A method for collecting electromyographic and blood oxygen signals based on electromyographic detection technology and near-infrared spectroscopy technology

Abstract

The present invention discloses a method for collecting electromyography and blood oxygen signal based on electromyography detection technology and near infrared spectroscopy technology, comprising the following steps: step 1, designing a surface electromyography and blood oxygen signal collection system; step 2, determining the model parameters of the blood oxygen signal collection system based on the modified Lambert-Beer theorem; step 3, respectively designing an intermittent grip test and a forearm blocking test to collect multivariate signal changes of muscles during the motion state and blood flow blocking process, and pre-processing the collected signals to verify the effectiveness of the surface electromyography and blood oxygen signal collection system to collect signals. The present invention can collect and pre-process the electromyography and near infrared signals of muscles during the motion process, which is of great value for real-time evaluation of fatigue degree during the rehabilitation training of athletes.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a method for collecting myoelectric and blood oxygen signals based on myoelectric detection technology and near-infrared spectroscopy technology. Background Art

[0002] Sports fatigue prevention and treatment is a hot topic in the field of sports human science research. The phenomenon of muscle fatigue mainly involves neurological and metabolic physiological factors. From the perspective of the nervous system, when muscles are fatigued, they will cause the superposition of action potential changes in time and space, which will lead to changes in sEMG, and then reflect the body state of the muscle. From the perspective of physiological metabolism, the vascular activity in the muscle is enhanced, and the consumption of blood oxygen in the local tissue increases, which will reduce the blood oxygen content in the tissue to another equilibrium point. Monitoring the degree of hypoxia in muscle tissue can be used to judge the degree of muscle fatigue.

[0003] sEMG detection technology has been studied by many researchers due to its good detectability, high sensitivity, non-invasiveness, safety and reliability, and has achieved some results. sEMG is an external comprehensive representation of the potential activity of human muscle cells, and its signal change characteristics are related to the biochemical and physiological changes that occur in skeletal muscle during fatigue contraction.

[0004] Near Infrared Spectroscopy (NIRS) is a non-destructive method for detecting changes in blood oxygen concentration in human tissues. It is based on the low absorption and high scattering characteristics of biological tissues for light with a wavelength between 700nm and 900nm. Photons in this band can penetrate into tissues several centimeters deep from the skin on the surface of the human body, greatly increasing the spatial range of optical analysis of biological tissues. The scattering effect of biological tissues on near-infrared light in the 700-900nm band is dominant, far greater than the absorption effect. This characteristic makes this band range called the "Optical Window", specifically referring to the fact that light in this band is easier to penetrate human tissues than light in other bands.

[0005] Current research is mainly based on single-channel unit signals. The collected signals are relatively simple and cannot fully reflect the changing laws of the intrinsic physical parameters during muscle fatigue. There are few studies that analyze the causes of muscle fatigue from multiple dimensions based on multi-signal fusion. In addition, most of the existing athlete fatigue detection instruments are bulky. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a method for collecting electromyography and blood oxygen signals based on electromyography detection technology and near-infrared spectroscopy technology, which can collect and preprocess both electromyography and near-infrared signals during muscle movement, and is of great value for real-time evaluation of fatigue level during athlete rehabilitation training.

[0007] In order to solve the above technical problems, the present invention provides a method for collecting myoelectric and blood oxygen signals based on electromyographic detection technology and near-infrared spectroscopy technology, comprising the following steps:

[0008] Step 1: Design the electromyography and blood oxygen signal acquisition system;

[0009] Step 2: Determine the model parameters of the blood oxygen signal acquisition system based on the modified Lambert-Beer theorem;

[0010] Step 3: Design an intermittent grip test and a forearm blocking test to collect the multivariate signal changes of muscles during the movement state and blood flow blocking process, and pre-process the collected signals to verify the effectiveness of the collected signals by the blood oxygen signal acquisition system.

