcontroller is connected with the smart device through the bluetooth module; the smart device is loaded with an adaptive electrocardiogram detection method and connected with the remote terminal through the network. This method is simple and fast, and is suitable for low-end microprocessors. When used in conjunction with its monitoring system, a portable ECG monitoring system can be realized and has the characteristics of multi-function, low power consumption, low cost, and small size.

Description

A self-adaptive electrocardiogram detection method and its monitoring system

technical field

The invention relates to medical device monitoring equipment, and more specifically, to an adaptive electrocardiogram detection method and a monitoring system thereof.

Background technique

Heart disease has always been the number one killer threatening human life, and it is one of the diseases with the highest morbidity and mortality. With the improvement of living standards and health awareness, people hope to monitor the health of the heart at any time and to carry out timely diagnosis and treatment in critical situations. Therefore, the development of a portable wireless ECG monitoring system is of great significance. mHealth is a useful and potentially powerful tool in improving the quality of health care.

At present, many methods popular at home and abroad all have advantages and disadvantages in different degrees to the methods of ECG signal processing. QRS detection based on wavelet transform, neural network method, adaptive filter, hidden Markov model, matched filter, genetic method, Hilbert transform, length and energy conversion, these methods have a common feature, that That is, the complexity of the method is high, which is not conducive to the real-time processing of digital signals, and has high requirements for the microprocessor.

At present, major medical equipment manufacturers and scientific research institutes in China have invested a lot of resources in the research and development of electrocardiographs, and have developed portable ECG instruments with their own characteristics, such as the HC- 201 standard version, and ECG Universal12-LeadPC-Basado Portable ECG from DRE Company of the United States. However, they have not been well popularized. The reason is that there are the following problems:

1. The provided ECG processing function and the recorded ECG information are limited, and it is difficult for doctors to obtain comprehensive ECG information, which reduces the doctor's correct rate of disease diagnosis.

2. Generally, a digital signal processor is used as the core device for ECG data analysis, while data communication, liquid crystal display, real-time clock, program memory, etc. all need to expand special functional devices, so the structure is more complex and the volume is relatively large. At the same time, the power consumption is relatively large, and the price is relatively expensive, which is unbearable for ordinary patients.

3. The supporting monitoring network and ECG data processing center are not yet perfect, and cannot catch up with the pace of the new era of the Internet of Things.

4. In this way, doctors are required to analyze and judge the ECG waveform, and it is difficult to make the electrocardiograph enter the ordinary hundred. There are defects in the current ECG analysis method, and the phenomenon of misjudgment of heart rate exists.

Contents of the invention

The main purpose of the present invention is to propose an adaptive ECG detection method, which enables monitoring a large amount of ECG signal data when using a low-end microprocessor.

In order to achieve the above object, its technical scheme is:

A kind of self-adaptive electrocardiogram detection method, comprises the following steps:

S1. Band-pass filtering the output ECG signal of the acquisition module;

S2. Perform an absolute value calculation on the signal processed in step S1;

S3. Low-pass filtering the signal processed in step S2;

S4. Setting a threshold to detect the R wave in the signal processed in step S3.

Preferably, in the step S1, a band-pass filter of 10-30 Hz is used for band-pass filtering, the transfer function of the band-pass filter is H _bandpass (z), the filter output is Y ₁ (z), and the ECG output from the acquisition module The signal sequence is x[n], and the z-transformation on it is X(z); then there is Y ₁ (z)=H _bandpass (z)X(z).

The bandpass filter is a Butterworth filter;

Let the system function of the Butterworth bandpass filter be: $h_{\mathrm{band}}\left(\mathrm{the\; s}\right)=\frac{1}{B_{\mathrm{no}}\left(\frac{{\mathrm{the\; s}}^2+{\mathrm{\ω}}_l{\mathrm{\ω}}_h}{\mathrm{the\; s}({\mathrm{\ω}}_h-{\mathrm{\ω}}_l)}\right)}$

in ${\mathrm{\ω}}_l={2f}_{\mathrm{the\; s}}\mathrm{the\; tan}\left(\frac{{\mathrm{\πf}}_l}{f_{\mathrm{the\; s}}}\right),$ ${\mathrm{\ω}}_h={2f}_{\mathrm{the\; s}}\mathrm{the\; tan}\left(\frac{{\mathrm{\πf}}_h}{f_{\mathrm{the\; s}}}\right),$ f _l is the low cut-off frequency, f _h is the high cut-off frequency, f _s is the sampling rate, n is the filter order, where B _n (s) is defined as:

