CN113100776B - Fatigue monitoring system and method for fusing myoelectricity and electrocardiosignal - Google Patents
Fatigue monitoring system and method for fusing myoelectricity and electrocardiosignal Download PDFInfo
- Publication number
- CN113100776B CN113100776B CN202110424867.XA CN202110424867A CN113100776B CN 113100776 B CN113100776 B CN 113100776B CN 202110424867 A CN202110424867 A CN 202110424867A CN 113100776 B CN113100776 B CN 113100776B
- Authority
- CN
- China
- Prior art keywords
- module
- driver
- electrocardiosignal
- signal
- fatigue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/18—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/20—Workers
- A61B2503/22—Motor vehicles operators, e.g. drivers, pilots, captains
Abstract
The invention discloses a fatigue monitoring system and method for fusing myoelectricity and electrocardiosignals, comprising a surface myoelectricity electrode module, a surface myoelectricity signal acquisition and conversion module, a surface myoelectricity signal data processing module, an electrocardiosignal acquisition and conversion module, an electrocardiosignal data processing module and a nonlinear support vector machine algorithm data fusion module; firstly, collecting a surface electromyographic signal of a driver, then carrying out signal amplification, filtering and A/D conversion, and then analyzing to obtain characteristic parameters of the surface electromyographic signal of the driver; collecting an electrocardiosignal of a driver, amplifying, filtering and A/D converting, and analyzing to obtain characteristic parameters of the electrocardiosignal of the driver; and carrying out feature layer fusion on the surface electromyographic signal feature parameters and the electrocardiosignal feature parameters of the driver to obtain a fatigue feature vector, and judging the fatigue condition of the driver by judging whether the feature vector accords with the fatigue feature. The invention reduces the risk of fatigue driving of the driver.
Description
Technical Field
The invention belongs to the technical field of biomedical electricity, and particularly relates to a fatigue monitoring system and method.
Background
Along with the economic development and the improvement of the life quality of people, the transportation industry is developed vigorously. Many new people are put into the traffic industry, but life work is irregular, drivers are inevitably dilemma when driving for a long time, and traffic accidents can be caused by inattention in a moment. Analysis of current traffic accidents shows that 75% of traffic accidents are due to human causes, and that the traffic accidents caused by fatigue driving account for 21% of the traffic accidents. The harm of fatigue driving is obvious, and the importance of preventing fatigue driving is self-evident. Therefore, fatigue monitoring systems employing myoelectric or electrocardiographic signals are becoming increasingly important. The invention patent 201610264968.4 discloses a fatigue detection system and a method based on a video intelligent algorithm, wherein the system detection comprises the steps of judging whether a worker is on duty or not and judging the opening and closing degree of eyes and mouths of the worker, but the fusion method adopts simple weighted average calculation, and is difficult to accurately detect in the traffic driving field with changeable environment and human behaviors. The invention patent 201521125712.2 discloses a driver fatigue detection system based on surface myoelectricity technology, wherein fatigue detection terminal equipment is designed into a wearable head ring, and the wearable head ring comprises an on/off button, a surface myoelectricity detection electrode, a data processing and controller, a vibration motor and a wireless transmitting device, but a single bioelectric signal obviously cannot meet the requirements of high-precision and error-free detection of variable fatigue information.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a fatigue monitoring system and a method for fusing myoelectricity and electrocardiosignals, wherein the fatigue monitoring system comprises a surface myoelectricity electrode module, a surface myoelectricity signal acquisition and conversion module, a surface myoelectricity signal data processing module, an electrocardiosignal acquisition and conversion module, an electrocardiosignal data processing module and a nonlinear support vector machine algorithm data fusion module; firstly, collecting a surface electromyographic signal of a driver, then carrying out signal amplification, filtering and A/D conversion, and then analyzing to obtain characteristic parameters of the surface electromyographic signal of the driver; collecting an electrocardiosignal of a driver, amplifying, filtering and A/D converting, and analyzing to obtain characteristic parameters of the electrocardiosignal of the driver; and carrying out feature layer fusion on the surface electromyographic signal feature parameters and the electrocardiosignal feature parameters of the driver to obtain a fatigue feature vector, and judging the fatigue condition of the driver by judging whether the feature vector accords with the fatigue feature. The invention reduces the risk of fatigue driving of the driver.
The technical scheme adopted for solving the technical problems is as follows:
the fatigue monitoring system for fusing myoelectricity and electrocardiosignals comprises a surface myoelectricity electrode module, a surface myoelectricity signal acquisition and conversion module, a surface myoelectricity signal data processing module, an electrocardiosignal acquisition and conversion module, an electrocardiosignal data processing module and a nonlinear support vector machine algorithm data fusion module;
the surface electromyographic signal electrode module collects surface electromyographic signals of a driver and sends the collected surface electromyographic signals of the driver to the surface electromyographic signal collection and conversion module; the surface electromyographic signal acquisition and conversion module is used for carrying out signal amplification, filtering and A/D conversion on the surface electromyographic signal of the driver and sending an output signal to the surface electromyographic signal data processing module; the surface electromyographic signal data processing module performs time domain analysis and frequency domain analysis on the input data to obtain characteristic parameters of the surface electromyographic signal of the driver;
the electrocardiosignal acquisition and conversion module acquires an electrocardiosignal of a driver, amplifies, filters and performs A/D conversion, and sends an output signal to the electrocardiosignal data processing module; the electrocardiosignal data processing module performs time domain analysis and frequency domain analysis on the input data to obtain characteristic parameters of the electrocardiosignal of the driver;
the surface electromyographic signal characteristic parameters and the electrocardiosignal characteristic parameters of the driver are sent to a nonlinear support vector machine algorithm data fusion module; and the nonlinear support vector machine algorithm data fusion module fuses the surface electromyographic signal characteristic parameters and the electrocardiosignal characteristic parameters of the driver into fatigue characteristic vectors by characteristic layers, and judges the fatigue condition of the driver by judging whether the characteristic vectors accord with the fatigue characteristics.