[0011] Preferably, in step 1, the electromyography and blood oxygen signal acquisition system includes an electromyography signal acquisition circuit and a blood oxygen signal acquisition circuit; wherein the electromyography signal acquisition circuit includes a differential circuit and a filter amplifier circuit, the differential circuit adopts a dual-end input single-output model and is placed at the forefront of the acquisition circuit, and a corresponding protection circuit is designed to suppress radio frequency interference in the signal, and the filter amplifier circuit includes a first-order RC low-pass filter and a second-order RC high-pass filter; the blood oxygen signal acquisition circuit includes a constant current source drive circuit, a timing separation circuit and a filter circuit, the constant current source drive circuit is used to drive a dual-wavelength LED light source, the timing separation circuit is based on a dual-channel analog switch, and the dual-channel switch is driven by a fixed-frequency PWM wave in a time-sharing manner, a voltage divider circuit is designed at the output end of the timing separation circuit to reduce channel attenuation, and the filter circuit includes a first-order RC low-pass filter and a second-order RC high-pass filter.

[0012] Preferably, in step 2, according to the scattering and reflection paths of photons in tissues simulated by Monte Carlo, the modified Lambert-Beer theorem formula is described as: Where L represents the physical distance between the near-infrared light source and the photoelectric sensor, B is the differential path factor used to describe the mathematical relationship between the actual propagation path of the photon and L, G is the dissipation constant used to represent the effect of photon attenuation, and I i is the incident light intensity, I is the outgoing light intensity, C represents the concentration of the absorbing substance, the unit is: mol / L; ε represents the molar absorption coefficient, the unit is: mM -1 cm -1 ; A represents optical density;

[0013] Since the loss G caused by photons during propagation cannot be calculated directly, the relative change of tissue blood oxygen concentration is considered. The specific formula is:

[0014] After adjusting the emission light intensity of the dual-wavelength light source, I0 represents the light intensity collected by the detector when the muscle tissue is in a resting state; I t represents the light intensity collected by the detector after the muscle state changes; ΔC represents the concentration change of the absorbing substance at time t; ΔA represents the change of light density at time t;

[0015] Use dual-wavelength LED to obtain the change in light density at each wavelength. The specific formula is as follows:

[0016]

[0017] in, And so on.

[0018] Preferably, in step 3, an intermittent grip test and a forearm blocking test are respectively designed to collect multivariate signal changes of muscles in the motion state and blood flow blocking process, and the collected signals are preprocessed to verify the effectiveness of the blood oxygen signal acquisition system in collecting signals. Specifically, an intermittent grip test is designed, sEMG is collected in the experiment, and the signal is processed by power frequency notch in MATLAB, and then wavelet denoising is performed and Fourier transform is performed to obtain the time domain and frequency domain characteristic components of the electromyographic signal, wherein the Fourier transform formula is as follows:

[0019]

[0020] Where ω represents frequency and t represents time. Since Fourier transform considers a periodic function to contain multiple frequency components, f(t) here is the superimposed signal.

[0021] The OxiplexTS tissue oximeter was used for a control test with this system. In the forearm blocking experiment, the changes in blood oxygen saturation were calculated in real time based on the differences in the absorption of red light and near-infrared light by Hb and HbO2, and the changes in the two hemoglobin parameters were calculated in real time based on the modified Lambert-Beer theorem model.

[0022] The beneficial effects of the present invention are as follows: (1) The present invention realizes the collection of multivariate signals of muscles based on electromyography detection technology and near-infrared spectroscopy technology, and can more accurately reflect the changes in muscle status; (2) The present invention adopts non-invasive and fast selection of Ag / AgCl gel electrodes to pick up electromyography signals, designs a stable signal processing circuit with strong anti-interference ability to complete the filtering processing of weak signals, and has low cost; designs a constant current source driving circuit to realize stable power driving of dual-wavelength near-infrared light, uses an analog switch to design a timing multiplexing circuit to extract the dual-wavelength optical signal in the channel, and then calculates the blood oxygen saturation and the relative changes of HbO2 and Hb according to the modified Lambert-Beer theorem; (3) The present invention realizes non-invasive and convenient electromyography and blood oxygen signal collection, which is of great value for muscle fatigue assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0024] Figure 2 It is a structural schematic diagram of the acquisition probe of the present invention.

[0025] Figure 3 It is a schematic diagram of the structure of the constant current source driving circuit of the present invention.

[0026] Figure 4 It is a schematic diagram of the structure of the sEMG signal processing circuit of the present invention.