n is an even number, $B_{\mathrm{no}}\left(\mathrm{the\; s}\right)=\underset{k=1}{\overset{\frac{\mathrm{no}}{2}}{\mathrm{\Π}}}[{\mathrm{the\; s}}^2-2\mathrm{the\; s}\cos \left(\frac{2k+\mathrm{no}-1}{2\mathrm{no}}\mathrm{\π}\right)+1],$

n is an odd number, $B_{\mathrm{no}}\left(\mathrm{the\; s}\right)=(\mathrm{the\; s}+1)\underset{k=1}{\overset{\frac{\mathrm{no}-1}{2}}{\mathrm{\Π}}}[{\mathrm{the\; s}}^2-2\mathrm{the\; s}\cos \left(\frac{2k+\mathrm{no}-1}{2\mathrm{no}}\mathrm{\π}\right)+1],$

By bilinearly transforming H _band (s), the transfer function of the Butterworth bandpass filter is obtained, namely $h_{\mathrm{bandpass}}\left(z\right)=h_{\mathrm{hand}}\left({2f}_{\mathrm{the\; s}}\frac{z-1}{z+1}\right).$

Preferably, the step S2 further includes discretizing the output signal of the step S1 to obtain a discrete sequence y ₁ [n] corresponding to Y ₁ (z), and performing an absolute value operation on the discrete sequence y ₁ [n], assuming The output sequence is y ₂ [n], then y ₂ [n]=|y ₁ [n]|.

Preferably, a low-pass filter with a cut-off frequency of 5 Hz is adopted in the step S3, and its transfer function is H _lowpass (z), and the output signal y ₂ [n] in the step S2 is z-transformed to obtain Y ₂ (z ), enter low-pass filter, set the filter output as Y ₃ (z), then Y ₃ (z)=H _lowpass (z)Y ₂ (z).

The low-pass filter adopts Butterworth low-pass filter

Let the system function of the Butterworth low-pass filter be

f为截止频率，f_s为采样率，n为滤波器阶数，其中B_n(s)定义为：in

f is the cutoff frequency, f _s is the sampling rate, and n is the filter order, where B _n (s) is defined as:

n is an even number, $B_{\mathrm{no}}\left(\mathrm{the\; s}\right)=\underset{k=1}{\overset{\frac{\mathrm{no}}{2}}{\mathrm{\Π}}}[{\mathrm{the\; s}}^2-2\mathrm{the\; s}\cos \left(\frac{2k+\mathrm{no}-1}{2\mathrm{no}}\mathrm{\π}\right)+1]$

n is an odd number, $B_{\mathrm{no}}\left(\mathrm{the\; s}\right)=(\mathrm{the\; s}+1)\underset{k=1}{\overset{\frac{\mathrm{no}-1}{2}}{\mathrm{\Π}}}[{\mathrm{the\; s}}^2-2\mathrm{the\; s}\cos \left(\frac{2k+\mathrm{no}-1}{2\mathrm{no}}\mathrm{\π}\right)+1]$

By performing bilinear transformation on H _low (s), the transfer function of the Butterworth low-pass filter is obtained, namely $h_{\mathrm{low\; pass}}\left(z\right)=h_{\mathrm{low}}\left({2f}_{\mathrm{the\; s}}\frac{z-1}{z+1}\right).$

选取极大值点的个数为N，阈值

a用于设置平均值比重的阈值，检测R波，对y₃[i]，若y₃[i]>h，则i为滤波后的心电信号序列y₃[n]中R波所在序号。Preferably, the specific implementation of step S4 is as follows: Y ₃ (z) output in step S3 is subjected to z-inverse transformation to obtain its corresponding discrete sequence y ₃ [n], and for the maximum value point of the filtered output sequence, set The maximum value point is i, where y ₃ [i]>y ₃ [i-1] and y ₃ [i]>y ₃ [i+1], set

Select the number of maximum points as N, and the threshold

a is used to set the threshold of the average specific gravity and detect the R wave. For y ₃ [i], if y ₃ [i]>h, then i is the serial number of the R wave in the filtered ECG signal sequence y ₃ [n] .