Further, the surface electromyographic signal electrode module adopts a bipolar electrode plate to measure the surface electromyographic signal, and a reference electrode is added between the two bipolar electrode plates to offset the common mode component of the signals acquired by the two bipolar electrode plates and keep a differential mode part.
Further, the surface electromyographic signal acquisition and conversion circuit module comprises a sensor module, an ADC conversion circuit and a terminal module;
the sensor module comprises an amplifying module, a filtering module and a notch processing module, and is used for amplifying, filtering and eliminating power frequency interference on the collected surface electromyographic signals of the driver;
the ADC conversion circuit converts the surface electromyographic signals processed by the sensor module from analog signals to digital signals and sends the digital signals to the terminal module, and the ADC conversion circuit is realized by adopting an Arduino circuit board;
the terminal module records and visualizes the surface electromyographic signals through software.
Further, the surface electromyographic signal data processing module performs time domain analysis and frequency domain analysis on the digitized surface electromyographic signal to obtain characteristic parameters of the electromyographic signal, including: integrated myoelectric value, root mean square value, average power frequency, median frequency.
Further, the electrocardiosignal acquisition and conversion module comprises an LED infrared light device, a sensing acquisition module, a main control module and an upper computer module;
the LED infrared light device is fixed at the index finger of a driver, and converts an optical signal into an electric signal by utilizing a photoelectric volume method;
the sensing acquisition module comprises an amplification module, a filtering module and a notch processing module, and is used for amplifying, filtering, eliminating baseline drift interference, eliminating power frequency interference and eliminating myoelectric signal interference on the acquired driver electrocardiosignals;
the main control module receives the electrocardiosignals processed by the sensing acquisition module, performs A/D conversion on the electrocardiosignals, and transmits the electrocardiosignals to the upper computer module;
the upper computer module comprises an electrocardiosignal processing and calculating module, a heart rate recording module and a heart rate real-time display module, wherein the electrocardiosignal processing and calculating module calculates a real-time heart rate value through a time domain analysis algorithm, the real-time heart rate value is stored in the heart rate recording module and is imported into the heart rate signal real-time display module, a window is created by using an upper computer screen, and the change condition of the heart rate value is displayed in real time.
Further, the electrocardiosignal data processing module adopts a differential threshold value to detect QRS waves, the QRS waves which are detected by mistake, missed and abnormal are deleted from the R-R interval sequence to form an N-N interval sequence, and the N-N interval sequence is used for carrying out time domain analysis and frequency domain analysis of HRV to obtain characteristic parameters of the electrocardiosignal.
Further, the nonlinear support vector machine algorithm data fusion module receives the electrocardiosignal characteristic parameters and the electromyographic characteristic parameters, and obtains a final decision boundary by adopting a support vector machine algorithm according to the fatigue state judged by the electrocardiosignal characteristic parameters or the electromyographic characteristic parameters, and judges that the decision boundary is in a non-fatigue state and the decision boundary is out of a fatigue state.
The surface electromyographic signal acquisition method specifically comprises the following steps:
step 1: the Arduino circuit board is connected with the terminal module, and digital signals are transmitted to the terminal module while power is supplied to the myoelectric signal acquisition module and the ADC conversion circuit;
step 2: according to the property of the muscle electric signals, the acquisition frequency of the terminal module is adjusted to be 10kHz;
step 3: removing the cuticle of the skin of the muscular abdomen of the driver, wiping the cuticle with alcohol, fixing the bipolar electrode plate on the muscular abdomen of the driver, and enabling the long axis of the bipolar electrode plate to be parallel to the long axis direction of muscle fibers;
step 4: and setting a terminal module, displaying the electromyographic signal image in real time and deriving electromyographic signal data in real time.
The surface electromyographic signal data processing method specifically comprises the following steps:
step 1: performing fast Fourier transform on the electromyographic signals processed by the surface electromyographic signal acquisition and conversion circuit module to obtain amplitude after Fourier transform;
step 2: calculating an integral myoelectricity value, a root mean square value, an average power frequency and a median frequency;
step 3: judging whether the driver is tired or not according to the trend of the median frequency, and if the median frequency is reduced, judging that the driver is tired, and if the median frequency is reduced, judging that the driver is not tired, and if the median frequency is increased, judging that the driver is not tired.
An electrocardiosignal data processing method specifically comprises the following steps:
step 1: setting electrocardiosignal acquisition frequency and sampling time to obtain the number of samples;
step 2: the R-wave threshold is obtained by DIFF (i) =f (i+1) -f (i-1) +2*f (i+2) -2*f (i-2); where f (-) represents sample data, i represents data values in samples, DIFF (i) =zeros (1, samples) represents Double zero-like matrices that generate a 1 sample number;
step 3: obtaining R-R interval data, and calculating:
wherein N represents the total number of data in the acquired sample,represents the average value of the time difference between adjacent main wave peaks, RR i Representing the time difference between the ith adjacent main wave crest, wherein SDNN represents the R-R interval standard deviation;
obtaining SDNN;
step 4: and analyzing the SDNN to judge the fatigue degree of the driver, wherein the value of the SDNN is positively related to the fatigue degree.