[0027] Figure 5 This is a topological diagram of the NIRS signal processing circuit of the present invention.

[0028] Figure 6 This is a schematic diagram of the blood oxygen non-invasive collection and detection model of the present invention.

[0029] Figure 7 (a) is a schematic diagram of the original sEMG signal measured under a certain group of intermittent grip strength experiments of the present invention.

[0030] Figure 7 (b) is a schematic diagram of FFT of the original sEMG signal measured in a certain group of intermittent grip strength experiments of the present invention.

[0031] Figure 7 (c) is a schematic diagram of wavelet soft threshold denoising of the sEMG original signal measured under a certain group of intermittent handgrip experiments of the present invention.

[0032] Figure 7 (d) is a schematic diagram of wavelet soft threshold denoising FFT of the sEMG original signal measured in a certain group of intermittent handgrip experiments of the present invention.

[0033] Figure 8This is a schematic diagram of the real-time changes of two hemoglobin concentrations measured by the system and the OxiplexTS tissue oximeter under a certain group of forearm blocking experiments of the present invention. DETAILED DESCRIPTION

[0034] like Figure 1 As shown, a method for collecting electromyographic and blood oxygen signals based on electromyographic detection technology and near-infrared spectroscopy technology includes the following steps:

[0035] Step 1: Design a blood oxygen signal acquisition system;

[0036] like Figure 2 As shown in the figure, it is a schematic diagram of the structure of the acquisition probe. The electromyographic signal is picked up by sticking the Ag / AgCl gel electrode on the target muscle group wiped with alcohol, and the button probe buckles the electrode patch. In addition, the electrodes are placed at three points, two of which are used as the differential input terminals of sEMG, and the other electrode is grounded to improve the accuracy of signal acquisition. As for the position of the electrode, the acquisition electrode is generally placed at the position of the target muscle group, and the ground electrode is placed at the resting muscle.

[0037] According to the characteristics of low absorption and high scattering of biological tissues for light with wavelengths between 700nm and 900nm. According to the modified Lambert-Beer theorem, changes in the blood oxygen state of biological tissues will cause obvious changes in the concentrations of oxyhemoglobin (Oxyhemoglobin, HbO2) and deoxyhemoglobin (Deoxyhemoglobin, Hb), which will cause corresponding changes in the absorption spectrum of tissues to light. Existing studies have shown that within a certain range on both sides of 800nm, the absorption coefficients of HbO2 and Hb change differently. Therefore, the present invention uses a dual-wavelength (735nm / 850nm) LED. According to existing studies, in order to achieve the purpose of collecting blood oxygen change information of deep muscles, the maximum distance between the sensor and the light source emitter is generally 4cm. The system selects the distances from the emitter to the light probe to be 3cm and 4cm respectively, and the distance between the light probes is 1cm, which is used to detect the intensity of the scattered near-infrared light, and convert the light intensity into a voltage signal and transmit it to the signal processing circuit module. Through testing, when the muscle contracts, the light intensity change can be effectively detected at this distance. In order to increase the fit with human muscles and reduce the impact of light transmission and leakage, the rest of the probe, except the sensor, is encapsulated with dense black silicone.

[0038] like Figure 3As shown, the integrated chip MAX6126 with strong anti-interference ability and constant current output function produced by MAXIM (Meishin) is used in the design of the adjustable constant current source drive circuit. The complementary output PWM drive signal of fixed frequency is generated by the timer function of the microprocessor STM32 to realize the time-sharing drive analog switch, thereby realizing the alternating light and dark of the dual-wavelength LED. The function of adding the corresponding filter circuit external capacitor C2 at the output end of the analog switch signal in the figure is to suppress the higher harmonics at the output end. The external resistor R1 and the impedance in the channel form a voltage divider circuit, which is used to reduce the influence of the conduction voltage drop during the conduction period of the analog switch channel. Existing studies have shown that near-infrared light propagates in tissues, and the intensity decays exponentially with distance. Therefore, it can be foreseen that the light intensity that the detector needs to detect is very weak, and a photoelectric sensor with higher sensitivity needs to be selected when selecting a photoelectric conversion device in the design. In order to make the photoelectric conversion circuit have a higher response speed and better suppression effect on interference, the photoelectric conversion chip of the present invention selects OPT101 designed and produced by Texas Instrument, and designs a corresponding peripheral drive circuit to achieve the purpose of outputting a voltage signal.