Another object of the present invention is to provide a monitoring system using the above-mentioned ECG detection method, which has the characteristics of multi-function, low power consumption, low cost and small size.

Its technical solution is:

A monitoring system applying an adaptive electrocardiographic detection method, comprising a sequentially connected electrocardiographic acquisition module, an intelligent device and a remote terminal, and the intelligent device is loaded with an adaptive electrocardiographic detection method.

Preferably, the ECG acquisition module includes a sequentially connected ECG preamplifier circuit, a 50Hz notch filter, a secondary amplifier circuit, an anti-aliasing filter circuit, an A/D conversion circuit, a microcontroller, and a Bluetooth module. The microcontroller is connected with the smart device through the bluetooth module.

Preferably, the ECG collection module further includes a button module and a display module, and the button module and the display module are connected to the microcontroller through a data bus.

Preferably, the microcontroller is a single-chip microcomputer.

Preferably, the smart device is connected to the remote terminal through a network, and the network is a GSM, WIFI, 3G or 4G network.

Preferably, the smart device is a smart phone or computer with bluetooth communication function.

Compared with the prior art, the beneficial effects of the present invention are: the ECG detection method proposed by the present invention is simple and fast, suitable for use with low-end microprocessors, and can realize a portable ECG monitoring system when used in conjunction with its monitoring system And has the characteristics of multi-function, low power consumption, low cost, small size and so on. Moreover, it can monitor the patient's ECG at any time, which greatly guarantees the safety of the patient.

Description of drawings

FIG. 1 is a flowchart of an adaptive ECG detection method.

Fig. 2 is a structural block diagram of an adaptive ECG monitoring system.

FIG. 3 is a diagram showing the effect of the detection method on R wave detection of ECG signals with EMG interference.

Fig. 4 is a diagram showing the effect of the detection method on R-wave detection of ECG signals with power frequency interference.

FIG. 5 is a diagram showing the effect of the detection method on R-wave detection of ECG signals with baseline drift.

FIG. 6 is a diagram showing the effect of the detection method on R-wave detection of ECG signals with step-jump interference.

Fig. 7 is an effect diagram of R wave detection of ECG signals with myoelectricity, power frequency, baseline drift, and step jump interference by the detection method.

Fig. 8 is a diagram showing the detection effect of the detection method of the present invention and the classical algorithm on a clean ECG signal.

Fig. 9 is a diagram showing the detection effect of the detection method of the present invention and the classical algorithm on the disturbed ECG signal.

Fig. 10 is a diagram of the detection effect of the detection method on the R wave of the heart in the present heart electricity monitoring system.

(Curve 1 is the original waveform of the collected ECG signal, Curve 2 is the R wave position detected and marked by this algorithm, Curve 3 and Curve 4 are the display of the process waveform of this algorithm)

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings, but the embodiments of the present invention are not limited thereto.

Fig. 1 is the flowchart of self-adaptive ECG detection method, comprises the following steps:

S1. Band-pass filtering the output ECG signal of the acquisition module;

S2. Perform an absolute value calculation on the signal processed in step S1;

S3. Low-pass filtering the signal processed in step S2;

S4. Setting a threshold to detect the R wave in the signal processed in step S3.

可算出ω_l与ω_h，再代入 $H_{\mathrm{band}}\left(s\right)=\frac{1}{B_n\left(\frac{s^2+{\mathrm{\ω}}_l{\mathrm{\ω}}_h}{s({\mathrm{\ω}}_h-{\mathrm{\ω}}_l)}\right)}$ 可得具体系统函数。In this embodiment, a sampling rate of 512 Hz is used to sample the ECG signal. The band-pass filter in step S1 is a Butterworth filter with a 6th-order pass band of 10 Hz to 30 Hz, the lower cut-off frequency f _l =10 Hz, and the higher Cutoff frequency f _h =30Hz, sampling rate f _s =512Hz, substitute into the formula

ω _l and ω _h can be calculated, and then substituted into $h_{\mathrm{band}}\left(\mathrm{the\; s}\right)=\frac{1}{B_{\mathrm{no}}\left(\frac{{\mathrm{the\; s}}^2+{\mathrm{\ω}}_l{\mathrm{\ω}}_h}{\mathrm{the\; s}({\mathrm{\ω}}_h-{\mathrm{\ω}}_l)}\right)}$ Specific system functions are available.

