CN104856666B - A kind of bioelectrical signals monitoring system based on LabVIEW - Google Patents

Abstract

The invention discloses a kind of bioelectrical signals monitoring system based on LabVIEW, by the bioelectrical signals of the bioelectrical signals of bioelectrical signals extraction equipment extraction multichannel, then parallel processing multichannel, monitored person is monitored using bioelectrical signals observation interface.In specific configuration, the need for monitored person can be according to itself, by the built-in function prestored in signal transacting interface selective calling system, bioelectrical signals are handled in real time, finally shown again by the display box on signal transacting interface, complete the monitoring to being monitored person's physical signs.

Description

A Bioelectrical Signal Monitoring System Based on LabVIEW

technical field

The invention belongs to the technical field of physiological monitoring, and more specifically relates to a bioelectrical signal monitoring system based on LabVIEW.

Background technique

Bioelectrical signals are important characteristic parameters of human health status, and also effectively reflect human brain activity and body movement. Through the analysis and monitoring of human bioelectric signals, the health status of the human body can be accurately grasped, and the operation and control of external equipment can be realized by identifying the characteristics of different bioelectric signals.

At present, the functions of the bioelectrical signal analysis and monitoring system on the market are too single, the signal processing algorithm is relatively simple, and most similar system programs do not have good portability, such as the EEG signal extraction module and the divine mind designed by TI based on the ADS1299 chip The technology is based on the EEG sensor designed by the TGAM chip. Therefore, it is difficult for the above two devices to be effectively promoted in the market.

More specifically, compared with the EEG signal extraction module designed by TI based on the ADS1299 chip, the present invention has the advantages of small size, multiple power supply modes, low thermal power consumption, built-in bioelectrical signal extraction device and more functions. Software display and other advantages. Compared with the EEG sensor designed by Shennian Technology based on the TGAM chip, the present invention has the advantages of low power consumption, stable communication, and multi-channel acquisition.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies of the prior art, provide a bioelectrical signal monitoring system based on LabVIEW, analyze the collected bioelectrical signals through a signal processing algorithm, and realize real-time monitoring of the human body state.

In order to achieve the above-mentioned purpose of the invention, the bioelectric signal monitoring system based on LabVIEW of the present invention is characterized in that, comprising:

A bioelectric signal extraction device, including EEG signal acquisition equipment and a plurality of limb signal acquisition equipment, both of which include spring steel sheets, elastic nylon belts and dry electrode sheets, wherein, in the EEG signal acquisition equipment, at least Including 3 pieces of dry electrode sheets; the dry electrode sheets are attached to the corresponding spring steel sheets by welding, and are stuck on the head and limbs when wearing the bioelectric signal extraction device, and the welding position corresponds to the active area of the bioelectric signal; elastic The nylon belt is located on the spring steel sheet, which can further strengthen the contact tightness between the dry electrode sheet and the human body, and increase the stability of bioelectrical signal collection; when extracting bioelectrical signals, the bioelectrical signal extraction equipment is worn on the head and limbs First, the bioelectrical signals of the brain and limbs are extracted respectively, and then sent to the PCB circuit integration module through the electromagnetic shielding signal line connected to the acquisition device;

Multiple electromagnetic shielding signal lines, adopting multi-layer anti-interference design, the inner layer is the signal transmission line, the outer layer is the rubber layer, and the outer layer of the signal transmission line is the metal wire envelope layer;

A PCB circuit integration module, including EMI filter, preamplifier and analog-to-digital converter; used to receive multiple bioelectric signals collected by bioelectric signal extraction equipment, process multiple bioelectric signals in parallel, and then send them to Zigbee networking communication module;

After the PCB circuit integration module receives the bioelectrical signal, it first filters out the electromagnetic noise with high out-of-band frequency through the EMI filter, and the filtered bioelectrical signal is amplified by the preamplifier and then input to the analog-to-digital converter. Then convert the analog bioelectric signal into a digital signal that is easy to send and receive, and send it to the zigbee networking communication module;