The beneficial effects of the invention are as follows:
1. the differential surface electromyographic signal acquisition electrodes are arranged on the surface electromyographic signal acquisition and conversion circuit, so that the invention is noninvasive and convenient.
2. The invention has the following characteristics in electrocardiosignal acquisition: the sensing acquisition module is provided with a digital filtering module and an LED infrared light device, the LED infrared light device can convert an optical signal into an electric signal, and the digital filtering module can amplify, filter and digital-to-analog convert the initial electric signal; the heart rate information obtained by the heart rate recording module is stored in a CSV file to be conveniently called by a follow-up fatigue algorithm, and the heart rate signal real-time display module provides a display area so that a user can monitor the running state of the equipment through visual information.
3. The invention is characterized in that the bioelectric signals comprise the electrocardiosignals and the electromyographic signals, and the two bioelectric signals with fatigue characteristics are fused, so that the accuracy and the stability of fatigue monitoring are improved.
4. The invention adopts the feature layer fusion, and compared with the data layer fusion, the bandwidth requirement is relatively lower, thereby reducing the requirement on hardware parts.
5. The method adopts the algorithm of the nonlinear support vector machine, improves the accuracy of the judging result, reduces the probability of misjudging whether the driver is tired by equipment, and reduces the risk of tired driving of the driver.
Drawings
Fig. 1 is a schematic structural diagram of a fatigue monitoring system fusing myoelectricity and electrocardiosignals.
Fig. 2 is a schematic diagram of a surface electromyographic signal electrode module structure according to the invention.
Fig. 3 is a schematic structural diagram of a surface electromyographic signal acquisition and conversion module of the invention.
Fig. 4 is a flow chart of the surface electromyographic signal acquisition processing of the invention.
Fig. 5 is a schematic structural diagram of an electrocardiosignal acquisition and conversion module of the invention.
Fig. 6 is a schematic diagram of a surface electromyographic signal and electrocardiosignal data processing module of the invention.
FIG. 7 is a schematic diagram of a nonlinear support vector machine algorithm data fusion template structure of the present invention.
FIG. 8 is a graph of the test results of the support vector machine on fatigue data set based on Gaussian kernel function of the present invention.
Detailed Description
The invention will be further described with reference to the drawings and examples.
In order to overcome the problems of high error and low stability caused by a single bioelectric signal in fatigue monitoring of a driver and a rough and simple data fusion mode, the invention provides a myoelectricity and electrocardio-integrated fatigue detection system and a myoelectricity and electrocardio-integrated fatigue detection method.
As shown in fig. 1, the fatigue monitoring system for fusing myoelectricity and electrocardiosignals comprises a surface myoelectricity electrode module, a surface myoelectricity signal acquisition and conversion module, a surface myoelectricity signal data processing module, an electrocardiosignal acquisition and conversion module, an electrocardiosignal data processing module and a nonlinear support vector machine algorithm data fusion module;
the surface electromyographic signal electrode module collects surface electromyographic signals of a driver and sends the collected surface electromyographic signals of the driver to the surface electromyographic signal collection and conversion module; the surface electromyographic signal acquisition and conversion module is used for carrying out signal amplification, filtering and A/D conversion on the surface electromyographic signal of the driver and sending an output signal to the surface electromyographic signal data processing module; the surface electromyographic signal data processing module performs time domain analysis and frequency domain analysis on the input data to obtain characteristic parameters of the surface electromyographic signal of the driver;
the electrocardiosignal acquisition and conversion module acquires an electrocardiosignal of a driver, amplifies, filters and performs A/D conversion, and sends an output signal to the electrocardiosignal data processing module; the electrocardiosignal data processing module performs time domain analysis and frequency domain analysis on the input data to obtain characteristic parameters of the electrocardiosignal of the driver;
the surface electromyographic signal characteristic parameters and the electrocardiosignal characteristic parameters of the driver are sent to a nonlinear support vector machine algorithm data fusion module; and the nonlinear support vector machine algorithm data fusion module fuses the surface electromyographic signal characteristic parameters and the electrocardiosignal characteristic parameters of the driver into fatigue characteristic vectors by characteristic layers, and judges the fatigue condition of the driver by judging whether the characteristic vectors accord with the fatigue characteristics.
Further, the surface electromyographic signal electrode module adopts a bipolar electrode plate to measure the surface electromyographic signal, and a reference electrode is added between the two bipolar electrode plates to offset the common mode component of the signals acquired by the two bipolar electrode plates and keep a differential mode part.
Further, the surface electromyographic signal acquisition and conversion circuit module comprises a sensor module, an ADC conversion circuit and a terminal module;
the sensor module comprises an amplifying module, a filtering module and a notch processing module, and is used for amplifying, filtering and eliminating power frequency interference on the collected surface electromyographic signals of the driver;
the ADC conversion circuit converts the surface electromyographic signals processed by the sensor module from analog signals to digital signals and sends the digital signals to the terminal module, and the ADC conversion circuit is realized by adopting an Arduino circuit board;
the terminal module records and visualizes the surface electromyographic signals through software.
Further, the surface electromyographic signal data processing module performs time domain analysis and frequency domain analysis on the digitized surface electromyographic signal to obtain characteristic parameters of the electromyographic signal, including: integrated myoelectric value, root mean square value, average power frequency, median frequency.