得到Butterworth带通滤波器的传递函数H_bandpass(z)表达式，进一步对采集到的心电信号序列x[n]进行z变换有X(z)，设滤波输出为Y₁(z)，则有Y₁(z)=H_bandpass(z)X(z)。By bilinearly transforming the H _band (s),

Obtain the expression of the transfer function H _bandpass (z) of the Butterworth bandpass filter, and further carry out z transformation to the collected ECG signal sequence x[n] to obtain X(z), and set the filter output as Y ₁ (z), then There is Y ₁ (z)=H _bandpass (z)X(z).

In the step S2, the absolute value operation is taken, and the z inverse transformation is performed on the filter output Y ₁ (z) of the step S1 to have its corresponding discrete sequence y ₁ [n], and the absolute value operation is performed on the discrete sequence y ₁ [n], assuming The output sequence is y ₂ [n], then y ₂ [n]=|y ₁ [n]|.

Adopt the Butterworth low-pass filter of 6th order cut-off frequency f=5Hz in the step S3, substitute into formula and can calculate the system function H _low (s) of Butterworth low-pass filter, then carry out bilinear transformation to it, Obtain the transfer function H _lowpass (z) of the Butterworth low-pass filter, and perform z-transformation on the signal y ₂ [n] output in step S3 to obtain Y ₂ (z). After low-pass filtering, the filter output is Y ₃ (z), then Y ₃ (z)=H _lowpass (z)Y ₂ (z).

选取极大值点的个数为N，阈值

a为另一阈值，检测R波，对y₃[i]，若y₃[i]>h，则i为滤波后的心电信号序列y₃[n]中R波所在序号。Perform z inverse transformation on Y ₃ (z) output by S3 to obtain its corresponding discrete sequence y ₃ [n]. For the maximum value point of the filtered output sequence, set the maximum value point to i, where y ₃ [i]> y ₃ [i-1] and y ₃ [i]>y ₃ [i+1], let

Select the number of maximum points as N, and the threshold

a is another threshold to detect the R wave. For y ₃ [i], if y ₃ [i]>h, then i is the serial number of the R wave in the filtered ECG signal sequence y ₃ [n].

As shown in Figure 2, a monitoring system based on an adaptive ECG detection method includes: an ECG acquisition module, an intelligent device and a remote terminal, and the ECG acquisition module is composed of an ECG preamplifier circuit, a 50Hz notch filter, It consists of a secondary amplifier circuit, an anti-aliasing filter circuit, an A/D conversion circuit, a microcontroller and a Bluetooth module. Amplifying circuit, anti-aliasing filter circuit, A/D conversion circuit, transmitted to the microcontroller, the microcontroller is connected to the smart device through the bluetooth module; the smart device is loaded with an adaptive ECG detection method and communicates with the remote terminal Connect via network.

The INA332 amplifying circuit with 3V operating voltage is used to amplify the collected ECG signal by 10 times.

The filter circuit adopts an asymmetric double-T active band-stop filter with adjustable Q value, and adjusts the Q value to a suitable position, so that the filter circuit has a 50 Hz notch function. After the relatively high-power interference signal is filtered out, a non-inverting amplifier circuit with a low-pass filter function is used to amplify the electrical signal by 100 times to make the signal reach V level or above.

Using 8-way 10-bit AD conversion module to convert the ECG signal into a digital signal and transmit it to the main controller of MSP430.

Use a professional bluetooth module, which is small in size and low in power consumption, and communicates with MSP430 through a serial port. MSP430 sends the collected ECG data to the smart device through the Bluetooth module.

The preset software on the smart device uses a new method to digitally filter the ECG data and extract the R-wave of the ECG, and finally displays it on the device screen. In this way, the patient's ECG can be monitored at any time through the smart device with Bluetooth function.