A zigbee networking communication module contains multiple communication ports, which can receive multiple bioelectrical signals sent by the PCB circuit integration module in parallel, and then forward them to the bioelectrical signal monitoring interface to realize multi-receiving and multi-sending network communication;

A signal processing function library, the signal processing interface processes the bioelectric signal by calling the functions in the signal processing function library, and then sends it to the bioelectric signal monitoring interface;

A bioelectrical signal monitoring interface, through the tab control on the bioelectrical signal monitoring interface, select to enter the signal monitoring interface or the signal processing interface;

On the signal monitoring interface, you can set the communication port between the zigbee networking communication module and the bioelectrical signal monitoring interface, and select the signal channel that needs to enter the signal processing interface. During communication, the signal quality of the bioelectrical signal can be displayed on the dashboard Direct observation, the relevant information of the bioelectrical signal displays the bioelectrical signal of the brain and limbs respectively through the 2-D display frame;

On the signal processing interface, the user can selectively call the functions in the signal processing function library according to the needs, and then set according to the called function. If the filter function is selected, the filter frequency band is set; if the notch function is selected, the Determine the power frequency interference frequency band removed by the notch function; if the wavelet function is selected, set the wavelet mother wave type, and select the number of layers of wavelet decomposition; if the neural network function is selected, the neural network classifier will be trained when the function is called Parameters; after the above-mentioned processing of the bioelectrical signal is completed, its relevant information displays the bioelectrical signal of the brain and limbs respectively through the 2-D display frame.

The purpose of the invention of the present invention is achieved like this:

The bioelectric signal monitoring system based on LabVIEW in the present invention extracts multiple bioelectric signals through a bioelectric signal extraction device, then processes the multiple bioelectric signals in parallel, and uses the bioelectric signal monitoring interface to monitor the monitored person. In the specific configuration, the monitored person can selectively call the pre-stored library functions in the system through the signal processing interface according to their own needs, and process the bioelectrical signal in real time, and finally display it through the display box on the signal processing interface to complete Monitoring of the physiological indicators of the monitored persons.

Simultaneously, the bioelectric signal monitoring system based on LabVIEW of the present invention also has the following beneficial effects:

(1) In the present invention, the bioelectric signal extraction device is based on spring steel, elastic nylon belt and dry electrode sheet. Existing similar products mostly use electrode caps and sticky patches to extract bioelectrical signals. Compared with electrode caps, the present invention has the advantages of light weight, small volume, and low cost; compared with sticky patches, the present invention has It can be used repeatedly, the tightness with the human body can be adjusted, and it is easy to put on and take off, etc.;

(2) Through the bioelectric signal monitoring interface, the bioelectric signals of multiple channels can be observed at the same time, and the communication interface can also be manually set, and the communication quality can be monitored in real time at the same time;

(3), the present invention has increased the design of library function; There are multiple signal processing functions pre-stored in the signal processing function library, and the user can realize real-time processing of bioelectrical signals extracted under various environments without programming; secondly, The signal processing function library can be expanded, and the signal processing function written by the user can be written to meet various needs of the user;

(4) In the signal processing interface, the user can selectively call the functions in the signal processing function library to process the bioelectrical signal according to the needs, and the operation process is simple.

Description of drawings

Fig. 1 is a kind of specific embodiment frame diagram of the bioelectrical signal monitoring system based on LabVIEW of the present invention;

Fig. 2 is a structural diagram of a specific embodiment of the bioelectrical signal extraction device shown in Fig. 1;

Fig. 3 is a structural diagram of a specific embodiment of the bioelectrical signal monitoring interface shown in Fig. 1 .

detailed description

Specific embodiments of the present invention will be described below in conjunction with the accompanying drawings, so that those skilled in the art can better understand the present invention. It should be noted that in the following description, when detailed descriptions of known functions and designs may dilute the main content of the present invention, these descriptions will be omitted here.

Example

LabVIEW (Laboratory Virtual Instrumentation Engineering Workbench): Laboratory virtual instrument engineering platform;

Zigbee: a low-power personal area network protocol based on the IEEE802.15.4 standard;

Fig. 1 is a structure diagram of a specific embodiment of the LabVIEW-based bioelectrical signal monitoring system of the present invention.