[0039] like Figure 4 As shown in the figure, the schematic diagram of the sEMG signal processing circuit structure is particularly critical for the surface electromyography sensor. The signal conditioning circuit is mainly composed of four parts: differential amplifier circuit, RC low-pass filter, high-pass filter and secondary amplifier. Since the differential amplifier circuit is the front end of signal acquisition, its design affects the signal-to-noise ratio of sEMG. Therefore, when designing the differential amplifier circuit, it is necessary to consider the influence of factors such as tiny device interference and bias current. The INA128 integrated chip designed by Texas Instruments is selected as the analog input front end of sEMG. When selecting parameters, it should also meet the requirements of amplifying differential mode signals and having strong suppression capabilities for common mode signals, as well as considering input and output impedance factors. In addition, since the chip itself will have radio frequency rectification during operation, the radio frequency interference (RFI) generated by it will cause the collected sEMG to produce DC output offset errors. Therefore, a low-pass filter is constructed at the input end of the operational amplifier for radio frequency filtering, and the cut-off frequency calculation formula is as follows:

[0040]

[0041] By C 11 , C 12 , R 13 and R 14 The cut-off frequency calculation formula of the second-order RC high-pass filter composed of a voltage-controlled voltage source is as follows:

[0042]

[0043] like Figure 5As shown in the figure, the topology diagram of the NIRS signal processing circuit. It is mainly composed of three parts: pre-amplifier, filter circuit, and analog timing separation circuit. There are only two photoelectric conversion sensors on the system probe. The signal received by each sensor is separated from the red light and near-infrared light after passing through the timing separation circuit composed of a dual-channel analog switch. Therefore, the four-channel signal generated by the circuit is adjusted through the potentiometer to adjust its amplification factor so that it can be effectively collected by the ADC function of the microprocessor STM32 and subsequently processed.

[0044] Step 2: Determine the model parameters of the blood oxygen signal acquisition system based on the modified Lambert-Beer theorem;

[0045] Based on the different absorption characteristics of oxygenated hemoglobin (Oxyhemoglobin, HbO2) and deoxyhemoglobin (Deoxyhemoglobin, Hb) in the "light window" of 600-900nm band, this system takes the Lambert-Beer theorem as the theoretical basis and combines the scattering and reflection paths of photons in tissues simulated by Monte Carlo to design a non-invasive blood oxygen signal acquisition model. Figure 7 As shown in (a)-(d):

[0046] The near-infrared light Ne and red light Re emitted by the dual-wavelength (735nm / 850nm) LED light source, a and b are two photoelectric sensors at different distances from the light source. Photons penetrate the multi-component mixed system such as skin and subcutaneous tissue and are received by photoelectric sensors a and b. Since the loss G caused by photons during propagation cannot be calculated directly, the relative change of tissue blood oxygen concentration is considered. The specific formula is as follows:

[0047]

[0048] After adjusting the emission light intensity of the dual-wavelength light source, A(0) represents the light intensity collected by the detector when the muscle tissue is in a resting state; A(t) represents the light intensity collected by the detector after the muscle state changes; ΔC represents the concentration change of the absorbing substance at time t; ΔA represents the change in light density at time t. For the photoelectric sensor used in this system, the change in light density is proportional to the collected voltage, so ΔA can be reflected by the voltage value. The specific formula is as follows:

[0049]

[0050] In formula (2.3), V0 represents the voltage value in the initial state, V t Represents the voltage value at time t, V b is the dark voltage value.

[0051] Due to the complex composition of biological tissues, the change in optical density is not caused by the absorption of a single substance, but a comprehensive superposition effect. In order to collect the concentration changes caused by the change in the motion state of each substance, the present invention uses a dual-wavelength LED to obtain the optical density change at each wavelength. The matrix form of the specific formula for the optical density change corresponding to each wavelength is as follows:

[0052]

[0053] in, And so on.