The ECG signal of the human body passes through the ECG preamplifier with high impedance and high common mode rejection ability, and is processed by a 50Hz trap circuit to obtain a signal that removes power frequency interference, and then passes through the secondary amplifier with strong amplification capability. , at this time the signal voltage reaches the volt level. In order to make the system satisfy the sampling law, the signal must be added with an anti-aliasing filter before entering the AD. The data sampled by AD is sent to the main control chip MSP430, and the data is packaged by MSP430 and sent to the smart phone through the Bluetooth module. The mobile phone will receive the collected ECG signal and use this method to detect the R wave, so as to calculate the number of heartbeats per minute, and can selectively send the ECG information to the relative's mobile phone or mobile phone through the GSM network or WIFI network. on the computer of the medical staff. If there is a problem with the patient's heartbeat, the smart phone carried by the patient can analyze the ECG signal to judge the situation, send an alarm or send a distress signal like a remote terminal.

The embodiments of the present invention described above are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the claims of the present invention.

Claims (10)

1. a self adaptation electrocardio detection method is characterized in that, may further comprise the steps:

S1. the output ECG signal to acquisition module carries out bandpass filtering;

S2. the computing that takes absolute value of the signal after step S1 being handled;

S3. the signal after step S2 being handled carries out low-pass filtering;

S4. the R ripple in the signal after setting threshold detects step S3 and handles.

2. self adaptation electrocardio detection method according to claim 1 is characterized in that, adopts the band filter of 10～30Hz to carry out bandpass filtering among the described step S1, and the transfer function of band filter is H _Bandpass(z), filtering is output as Y ₁(z), the ECG signal sequence of acquisition module output is x[n], it is carried out the z conversion X (z); Y is then arranged ₁(z)=H _Bandpass(z) X (z).

3. self adaptation cardioelectric monitor method according to claim 2 is characterized in that, described step S2 comprises that also the signal to step S1 output carries out discretization, obtains Y ₁(z) Dui Ying discrete series y ₁[n] is to discrete series y ₁[n] computing that takes absolute value, establishing output sequence is y ₂[n] then has y ₂[n]=| y ₁[n] |.

4. self adaptation cardioelectric monitor method according to claim 3 is characterized in that, adopting cut-off frequency among the described step S3 is the low pass filter of 5Hz, and establishing its transfer function is H _Lowpass(z), to the output signal y among the step S2 ₂[n] carries out the z conversion and obtains Y ₂(z), through low-pass filtering, establish filtering and be output as Y ₃(z), Y then ₃(z)=H _Lowpass(z) Y ₂(z).

5. self adaptation cardioelectric monitor method according to claim 4 is characterized in that the specific implementation of described step S4 is: to the Y of step S3 output ₃(z) carry out the z inverse transformation and obtain its corresponding discrete series y ₃[n], for the maximum point of filtering output sequence, establishing maximum point is i, wherein y ₃[i]〉y ₃[i-1] and y ₃[i]〉y ₃[i+1] establishes

The number of choosing maximum point is N, and threshold value h=a * dN, a are used to be provided with the threshold value of meansigma methods proportion, detects the R ripple, to y ₃[i] is if y ₃[i]〉h, then i is filtered electrocardiosignal sequences y ₃R ripple place sequence number in [n].

6. an application rights requires the monitoring system of the described self adaptation electrocardio of 1-5 detection method, comprise the electrocardiogram acquisition module that connects in turn, smart machine and remote terminal is characterized in that, are mounted with the described self adaptation electrocardio of claim 1-5 detection method on the described smart machine.

7. monitoring system according to claim 6, it is characterized in that, described electrocardiogram acquisition module comprises and connects electrocardio pre-amplification circuit, 50Hz wave trap, second amplifying circuit, anti-aliasing filter circuit, A/D change-over circuit, microcontroller and bluetooth module in turn that described microcontroller is connected with smart machine by bluetooth module.

8. monitoring system according to claim 6 is characterized in that, described electrocardiogram acquisition module also comprises key-press module and display module, and key-press module is connected with microcontroller by data/address bus with display module; Described microcontroller is a single-chip microcomputer.

9. monitoring system according to claim 6 is characterized in that described smart machine is connected with remote terminal by network, and described network is GSM, WIFI, 3G or 4G network.

10. monitoring system according to claim 6 is characterized in that, described smart machine is smart mobile phone or the computer with bluetooth communication function.

Images