In this embodiment, as shown in Figure 1, a bioelectrical signal monitoring system based on LabVIEW of the present invention includes: bioelectrical signal extraction device 1, electromagnetic shielding signal line 2, PCB circuit integration module 3, zigbee networking communication Module 4, signal processing function library 5 and bioelectrical signal monitoring interface 6.

As shown in FIG. 2 , the bioelectric signal extraction device 1 further includes an EEG signal acquisition device and a plurality of limb signal acquisition devices, wherein the left side of FIG. 2 is the EEG signal collection device, and the right side of FIG. 2 is the limb signal collection device. Both collection devices include a spring steel sheet 1.1, an elastic nylon belt 1.2 and dry electrode sheets 1.3; wherein, in the EEG signal collection device, there are at least three dry electrode sheets 1.3. The dry electrode sheet 1.3 is attached to the corresponding spring steel sheet 1.1 by welding, which can be well stuck on the head and limbs, and the welding position corresponds to the active area of the bioelectric signal; the elastic nylon band 1.2 can further strengthen the dry electrode sheet 1.3 Close contact with the human body to increase the stability of bioelectrical signal acquisition; when extracting bioelectrical signals, wear the EEG signal acquisition equipment on the head, and wear multiple limb signal acquisition equipment on the wrist and big arm respectively and forearm etc. to extract the bioelectrical signals from the brain and limbs respectively; an electromagnetic shielding signal line 2 is connected to each acquisition device, and the collected bioelectrical signals are then sent to the PCB circuit through the electromagnetic shielding signal line 2 integrated module;

The electromagnetic shielding signal line 2 adopts a multi-layer anti-interference design, wherein the inner layer is a signal transmission line, the outer layer is a rubber layer, and the periphery of the signal transmission line is a metal wire envelope layer, thereby realizing electromagnetic shielding to a certain extent and enhancing electrical protection. The anti-interference ability of the signal increases the effective transmission distance;

The PCB circuit integration module 3 includes an EMI filter, a preamplifier, and an analog-to-digital converter; it is used to receive multiple bioelectric signals collected by the bioelectric signal extraction device 1, and process the multiple bioelectric signals in parallel, and then Send to zigbee networking communication module 4;

After the PCB circuit integration module 3 receives the bioelectric signal, it first filters out the electromagnetic noise with high out-of-band frequency through the EMI filter, and the filtered bioelectric signal is amplified by the preamplifier and then input to the analog-to-digital converter. The device converts the analog bioelectric signal into a digital signal that is easy to send and receive, and sends it to the zigbee networking communication module 4;

The zigbee networking communication module contains multiple ports, which can receive multiple bioelectrical signals sent by the PCB circuit integration module 3 in parallel, and then forward them to the bioelectrical signal monitoring interface 6, realizing multi-receiving and multi-sending network communication;

The signal processing function library 5, the signal processing interface processes the bioelectric signal by calling the function in the signal processing function library, and then sends it to the bioelectric signal monitoring interface 6; in this embodiment, the signal processing function library includes filter functions, Various functions such as notch function, wavelet function and artificial neural network function;

As shown in Figure 3, the bioelectrical signal monitoring interface 6 includes a signal monitoring interface and a signal processing interface; through the tab control 6.1 on the bioelectrical signal monitoring interface, select to enter the signal monitoring interface or signal processing interface, wherein, Figure 3 ( a) is the signal monitoring interface, and Fig. 3(b) is the signal processing interface;

On the signal monitoring interface, you can set the communication port between the zigbee networking communication module 4 and the bioelectrical signal monitoring interface 6 at 6.2, and select the signal channel that needs to enter the signal processing interface. During communication, the signal quality of the bioelectrical signal It can be directly observed on the instrument panel 6.3, and the relevant information of the bioelectric signal displays the bioelectric signals of the brain and limbs through the 2-D display box, that is, the bioelectric signals of the brain and limbs are displayed at 6.5 and 6.6 respectively. Signal;

In this embodiment, as shown in Figure 3(a), the instrument panel 6.3 is divided into 0-10 scales. When the pointer on the instrument panel 6.3 points to a certain scale, it indicates the degree to which the bioelectric signal is disturbed by the external environment , where "0" means ideal interference-free communication, and "10" means severe interference from the external environment;