[0054] Although the value of L will affect the change of the differential path factor B, according to existing studies, when L is greater than a certain value (2.5cm), B will tend to be stable. Based on the results of existing studies, this paper combines the interpolation method to determine the B value that meets this system. The molar absorption coefficient ε in the formula can also be obtained from the literature. At this point, the relative changes in the concentrations of the two hemoglobins in the tissue can be quantitatively calculated. The formula is as follows:

[0055]

[0056] Step 3: Design an intermittent grip test and a forearm blocking test to collect the multivariate signal changes of muscles during the movement state and blood flow blocking process, and pre-process the collected signals to verify the effectiveness of the collected signals by the blood oxygen signal acquisition system.

[0057] An intermittent grip test was designed. sEMG was collected in the experiment. The signal was processed with power frequency notch in MATLAB, and then wavelet denoising and Fourier transform were performed to obtain the time domain and frequency domain characteristic components of the electromyographic signal. The details are as follows:

[0058] like Figure 7 (a)-(d) show the time domain and frequency domain changes of sEMG data measured by the present invention under the intermittent grip test. The experiment was conducted on the middle part of the brachioradialis muscle of the right arm of the human body. The subject sat quietly in a laboratory with a constant temperature for 5 minutes to adjust the human body state to a calm and appropriate state. Use medical alcohol cotton to wipe the brachioradialis muscle position on the subject's right forearm to remove dust and keratin on the skin to ensure good contact between the surface muscle electrode and the subject's skin. Use Ag / AgCl gel electrode to stick to the middle part of the brachioradialis muscle of the right arm, grasp the hand grip dynamometer, and grasp intermittently continuously, and repeat the above steps.

[0059] In the wavelet domain, sEMG signals are distinguished from noise to achieve signal denoising. Sys4 is used to select the mother wavelet. The detailed steps are as follows: the collected electromyographic signals are decomposed, and the threshold denoising is performed by the maximum and minimum thresholds. Since the spectrum of the electromyographic signal after Fourier transform is mainly distributed in 10-500Hz, according to the Nyquist sampling theorem, in order to make the sampled signal more complete, the sampling frequency of the signal is set to 1kHz, and the fifth-order approximation frequency band of the signal will be lower than 15.6Hz. The elimination of low-frequency noise interference can be achieved by subtracting the reconstructed signal of the fifth-order approximation frequency band from the original signal reconstructed by wavelet. Figure 7 (a) and (c) are the comparisons before and after sEMG soft threshold denoising. It can be seen that the signal can effectively retain useful signals and suppress noise signals after wavelet threshold denoising. Figure 7 The signal spectra of (b) and (d) are mainly distributed in the range of 10 to 500 Hz, and the main energy is concentrated between 50 and 150 Hz, which also shows the effectiveness of the sEMG signal acquisition module in this system.

[0060] A typical forearm blocking experiment is used to verify the effectiveness of the system in collecting blood oxygen signals, as follows:

[0061] like Figure 8 As shown in the figure, the data measured by the system and OxiplexTS tissue oximeter of the present invention under a certain group of forearm blocking experiments are changed in real time. In the forearm blocking experiment, the sensor probe is tightly attached to the inner side of the forearm, the bandage of the mercury sphygmomanometer is tied to the upper arm, and the arm to be tested is placed horizontally on the table and the test is started. After the system is powered on and preheated for 20 minutes, it starts to apply pressure to the arm to be tested, and the pressure in the cuff is quickly increased to 120mmHg, so that the blood flow through the veins and arteries is completely blocked. Since the local blood circulation is blocked and the biological tissues of the forearm are constantly consuming oxygen at the same time, the concentration of oxygenated hemoglobin in the forearm shows a downward trend, and the concentration of deoxygenated hemoglobin will gradually increase. The pressure in the cuff is quickly released after a period of time, the blood circulation in the forearm is normal, the oxygen content in the blood in the forearm will quickly return to its original state, and the concentration of oxygenated hemoglobin and deoxygenated hemoglobin in the blood will return to the initial state.