Further, the electrocardiosignal acquisition and conversion module comprises an LED infrared light device, a sensing acquisition module, a main control module and an upper computer module;
the LED infrared light device is fixed at the index finger of a driver, and converts an optical signal into an electric signal by utilizing a photoelectric volume method;
the sensing acquisition module comprises an amplification module, a filtering module and a notch processing module, and is used for amplifying, filtering, eliminating baseline drift interference, eliminating power frequency interference and eliminating myoelectric signal interference on the acquired driver electrocardiosignals;
the main control module receives the electrocardiosignals processed by the sensing acquisition module, performs A/D conversion on the electrocardiosignals, and transmits the electrocardiosignals to the upper computer module;
the upper computer module comprises an electrocardiosignal processing and calculating module, a heart rate recording module and a heart rate real-time display module, wherein the electrocardiosignal processing and calculating module calculates a real-time heart rate value through a time domain analysis algorithm, the real-time heart rate value is stored in the heart rate recording module and is imported into the heart rate signal real-time display module, a window is created by using an upper computer screen, and the change condition of the heart rate value is displayed in real time.
Further, the electrocardiosignal data processing module adopts a differential threshold value to detect QRS waves, the QRS waves which are detected by mistake, missed and abnormal are deleted from the R-R interval sequence to form an N-N interval sequence, and the N-N interval sequence is used for carrying out time domain analysis and frequency domain analysis of HRV to obtain characteristic parameters of the electrocardiosignal.
Further, the nonlinear support vector machine algorithm data fusion module receives the electrocardiosignal characteristic parameters and the electromyographic characteristic parameters, and obtains a final decision boundary by adopting a support vector machine algorithm according to the fatigue state judged by the electrocardiosignal characteristic parameters or the electromyographic characteristic parameters, and judges that the decision boundary is in a non-fatigue state and the decision boundary is out of a fatigue state.
Further, the bipolar electrode plate substrate is made of silver, and the surface of the bipolar electrode plate substrate is plated with silver chloride.
The surface electromyographic signal acquisition method specifically comprises the following steps:
step 1: the Arduino circuit board is connected with the terminal module, and digital signals are transmitted to the terminal module while power is supplied to the myoelectric signal acquisition module and the ADC conversion circuit;
step 2: according to the property of the muscle electric signals, the acquisition frequency of the terminal module is adjusted to be 10kHz;
step 3: removing the cuticle of the skin of the muscular abdomen of the driver, wiping the cuticle with alcohol, fixing the bipolar electrode plate on the muscular abdomen of the driver, and enabling the long axis of the bipolar electrode plate to be parallel to the long axis direction of muscle fibers;
step 4: and setting a terminal module, displaying the electromyographic signal image in real time and deriving electromyographic signal data in real time.
The surface electromyographic signal data processing method specifically comprises the following steps:
step 1: performing fast Fourier transform on the electromyographic signals processed by the surface electromyographic signal acquisition and conversion circuit module to obtain amplitude after Fourier transform;
step 2: calculating an integral myoelectricity value, a root mean square value, an average power frequency and a median frequency;
step 3: judging whether the driver is tired or not according to the trend of the median frequency, and if the median frequency is reduced, judging that the driver is tired, and if the median frequency is reduced, judging that the driver is not tired, and if the median frequency is increased, judging that the driver is not tired.
An electrocardiosignal data processing method specifically comprises the following steps:
step 1: setting electrocardiosignal acquisition frequency and sampling time to obtain the number of samples;
step 2: the R-wave threshold is obtained by DIFF (i) =f (i+1) -f (i-1) +2*f (i+2) -2*f (i-2); where f (-) represents sample data, i represents data values in samples, DIFF (i) =zeros (1, samples) represents Double zero-like matrices that generate a 1 sample number;
step 3: obtaining R-R interval data, and calculating:
wherein N represents the total number of data in the acquired sample,represents the average value of the time difference between adjacent main wave peaks, RR i Representing the time difference between the ith adjacent dominant wave peak; SDNN represents R-R interval standard deviation;
obtaining SDNN;
step 4: and analyzing the SDNN to judge the fatigue degree of the driver, wherein the Standard Deviation (SDNN) of the normal R-R interval obviously rises with the deepening of the fatigue degree, and other time domain index changes are not obvious.
Specific examples:
as shown in figure 1 which is a schematic diagram of the structure of a fatigue monitoring system that fuses myoelectric and electrocardiographic signals,
fig. 2 is a schematic structural diagram of a surface electromyographic signal electrode module, wherein an electrode slice substrate used for the surface electromyographic signal electrode module is made of silver, and the surface of the electrode slice substrate is plated with silver chloride. The invention adopts the common bipolar to measure the electromyographic signals, and a reference electrode is added in the middle for reducing noise. The input signals detected by the two electrodes are subtracted to cancel out the common mode components of the two electrodes, so as to amplify the differential mode. Because the electromyographic signals are transmitted in the human body at a high attenuation speed, the electrode is placed on the abdomen of the muscle. Meanwhile, in order to better collect the electromyographic signals, the cuticle of the epidermis is removed before the electrode is placed, and the electrode is rubbed by alcohol, and when the electrode is placed, the electrode should be carefully aligned with the direction of the muscle fibers, that is, the long axis of the electrode should be parallel to the long axis of the muscle fibers.
Fig. 3 is a schematic structural diagram of a surface electromyographic signal acquisition and conversion module, where the surface electromyographic signal circuit structure includes a front-section acquisition circuit, which is essentially an amplifying and filtering circuit, and includes a built-in amplifying circuit wood block of the surface electromyographic signal acquisition module, a built-in high-pass filter and low-pass filter module of the surface electromyographic signal acquisition module, a notch processing module, and a built-in bias circuit module. The main functions are as follows: (1) Amplifying sEMG signals to meet the conditions according to the requirements of A/D acquisition; (2) filtering out noise signals when amplifying sMEG.