On the signal processing interface, the signal channels waiting to be processed are displayed through the indicator light 6.10, and the user can selectively call the functions in the signal processing function library according to the needs, that is, one or more functions can be called to control the biological signals in the signal channel. The electrical signal is processed. When a function is called, the corresponding function is set. The specific setting process is as follows:

1), the filter function in the signal processing function library can be called at 6.7, and the filter frequency band can be set

In this embodiment, the filter in the library function is a Butterworth filter (Butterworth), and the user can set the filter frequency band according to the needs, extract the specific signal frequency band that you want to monitor in depth, or use it to filter out low-frequency clutter and high-frequency invalid waves; for example, if you want to observe the alpha-band (8Hz-12Hz) brainwaves that reflect motor imagination in the EEG signal alone, you can use the filter function to filter out the EEG in this frequency band on the basis of the original EEG signal Signal;

2), at 6.8, the notch function in the signal processing function library can be called, and the power frequency interference frequency band removed by the notch function can be set

In this embodiment, the notch is realized by an FIR-based band-stop filter, and the notch frequency is 50 Hz or 60 Hz;

3) In 6.9, the wavelet function in the signal processing function library can be called, the type of wavelet mother wave can be set, and the number of layers of wavelet decomposition can be selected

In this embodiment, the mallat algorithm is used to construct the wavelet function, and its type is haar or db wavelet mother wave, and by setting the number of wavelet decomposition layers, the original signal is subjected to different degrees of wavelet smoothing processing, in principle, the more the number of decomposition layers , the better the smoothing effect is, but some useful feature information may be lost. Here, the decomposition level is selected as 3 layers;

4) The system can also call the neural network classification function in the signal processing function library. The parameters of the neural network classifier do not need to be set, but need to be trained. Because the reference energy of different human body bioelectric signals is quite different, so when calling the neural network classification function Before, the neural network classifier parameters need to be trained, the training steps are as follows:

S1: During the k-th training and learning, extract a large number of bioelectrical signals from the monitored person as training sample data, marked as: x(k)=(x ₁ (k), x ₂ (k),...,x _n (k)), where k=1,2,...,m, m∈M;

S2: During the serial port communication at 38400 baud rate, send the training sample data to the signal processing interface; at this time, the signal processing interface defaults to the neural network training mode;

S3: Complete the parameter training of the neural network classifier

The signal processing interface reads the training sample data, and according to the sample data, sends the neural network classifier modeling parameters under the serial port communication with a baud rate of 38400, and sets the neural network model parameters, including the number n of neurons in the input layer, the number of neurons in the hidden layer The number of units p, the number of neurons in the output layer q, the allowable error ε and the maximum number of learning times M;

Calculated by the software in step S2:

Hidden layer input:

Hidden layer output: ho _h (k)=f(hi _h (k)) h=1,2,...,p;

Output layer input:

Output layer output: yo _o (k)=f(yi _o (k)) o=1,2,...q;

Error function:

Among them, b _h and b _o are constant vectors, and the vector length is the same as the number of hidden layer/output layer neurons; who _ho is the connection weight of the hidden layer; w _ih is the connection weight of the input layer; f( ) is the derivable The sigmoid function, net＝x ₁ w ₁ +x ₂ w ₂ +...+x _n w _n , where x _n and w _n are the input and connection weights of f( ) respectively;

Suppose the expected output: d _o (k)=(d ₁ (k),d ₂ (k),...,d _q (k)), according to the output layer output and the expected output, get the error function, and then calculate The partial derivative of the error function with respect to the neurons in the output layer and the neurons in the hidden layer, namely:

The partial derivative of the error function with respect to the input to the output layer:

in, as a substitute symbol, that is, use -δ _o (k) to replace -(d _o (k)-yo _o (k))f′(yi _o (k));

The partial derivative of the output layer input to the hidden layer connection weights:

The partial derivative of the error function to the hidden layer connection weights:

The partial derivative of the error function with respect to the hidden layer input:

The partial derivative of the error function with respect to the input layer connection weights:

The partial derivative of the hidden layer input to the input layer connection weights:

According to the above partial derivative equation, modify the connection weights of the hidden layer and the connection weights of the input layer

Hidden layer connection weights:

Input layer connection weights:

μ, η are constants, are all set to 0.01 in the present embodiment; is the hidden layer connection weight obtained during the kth training and learning, It is the input layer connection weight obtained during the kth training and learning; at this time, the number of times m of training and learning that has been completed will increase once;

Calculate the global error during the first k training sessions, the global error:

Determine whether the global error reaches the preset accuracy or the number of learning times m reaches the upper limit M. If so, the training of the parameters of the neural network classifier ends; otherwise, return to step S1 for the k+1th training and learning.

Among them, the neural network classifier cannot be used for different people, or even for the same person in different states. Therefore, after the signal processing interface is completely exited, the training results will be automatically cleared, and the next time it is used, it needs to be collected again. training data for training;

After the above-mentioned processing of bioelectrical signals is completed, the relevant information displays the bioelectrical signals of the brain and limbs respectively through the 2-D display frame, that is, the bioelectrical signals of the brain and limbs are displayed at 6.11 and 6.12 respectively.

Although the illustrative specific embodiments of the present invention have been described above, so that those skilled in the art can understand the present invention, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, As long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are included in the protection list.

Claims (5)

1. A bioelectric signal monitoring system based on LabVIEW, characterized in that, comprising: A bioelectrical signal extraction device, including an EEG signal acquisition device and a plurality of limb signal acquisition devices, the EEG signal acquisition device and a plurality of limb signal acquisition devices all include a spring steel sheet, an elastic nylon belt and a dry electrode sheet, wherein, in The EEG signal acquisition equipment includes at least 3 pieces of dry electrode sheets; the dry electrode sheets are attached to the corresponding spring steel sheets by welding, and are stuck on the head and limbs when wearing the bioelectric signal extraction equipment, and the welding position is the same as the bioelectric signal extraction equipment. Corresponding to the active area of the signal; the elastic nylon belt is located on the spring steel sheet, which can further strengthen the contact tightness between the dry electrode sheet and the human body, and increase the stability of bioelectrical signal acquisition; when extracting bioelectrical signals, wear the bioelectrical signal extraction equipment On each part of the head and limbs, the bioelectrical signals of the brain and limbs are respectively extracted, and then sent to the PCB circuit integration module through the electromagnetic shielding signal line connected to the acquisition device; Electromagnetic shielding signal line adopts multi-layer anti-interference design, the inner layer is the signal transmission line, the outer layer is the rubber layer, and the outer layer of the signal transmission line is the metal wire envelope layer; PCB circuit integration module, including EMI filter, preamplifier and analog-to-digital converter; used to receive multiple bioelectric signals collected by bioelectric signal extraction equipment, process multiple bioelectric signals in parallel, and then send them to zigbee Network communication module; After the PCB circuit integration module receives the bioelectrical signal, it first filters out the electromagnetic noise with high out-of-band frequency through the EMI filter, and the filtered bioelectrical signal is amplified by the preamplifier and then input to the analog-to-digital converter. Then convert the analog bioelectric signal into a digital signal that is easy to send and receive, and send it to the zigbee networking communication module; The zigbee networking communication module contains multiple communication ports, which can receive multiple bioelectrical signals sent by the PCB circuit integration module in parallel, and then forward them to the bioelectrical signal monitoring interface to realize multi-receiving and multi-sending network communication; The signal processing function library, the signal processing interface processes the bioelectrical signal by calling the functions in the signal processing function library, and then sends it to the bioelectrical signal monitoring interface; Bioelectrical signal monitoring interface, including signal monitoring interface and signal processing interface; through the tab control on the bioelectrical signal monitoring interface, choose to enter the signal monitoring interface or signal processing interface; Wherein, the signal monitoring interface includes a communication port, an instrument panel, a first 2-D display frame, and a first bioelectrical signal channel; For communication, select the signal channel that needs to enter the signal processing interface through the first bioelectrical signal channel. During communication, directly observe the signal quality of the bioelectrical signal through the deflection scale of the pointer on the instrument panel, and finally pass the first 2-D The display box displays the bioelectric signals of the brain and limbs separately; The signal processing interface includes a second bioelectrical signal channel and a second 2-D display frame; on the signal processing interface, the signal channel that needs to enter the signal processing interface is selected through the second bioelectrical signal channel. It is necessary to selectively call the function in the signal processing function library, and then set it according to the called function. If the filter function is selected, the filtered frequency band is set; if the notch function is selected, the frequency band removed by the notch function is set. If the wavelet function is selected, set the wavelet mother wave type and select the number of layers of wavelet decomposition; if the neural network function is selected, the parameters of the neural network classifier will be trained when the function is called; the bioelectrical signal passes through the After the function processing in the signal processing function library is completed, the bioelectric signals of the brain and limbs among the processed bioelectric signals are displayed separately through the second 2-D display frame. 2. The bioelectrical signal monitoring system based on LabVIEW according to claim 1, wherein the signal processing function library at least includes: filter function, notch function, wavelet function and artificial neural network function. 3. The bioelectrical signal monitoring system based on LabVIEW according to claim 1, wherein the instrument panel is divided into 0--10 scales, and when the pointer on the instrument panel points to a certain scale, it means that the biological The degree to which electrical signals are interfered by the external environment, where "0" indicates ideal interference-free communication, and "10" indicates severe external environment interference. 4. the bioelectrical signal monitoring system based on LabVIEW according to claim 1, is characterized in that, described training neural network classifier parameter method is: (4.1) During the k-th training and learning, a large number of bioelectrical signals are extracted from the monitored person as training sample data, marked as: x(k)=(x ₁ (k), x ₂ (k),...,x _n (k)), where k=1,2,...,m, m∈M; (4.2) During serial communication at 38400 baud rate, send the training sample data to the signal processing interface; (4.3) Complete the parameter training of the neural network classifier The signal processing interface reads the training sample data. According to the sample data, the neural network classifier modeling parameters are sent under the serial port communication at 38400 baud rate, and the neural network model parameters are set, including the number n of neurons in the input layer, the number of neurons in the hidden layer The number of units p, the number of neurons in the output layer q, the allowable error ε and the maximum number of learning times M; By calculation: Hidden layer input: Hidden layer output: ho _h (k)=f(hi _h (k))h=1,2,...,p; Output layer input: Output layer output: yo _o (k)=f(yi _o (k))o=1,2,...q; Error function: Among them, f() is a derivable sigmoid function; Suppose the expected output: d _o (k)=(d ₁ (k),d ₂ (k),…,d _q (k)), obtain the error function according to the output layer output and the expected output, and then calculate the error function For b _h and b _o are constant vectors, the length of the vector is the same as the number of neurons in the hidden layer/output layer; who _ho is the connection weight of the hidden layer; w _ih is the connection weight of the input layer; The partial derivatives of layer neurons, namely: The partial derivative of the error function with respect to the input to the output layer: The partial derivative of the output layer input to the hidden layer connection weights: The partial derivative of the error function to the hidden layer connection weights: The partial derivative of the error function with respect to the hidden layer input: is a substitute symbol; The partial derivative of the error function with respect to the input layer connection weights: The partial derivative of the hidden layer input to the input layer connection weights: According to the above partial derivative equation, modify the connection weights of the hidden layer and the connection weights of the input layer Hidden layer connection weights: Input layer connection weights: μ and η are constants, is the hidden layer connection weight obtained during the kth training and learning, is the input layer connection weight obtained during the kth training and learning; Calculate the global error during the first k training sessions, the global error: Determine whether the global error reaches the preset accuracy or the number of learning times m reaches the upper limit M. If so, the training of the neural network classifier parameters ends; otherwise, return to step (4.1) for the k+1th training and learning. 5. The bioelectrical signal monitoring system based on LabVIEW according to claim 1 or 4, wherein the neural network classifier will automatically clear the training results this time after the signal processing interface completely exits, and wait for the next time When in use, the training data needs to be collected again for training.