[0062] Figure 8 (I) is the data measured by this system and (II) is the data measured by the OxiplexTS tissue oximeter in a forearm occlusion experiment. It can be seen that the concentration trend and amplitude of oxygenated hemoglobin and deoxygenated hemoglobin obtained by this system under forearm occlusion are basically consistent with the data collected by the OxiplexTS tissue oximeter, and are basically consistent with the existing research conclusions, indicating the effectiveness of this system in collecting tissue blood oxygen signals based on near-infrared technology.

Claims

1. A method for collecting electromyographic and blood oxygen signals based on electromyographic detection technology and near-infrared spectroscopy technology, characterized in that: The steps include: Step 1, design an electromyography and blood oxygen signal acquisition system; the electromyography and blood oxygen signal acquisition system includes an electromyography signal acquisition circuit and a blood oxygen signal acquisition circuit; the electromyography signal acquisition circuit includes a differential circuit and a filter amplifier circuit, the differential circuit adopts a double-end input single-output model and is placed at the front end of the acquisition circuit, and a corresponding protection circuit is designed to suppress radio frequency interference in the signal, and the filter amplifier circuit includes a first-order RC low-pass filter and a second-order RC high-pass filter; The blood oxygen signal acquisition circuit includes a constant current source drive circuit, a timing separation circuit and a filter circuit. The constant current source drive circuit is used to drive a dual-wavelength LED light source. The timing separation circuit is based on a dual-channel analog switch and drives the dual-channel switch in a time-sharing manner with a fixed-frequency PWM wave. A voltage divider circuit is designed at the output end of the timing separation circuit to reduce channel attenuation. The filter circuit includes a first-order RC low-pass filter and a second-order RC high-pass filter. Step 2: Determine the model parameters of the blood oxygen signal acquisition system based on the modified Lambert-Beer theorem; according to the scattering and reflection paths of photons in tissues simulated by Monte Carlo, the modified Lambert-Beer theorem formula is described as: Where L represents the physical distance between the near-infrared light source and the photoelectric sensor, B is the differential path factor used to describe the mathematical relationship between the actual propagation path of the photon and L, G is the dissipation constant used to represent the effect of photon attenuation, and I i is the incident light intensity, I is the outgoing light intensity, C represents the concentration of the absorbing substance, the unit is: mol / L; ε represents the molar absorption coefficient, the unit is: mM -1 cm -1 ; A represents optical density; Since the loss G caused by photons during propagation cannot be calculated directly, the relative change of tissue blood oxygen concentration is considered. The specific formula is: After adjusting the emission light intensity of the dual-wavelength light source, I0 represents the light intensity collected by the detector when the muscle tissue is in a resting state; I t represents the light intensity collected by the detector after the muscle state changes; ΔC represents the change in concentration of the absorbing substance at time t; ΔA represents the change in optical density at time t; Use dual-wavelength LED to obtain the change in light density at each wavelength. The specific formula is as follows: in, The rest are similar; Step 3: Design an intermittent grip test and a forearm blocking test to collect the changes in multivariate signals of muscles during movement and blood flow blocking. Pre-process the collected signals to verify the effectiveness of the signals collected by the electromyography and blood oxygen signal acquisition system.

2. The method for collecting myoelectric and blood oxygen signals based on myoelectric detection technology and near infrared spectroscopy technology as claimed in claim 1, characterized in that: In step 3, an intermittent grip test and a forearm blocking test are designed to collect the multivariate signal changes of the muscles during the motion state and blood flow blocking process. The collected signals are preprocessed to verify the effectiveness of the blood oxygen signal acquisition system in collecting signals. Specifically, an intermittent grip test is designed, sEMG is collected in the experiment, and the signal is processed by power frequency notch in MATLAB, and then wavelet denoising is performed and Fourier transform is used to obtain the time domain and frequency domain characteristic components of the electromyographic signal. The Fourier transform formula is as follows: Where ω represents frequency and t represents time. Since Fourier transform considers a periodic function to contain multiple frequency components, f(t) here is the superimposed signal. The OxiplexTS tissue oximeter was used for a control test with this system. In the forearm blocking experiment, the changes in blood oxygen saturation were calculated in real time based on the differences in the absorption of red light and near-infrared light by Hb and HbO2, and the changes in the two hemoglobin parameters were calculated in real time based on the modified Lambert-Beer theorem model.

Images