The surface muscle electric signal acquisition module is internally provided with an amplifying circuit module and is provided with a pre-amplifying circuit and a post-amplifying circuit. The total of the two-stage amplification factors is about 1000 times, namely the amplitude of the collected electromyographic signals is amplified by 1000 times. In order to reduce the influence of distributed capacitance in the signal transmission process, a pre-amplifying circuit is closely attached to the surface electrode. Meanwhile, since the electromyographic signals are characterized by weak and doped with interference signals, the preamplifiers must employ low-noise high-performance amplifiers and have shorter response times to improve the signal-to-noise ratio. The post-amplifier circuit sEMG signal is amplified to a range that is easy to read.
The surface muscle electric signal acquisition module is internally provided with a high-pass filter and a low-pass filter module, and 4-order Butterworth high-pass filtering is used for removing motion artifacts and electric noise caused by cables, and filtering out signals with the frequency greater than 1000 Hz; the low pass filtering filters out signals having a frequency of less than 10Hz of the sEMG signal. The portion of the sEMG signal with the greatest signal energy frequency is retained.
The notch processing module is used for removing power frequency interference of 50 Hz. For sEMG signals, a band-reject filter method based on an operational amplifier is adopted; the zero-setting voltage-stabilizing circuit has the function of ensuring the stable and consistent input and output of the circuit, namely inhibiting the zero drift of the circuit.
The bias circuit module is arranged in the module so as to facilitate reading of the amplified sEMG signal. After the pre-amplifying circuit and the post-amplifying circuit, the value of the sEMG signal fluctuates between +/-2V, so that the built-in bias circuit is designed to bias the myoelectric signal level upwards by 2.5V. At this time, the sEMG signal range is approximately 0-5V, which is convenient for the reading of most ADC conversion modules.
The surface electromyographic signal ADC conversion circuit module is used for converting the analog quantity output by the surface electromyographic signal into the digital quantity through the ADC module. The invention adopts the Arduino board to carry out analog-to-digital conversion, and the Arduino IDE writes a program code to upload the program to the Arduino circuit board, so that the Arduino circuit board can convert an analog signal into a digital signal.
As shown in fig. 4, the surface electromyographic signal acquisition processing flow chart includes the following steps:
step 1: the Arduino circuit board is connected with the terminal module, and digital signals are transmitted to the terminal module while power is supplied to the myoelectric signal acquisition module and the ADC conversion circuit;
step 2: according to the property of the muscle electric signals, the acquisition frequency of the terminal module is adjusted to be 10kHz;
step 3: removing the cuticle of the skin of the muscular abdomen of the driver, wiping the cuticle with alcohol, fixing the bipolar electrode plate on the muscular abdomen of the driver, and enabling the long axis of the bipolar electrode plate to be parallel to the long axis direction of muscle fibers;
step 4: and a terminal module is arranged, the electromyographic signal image is displayed in real time, electromyographic signal data is derived in real time, and the data of the electromyographic signal is derived into an Excel file, so that the signal fusion is convenient to call. .
Fig. 5 shows an electrocardiosignal acquisition and processing module diagram, wherein the electrocardiosignal acquisition and processing module comprises a sensing acquisition module, a main control module and an upper computer module, and the sensing acquisition module is connected by using a data line, an LED infrared light device of the sensing acquisition module emits light at the moment, and a display device is powered. The index finger of the tested person is clung to the sensing acquisition module and fixed by the black binding belt, so that the interference of the ambient light change on the acquisition of the photoelectric signal can be eliminated to a great extent. The upper computer module is controlled to start data acquisition, the LED infrared light device of the sensing acquisition module emits red light with the wavelength of 660nm and infrared light with the wavelength of 900nm, under the action of heart beat, the hemoglobin content in blood vessels is changed, the light transmittance of the infrared light is changed, the intensity of light reflected by human tissues and received by the photoelectric converter in the infrared light device is changed periodically, and accordingly, the periodic change of an electric signal obtained by the photoelectric converter can reflect the periodic change of biological signals.
The digital filter module in the sensing acquisition module adopts a pre-amplifying circuit firstly, so that the output impedance of the amplifier needs to be improved in order to reduce the influence of the internal resistance of a signal source; in order to reduce the power frequency interference carried by the human body and the physiological signal interference except the measurement signal, a high differential amplification coefficient is required. Therefore, the high-precision instrument amplifier AD620 is adopted, the power supply range is wide from-2.3V to +2.3V, and the electrocardiosignal amplifier is suitable for low-voltage and low-power consumption application occasions and finally amplifies the electrocardiosignal by 10 times. Then filtering the noise, wherein the noise of the electrocardiosignal mainly comprises 50Hz power frequency interference, and the elimination can be realized by adopting a power frequency trap; the baseline drift is mainly generated by the respiration of a human body, so that the reference voltage of an electrocardiosignal is offset, and the electrocardiosignal is wholly floated. The myoelectric signal noise is mainly removed by a low-pass filter, and the frequency is mainly 50-5000 Hz; before low-pass filtering, the amplitude of the electrocardiosignal is still very small, and the 10-time amplification realized by the pre-amplifying circuit is still far from the requirement of digital-to-analog conversion, so that a 100-time main amplifying circuit is needed to be added to amplify the electrocardiosignal 1000 times, and the voltage input range of the digital-to-analog conversion chip is reached. Finally, the primarily processed electrocardiosignals are obtained and transmitted to a main control module through a data line.
The main control module is connected with the upper computer module and the sensing acquisition module, and is mainly used for converting electrocardiosignals from the sensing acquisition module into a communication protocol and transmitting the communication protocol to the upper computer module through the USB interface for further analysis and calculation. And the USB power supply is also responsible for providing the working voltage of 1.8V for the sensing acquisition module.
The upper computer module receives the transmitted electrocardiosignals, and the electrocardiosignal processing calculation module designs an algorithm to calculate the heart rate value, and the basic principle is as follows: and performing time domain analysis by using the electrocardiosignals subjected to amplification filtering processing, and performing heart rate estimation according to the number or frequency of waveform peaks in a specific time period. And then, transmitting the real-time heart rate value data to a heart rate recording module, continuously importing the data into a CSV file and storing the data, so that the subsequent fatigue algorithm can be conveniently invoked. And meanwhile, heart rate data are transmitted to a heart rate signal real-time display module, and the change of the heart rate value is displayed in real time by utilizing a display screen of the upper computer, so that a user can conveniently judge the working state of the whole device according to the normal or abnormal state of the heart rate value.
Fig. 6 is a schematic diagram of a surface electromyographic signal (sEMG) and an electrocardiosignal data processing module, and the surface electromyographic signal (sEMG) is a very important bioelectric signal generated along with muscle activity. The formation of the electromyographic signals is based on the excitation and contraction of muscles. Research shows that the surface electromyographic signals show a certain regularity along with the deepening of the fatigue degree of the human body, the amplitude of the surface electromyographic signals can be increased along with the deepening of the fatigue degree of the human body, and the frequency can be reduced. The characteristic parameters of the electromyographic signals can be obtained by carrying out time domain analysis and frequency domain analysis on the processed data: integrated myoelectricity value (IEMG), root mean square value (RMS), average power frequency (MPF), median Frequency (MF). A common indicator of the current phase surface electromyographic signal analysis in the frequency domain is mainly the Median Frequency (MF) and the average power frequency (MPF).
The specific treatment process comprises the following steps:
step 1: performing fast Fourier transform on the electromyographic signals processed by the surface electromyographic signal acquisition and conversion circuit module to obtain amplitude after Fourier transform;
step 2: calculating an integral myoelectricity value, a root mean square value, an average power frequency and a median frequency;
step 3: judging whether the driver is tired or not according to the trend of the median frequency, and if the median frequency is reduced, judging that the driver is tired, and if the median frequency is reduced, judging that the driver is not tired, and if the median frequency is increased, judging that the driver is not tired.
The electrocardiosignal data processing template consists of a P wave, a QRS wave group and a T wave according to the waveform of a normal electrocardiosignal in one period. The heart rate variability index is mainly composed of P wave, QRS wave group, T wave and U wave, wherein the heart rate variability index is a common index for researching driving fatigue. When the fatigue degree changes, the Standard Deviation (SDNN) of the normal R-R interval changes obviously, so that the SDNN is selected as the time domain index. The specific treatment process comprises the following steps:
step 1: setting electrocardiosignal acquisition frequency and sampling time to obtain the number of samples;
step 2: the R-wave threshold is obtained by DIFF (i) =f (i+1) -f (i-1) +2*f (i+2) -2*f (i-2); where f (-) represents sample data, i represents data values in samples, DIFF (i) =zeros (1, samples) represents Double zero-like matrices that generate a 1 sample number;
step 3: obtaining R-R interval data, and calculating:
wherein N represents the total number of data in the acquired sample,represents the average value of the time difference between adjacent main wave peaks, RR i Representing the time difference between the ith adjacent dominant wave peak;
obtaining SDNN;
step 4: and analyzing the SDNN to judge the fatigue degree of the driver, wherein the Standard Deviation (SDNN) of the normal R-R interval obviously rises with the deepening of the fatigue degree, and other time domain index changes are not obvious.
As shown in fig. 7, which is a schematic structural diagram of a nonlinear support vector machine algorithm data fusion template, the data of a surface electromyographic signal and electrocardiosignal data processing template belong to nonlinear data, so that the data cannot be directly solved by a support vector machine, the invention adopts a feature layer fusion method, after receiving the data from the surface electromyographic signal and electrocardiosignal data processing template, a gaussian kernel function K (x 1, x 2) =exp (-gamma abs (x 1-x 2)/(2)) is constructed after selecting proper electromyographic parameters, then a data linearization fusion processor special for electromyographic and electrocardiosignal is established, the electromyographic and electrocardiosignal data are linearized and fused into fatigue feature vectors, the fatigue condition of a driver is comprehensively judged by judging whether the feature vectors accord with fatigue features, the fatigue state is marked as 1, the non-fatigue state is marked as-1, and the electromyographic features and whether the fatigue is made into a data set named as test; then the established Gaussian kernel function is adopted to establish a decision function by adopting a linearization support vector machine methodAn iteration is performed which is carried out in such a way that,and obtaining a final decision boundary, importing the Testing data set into the decision boundary, and classifying the fatigue state of the driver.
As shown in fig. 8, which is a graph of the test result of the support vector machine on the fatigue data set based on the gaussian kernel function, the abscissa is the EMG myoelectric processing signal, the ordinate is the ECG electrocardiographic processing signal, the decision within the decision boundary is a non-fatigue state, and the decision outside the decision boundary is a fatigue state.
Claims (5)
1. The fatigue monitoring system for fusing myoelectricity and electrocardiosignals is characterized by comprising a surface myoelectricity electrode module, a surface myoelectricity signal acquisition and conversion module, a surface myoelectricity signal data processing module, an electrocardiosignal acquisition and conversion module, an electrocardiosignal data processing module and a nonlinear support vector machine algorithm data fusion module;
the surface electromyographic signal electrode module collects surface electromyographic signals of a driver and sends the collected surface electromyographic signals of the driver to the surface electromyographic signal collection and conversion module; the surface electromyographic signal acquisition and conversion module is used for carrying out signal amplification, filtering and A/D conversion on the surface electromyographic signal of the driver and sending an output signal to the surface electromyographic signal data processing module; the surface electromyographic signal data processing module performs time domain analysis and frequency domain analysis on the input data to obtain characteristic parameters of the surface electromyographic signal of the driver;
the electrocardiosignal acquisition and conversion module acquires an electrocardiosignal of a driver, amplifies, filters and performs A/D conversion, and sends an output signal to the electrocardiosignal data processing module; the electrocardiosignal data processing module performs time domain analysis and frequency domain analysis on the input data to obtain characteristic parameters of the electrocardiosignal of the driver;
the surface electromyographic signal characteristic parameters and the electrocardiosignal characteristic parameters of the driver are sent to a nonlinear support vector machine algorithm data fusion module; the nonlinear support vector machine algorithm data fusion module fuses the surface electromyographic signal characteristic parameters and the electrocardiosignal characteristic parameters of the driver into fatigue characteristic vectors, and judges the fatigue condition of the driver by judging whether the characteristic vectors accord with the fatigue characteristics or not;
the surface electromyographic signal acquisition and conversion module comprises a sensor module, an ADC conversion circuit and a terminal module;
the sensor module comprises an amplifying module, a filtering module and a notch processing module, and is used for amplifying, filtering and eliminating power frequency interference on the collected surface electromyographic signals of the driver;
the ADC conversion circuit converts the surface electromyographic signals processed by the sensor module from analog signals to digital signals and sends the digital signals to the terminal module, and the ADC conversion circuit is realized by adopting an Arduino circuit board;
the terminal module records and visualizes the surface electromyographic signals through software;
the electrocardiosignal acquisition and conversion module comprises an LED infrared light device, a sensing acquisition module, a main control module and an upper computer module;
the LED infrared light device is fixed at the index finger of a driver, and converts an optical signal into an electric signal by utilizing a photoelectric volume method;
the sensing acquisition module comprises an amplification module, a filtering module and a notch processing module, and is used for amplifying, filtering, eliminating baseline drift interference, eliminating power frequency interference and eliminating myoelectric signal interference on the acquired driver electrocardiosignals;
the main control module receives the electrocardiosignals processed by the sensing acquisition module, performs A/D conversion on the electrocardiosignals, and transmits the electrocardiosignals to the upper computer module;
the upper computer module comprises an electrocardiosignal processing and calculating module, a heart rate recording module and a heart rate real-time display module, wherein the electrocardiosignal processing and calculating module calculates a real-time heart rate value through a time domain analysis algorithm, stores the real-time heart rate value in the heart rate recording module, imports the heart rate signal real-time display module, creates a window by utilizing an upper computer screen, and displays the change condition of the heart rate value in real time;
the surface electromyographic signal acquisition method specifically comprises the following steps:
step 1: the Arduino circuit board is connected with the terminal module, and digital signals are transmitted to the terminal module while power is supplied to the myoelectric signal acquisition module and the ADC conversion circuit;
step 2: according to the property of the muscle electric signals, the acquisition frequency of the terminal module is adjusted to be 10kHz;
step 3: removing the cuticle of the skin of the muscular abdomen of the driver, wiping the cuticle with alcohol, fixing the bipolar electrode plate on the muscular abdomen of the driver, and enabling the long axis of the bipolar electrode plate to be parallel to the long axis direction of muscle fibers;
step 4: setting a terminal module, displaying an electromyographic signal image in real time and deriving electromyographic signal data in real time;
the surface electromyographic signal data processing method specifically comprises the following steps:
step 1: performing fast Fourier transform on the electromyographic signals processed by the surface electromyographic signal acquisition and conversion circuit module to obtain amplitude after Fourier transform;
step 2: calculating an integral myoelectricity value, a root mean square value, an average power frequency and a median frequency;
step 3: judging whether the driver is tired or not according to the trend of the median frequency, and if the median frequency is reduced, judging that the driver is tired, and if the median frequency is reduced, judging that the driver is not tired, and if the median frequency is increased, judging that the driver is not tired;
an electrocardiosignal data processing method specifically comprises the following steps:
step 1: setting electrocardiosignal acquisition frequency and sampling time to obtain the number of samples;
step 2: the R-wave threshold is obtained by DIFF (i) =f (i+1) -f (i-1) +2*f (i+2) -2*f (i-2); where f (-) represents sample data, i represents data values in samples, DIFF (i) =zeros (1, samples) represents Double zero-like matrices that generate a 1 sample number;
step 3: obtaining R-R interval data, and calculating:
wherein N represents the total number of data in the acquired sample,represents the average value of the time difference between adjacent main wave peaks, RR i Representing the time difference between the ith adjacent main wave crest, wherein SDNN represents the R-R interval standard deviation;
obtaining SDNN;
step 4: and analyzing the SDNN to judge the fatigue degree of the driver, wherein the value of the SDNN is positively related to the fatigue degree.
2. The fatigue monitoring system for fusing myoelectricity and electrocardiosignals according to claim 1, wherein the surface myoelectricity signal electrode module measures the surface myoelectricity signal by using bipolar electrode plates, and a reference electrode is added between the two bipolar electrode plates to offset common mode components of signals acquired by the two bipolar electrode plates, and a differential mode part is reserved.
3. The fatigue monitoring system for fusing myoelectric and electrocardiosignal according to claim 1, wherein the surface myoelectric signal data processing module performs time domain analysis and frequency domain analysis on the digitized surface myoelectric signal to obtain characteristic parameters of the myoelectric signal, and the fatigue monitoring system comprises: integrated myoelectric value, root mean square value, average power frequency, median frequency.
4. The fatigue monitoring system for fusing myoelectricity and electrocardiosignals according to claim 1, wherein the electrocardiosignal data processing module detects QRS waves by adopting a differential threshold value, deletes false, missed and abnormal QRS waves from the R-R interval sequence to form an N-N interval sequence, and performs time domain analysis and frequency domain analysis of HRV by using the N-N interval sequence to obtain characteristic parameters of the electrocardiosignals.
5. The system for monitoring fatigue by fusing myoelectricity and electrocardiosignals according to claim 1, wherein the nonlinear support vector machine algorithm data fusion module is used for receiving the characteristic parameters of the electrocardiosignals and the characteristic parameters of the myoelectricity, obtaining a final decision boundary by adopting a support vector machine algorithm according to the fatigue state judged by the characteristic parameters of the electrical signals or the characteristic parameters of the myoelectricity, judging the state within the decision boundary as a non-fatigue state, and judging the state outside the decision boundary as a fatigue state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110424867.XA CN113100776B (en) | 2021-04-20 | 2021-04-20 | Fatigue monitoring system and method for fusing myoelectricity and electrocardiosignal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110424867.XA CN113100776B (en) | 2021-04-20 | 2021-04-20 | Fatigue monitoring system and method for fusing myoelectricity and electrocardiosignal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113100776A CN113100776A (en) | 2021-07-13 |
| CN113100776B true CN113100776B (en) | 2023-08-29 |
Family
ID=76718941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110424867.XA Active CN113100776B (en) | 2021-04-20 | 2021-04-20 | Fatigue monitoring system and method for fusing myoelectricity and electrocardiosignal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113100776B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114625257B (en) * | 2022-05-16 | 2022-08-16 | 浙江强脑科技有限公司 | Action recognition method and device based on electromyographic signals |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108765876A (en) * | 2018-05-31 | 2018-11-06 | 东北大学 | Driving fatigue depth analysis early warning system based on multimode signal and method |
| WO2020042709A1 (en) * | 2018-08-31 | 2020-03-05 | 江苏盖睿健康科技有限公司 | Wearable device capable of reflecting degree of fatigue of human body |
| WO2020151144A1 (en) * | 2019-01-24 | 2020-07-30 | 五邑大学 | Generalized consistency-based fatigue classification method for constructing brain function network and relevant vector machine |
-
2021
- 2021-04-20 CN CN202110424867.XA patent/CN113100776B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108765876A (en) * | 2018-05-31 | 2018-11-06 | 东北大学 | Driving fatigue depth analysis early warning system based on multimode signal and method |
| WO2020042709A1 (en) * | 2018-08-31 | 2020-03-05 | 江苏盖睿健康科技有限公司 | Wearable device capable of reflecting degree of fatigue of human body |
| WO2020151144A1 (en) * | 2019-01-24 | 2020-07-30 | 五邑大学 | Generalized consistency-based fatigue classification method for constructing brain function network and relevant vector machine |
Non-Patent Citations (1)
| Title |
|---|
| 王琳 ; 张陈 ; 尹晓伟 ; 付荣荣 ; 王宏 ; .一种基于驾驶员生理信号的非接触式驾驶疲劳检测技术.汽车工程.2018,(第03期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113100776A (en) | 2021-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP0335357B1 (en) | Improved method and apparatus for detecting optical pulses | |
| US7848795B2 (en) | Determination of sleep depth | |
| US20090024044A1 (en) | Data recording for patient status analysis | |
| CN107080527B (en) | Mental state monitoring method based on wearable vital sign monitoring device | |
| CN110338813B (en) | Noninvasive blood glucose detection method based on spectrum analysis | |
| CN109259762A (en) | A kind of muscular fatigue comprehensive test device based on multivariate data fusion | |
| CN102462494A (en) | Novel intelligent electrocardiogram test healthcare apparatus | |
| CN112006686A (en) | Neck muscle fatigue analysis method and system | |
| CN115500800A (en) | Wearable physiological parameter detection system | |
| CN108324286B (en) | Infrared noninvasive blood glucose detection device based on PCA-NARX correction algorithm | |
| CN114886388B (en) | Evaluation method and device for quality of electroencephalogram signal in anesthesia depth monitoring process | |
| CN111166306A (en) | Physiological signal acquisition method, computer device and storage medium | |
| KR20160107390A (en) | Apparatus for measuring bio-signal | |
| CN113100776B (en) | Fatigue monitoring system and method for fusing myoelectricity and electrocardiosignal | |
| Mukti et al. | Performance analysis of twelve lead ECG based on delivery distance using bluetooth communication | |
| CN201894645U (en) | Novel intelligent electrocardiogram test healthcare apparatus | |
| CN111134641A (en) | Sleep monitoring chip system and sleep monitoring chip | |
| Ottenbacher et al. | An approach to reliable motion artifact detection for mobile long-term ECG monitoring systems using dry electrodes | |
| CN203208022U (en) | Electroencephalograph | |
| CN201200408Y (en) | Lie detector with function for detecting brain electricity | |
| CN114903445A (en) | Intelligent monitoring and early warning system for cardiovascular and cerebrovascular diseases | |
| Anas et al. | On-line monitoring and analysis of bioelectrical signals | |
| CN203208024U (en) | Brain wave training instrument | |
| CN114052682B (en) | Physiological parameter monitoring system based on raspberry group | |
| Jegan et al. | Methodological role of mathematics to estimate human blood pressure through biosensors |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |