Electrode shedding detection method

Abstract

The invention provides an electrode falling detection method. The electrode falling detection method comprises the following steps: acquiring an original signal of a section of biological signal of an organism; filtering 50Hz power frequency interference and other noise of the original signal by using a digital filter to obtain an effective signal; acquiring 50Hz power frequency noise of the original signal by using a digital filter; calculating a signal-to-noise ratio: firstly, respectively calculating the signal intensity of an effective signal and the signal intensity of 50Hz power frequency noise through a formula; secondly, calculating the signal-to-noise ratio of the effective signal of the original signal to the power frequency noise through a formula; the signal-to-noise ratio of the original signal is compared with a preset threshold value, and then whether the electrode falls off or not can be judged. According to the electrode falling detection method, falling detection of pure software can be achieved on the basis of the collected bio-electricity signals without additionally designing or newly adding a sensor, and higher universality and flexibility are achieved; and the algorithm of the detection method is simple.

Description

Technical Field

[0001] The invention relates to a detection method, in particular to an electrode shedding detection method. Background Art

[0002] Along with physiological activities such as muscle contraction and nerve impulses, the human body will spontaneously generate electrical signals. These electrical signals can pass through biological tissues and be detected on the surface of the human body. However, since these signals are weak and interfere with each other, most of them are difficult to distinguish. At present, the common signals that can be collected and applied are electrocardiogram, electroencephalogram and electromyography signals.

[0003] Taking ECG signals as an example, normal heart excitation originates from the sinoatrial node and the upper part of the right atrium, and is transmitted to the atria and ventricles in turn through the conduction system, causing the potential difference inside and outside the myocardial cells to disappear or recover, i.e., depolarization or repolarization. Since human body fluids contain electrolytes and have conductive properties, current will be generated inside and on the surface of the human body, and the potential distributed on the surface of the human body will also change accordingly. By detecting the potential difference at different locations on the surface of the human body, we can record this change. Different detection locations will generate different signals.

[0004] In the process of collecting bioelectric signals, the electrodes attached to the surface of the skin are prone to oxidation, and the quality of the electrodes is also uneven, and they cannot maintain sufficient and effective contact with the organism. At the same time, the subconscious activities of the collected object and other reasons can also cause the electrodes to fall off. These factors make the collected signals inaccurate, which hinders researchers from developing applications based on them. Therefore, it is necessary to detect electrode shedding in a timely manner in practical applications. Traditional electrophysiological signals (ECG, EEG, EMG) are low-frequency and low-amplitude signals, and their electrode shedding detection is currently often completed by hardware circuits.

[0005] Traditional electrophysiological signals (ECG, EEG, EMG) are low-frequency and low-amplitude signals, and their electrode detachment detection is currently usually completed by hardware circuits. The patent "An Electrode Detachment Detection Circuit and Detection Method" (patent number CN110772232B) designs an impedance detection circuit. Due to the physical characteristics of the organism, the resistance value of the organism cannot be infinite, so its resistance value must be much lower than the resistance value when the electrode is detached. In this way, by detecting whether the impedance value of the load contacted by the electrode exceeds a predetermined threshold, it can be determined whether the electrode has fallen off from the organism. The detection circuit designed in the patent "A Lead Detachment Detection Circuit" (patent number CN218606624U) is realized by detecting the electrode input voltage. Since the voltage of the organism passes through a specific amplifier, its voltage amplitude is stable within a certain range. By comparing the voltage value after amplification or other processing with the preset threshold, the detachment detection can be achieved.

[0006] In addition to the fall-off detection by hardware circuits, there are also many methods for fall-off detection based on specific signals and algorithms. These methods do not require the design of new hardware circuits, and can be implemented by using existing functional modules and specific algorithms. For example, the method discussed in the patent "Wearable device and its fall-off detection method, device, and storage medium" (patent number CN110954972B). This method is based on the existing mature acceleration measurement module and determines whether it falls off by real-time monitoring of the acceleration during normal use. The principle of this method is that during normal use, since the electrode is attached to the surface of the organism, its motion state will fall rapidly. Therefore, by judging whether the current acceleration direction is downward and whether the acceleration value exceeds the threshold, the fall-off judgment of the wearable device can be realized. The patent "Sensor fall-off detection method, device and storage medium" (patent number CN110353699B) uses a blood oxygen sensor module to obtain the variability of the characteristic parameters by measuring the average deviation, standard deviation and slope ratio of the preset parameter sequence of the signal, and then performs fall-off detection on the device. The principle is that when the system is working normally, the blood oxygen value, pulse rate and ambient light intensity are always stable within a certain range. When the electrode falls off, these characteristic values ​​deviate from the normal range, which can be used to determine whether the electrode has fallen off.

[0007] Traditional hardware dropout detection methods require the design of additional specific circuits. However, due to the different circuit characteristics of different acquisition devices, new hardware circuits need to make partial or even overall adjustments to their components and circuit parameters, which lacks versatility. At the same time, when these circuits are mounted on integrated circuits, the chip development cost and R&D cycle will be extended, and even the effectiveness of their functions cannot be guaranteed, affecting the overall function of the chip.

[0008] The traditional hardware shedding detection method actually still needs to rely on existing mature sensing equipment. Although it avoids the complexity of adding new circuits and the trouble of adaptation, it still needs to add a dedicated shedding detection circuit, which increases the system manufacturing cost. In addition, the algorithms for characteristic values ​​such as blood oxygen values ​​are mature but complicated, which increases the calculation amount of the MCU.

[0009] Therefore, it is necessary to design a new electrode detachment detection method to overcome the above problems. Summary of the invention

[0010] The purpose of the present invention is to overcome the defects of the prior art and provide an electrode detachment detection method. Based on the collected bioelectric signal itself, it can realize pure software detachment detection without the need for additional design or addition of new sensors, and has higher versatility and flexibility; and the algorithm of the detection method is simple.

[0011] The present invention is achieved in that:

[0012] The present invention provides an electrode detachment detection method, comprising the following steps:

[0013] Step 1: Collect the original signal of a biological signal of an organism;

[0014] Step 2: Use a digital filter to filter out the 50Hz power frequency interference and other noise of the original signal to obtain a valid signal;

[0015] Step 3: Use a digital filter to obtain the 50Hz power frequency noise of the original signal;

[0016] Step 4: Calculate the signal-to-noise ratio: First, use formula 1 to calculate the signal strength of the effective signal and the signal strength of the 50Hz power frequency noise respectively; secondly, use formula 2 to calculate the signal-to-noise ratio of the effective signal and the power frequency noise of this section of the original signal; compare the signal-to-noise ratio of the original signal with the preset threshold, and then determine whether the electrode has fallen off;

[0017] Formula 1:

[0018] Formula 2:

[0019] Among them, P ∞ represents the signal strength of the current signal, n is the serial number of the discrete signal, x[n] represents the sequence value of the nth sample point, 2N+1 represents the effective sample length of a single discrete signal sequence, and N represents the maximum number of samples on both sides of the vertical axis of the rectangular coordinate system; SNR is the signal-to-noise ratio, P S is the effective signal strength, P N is the noise signal strength.

[0020] Furthermore, in step one, the original biological signal of the organism is collected by an electrocardiogram detector, a dynamic electroencephalograph or a portable electroencephalograph.

[0021] Furthermore, in step 2 and step 3, the 50 Hz power frequency noise and the effective signal are obtained by filtering the same section of the original signal, and the signal length remains consistent.

[0022] Furthermore, in step 4, the preset threshold is set based on different collected signals and on the basis of effective experiments.

[0023] The present invention has the following beneficial effects:

[0024] The electrode detachment detection method provided by the present invention is based on the collected bioelectric signal itself. Without the need for additional design or the addition of new sensors, it can realize pure software detachment detection, and has higher versatility and flexibility. It uses the strength of the effective signal and the noise signal as the calculation basis, and the algorithm is simple. It is easy to operate, and only a simple filter is needed to obtain a clear and obvious threshold dividing boundary. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0026] Figure 1 A flow chart of an electrode detachment detection method provided by an embodiment of the present invention;

[0027] Figure 2 A schematic diagram of an effective signal of no electrode shedding provided by an embodiment of the present invention;

[0028] Figure 3 A schematic diagram of a power frequency noise signal without electrode shedding provided in an embodiment of the present invention;

[0029] Figure 4 A schematic diagram of the signal-to-noise ratio of an electrode without dropout provided in an embodiment of the present invention;

[0030] Figure 5 A schematic diagram of an effective signal when an electrode falls off provided by an embodiment of the present invention;

[0031] Figure 6 A schematic diagram of a power frequency noise signal when an electrode falls off provided by an embodiment of the present invention;

[0032] Figure 7 A schematic diagram of the signal-to-noise ratio when an electrode falls off provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] like Figure 1 As shown, the present invention provides an electrode detachment detection method, comprising the following steps:

[0035] Step 1: Collect the original signal of a biological signal of an organism. The original signal of a biological signal of an organism can be collected through an existing hardware system. The existing hardware system includes an electrocardiogram detector, a dynamic electroencephalograph, a portable electroencephalograph, etc.

[0036] Step 2: Use a digital filter to filter out the 50Hz power frequency interference and other noise of the original signal to obtain a valid signal. The bioelectric signal at this time may be a signal of electrode detachment, etc., which is actually a noise signal.

[0037] Step 3: Use a digital filter to obtain the 50 Hz power frequency noise of the original signal. At this time, the 50 Hz power frequency noise and the effective signal are obtained by filtering the same section of the original signal, and the signal length remains consistent.

[0038] Step 4: Calculate the signal-to-noise ratio: First, calculate the signal strength of the effective signal and the signal strength of the 50Hz power frequency noise respectively by formula 1; secondly, calculate the signal-to-noise ratio of the effective signal and the power frequency noise of this section of the original signal by formula 2; compare the signal-to-noise ratio of the original signal with the preset threshold, and then determine whether the electrode is detached. Among them, the preset threshold is set based on different collected signals and effective experiments, and is set according to actual working conditions.

[0039] Formula 1:

[0040] Formula 2:

[0041] In Formula 1, P ∞ represents the signal strength of the current signal, n is the serial number of the discrete signal, x[n] represents the sequence value of the nth sample point; 2N+1 represents the effective sample length of a single discrete signal sequence, and N represents the maximum number of samples on both sides of the vertical axis of the rectangular coordinate system. In formula 2, SNR is the signal-to-noise ratio, P S is the effective signal strength, P N is the noise signal strength, S is the abbreviation of signal, indicating that the power value is the effective signal power value, and N is the abbreviation of noise, indicating that the power value is the noise signal power value.

[0042] The application scenario of the electrode detachment detection method provided by the present invention is that during the bioelectric signal collection process of a living body (not limited to the human body), the conductor attached to the body surface or entering the body of the living body falls off, the connection with the host is unstable, the electrode line is damaged, etc., causing the host to collect signals that are wrong or seriously interfered. This method can be applied to real-time detection during the hardware system collection process, and can also be applied to data analysis of the collection results after the collection is completed.

[0043] The method provided by the present invention is implemented based on the principle that the human body suppresses power frequency noise. Among them, power frequency noise mainly refers to the fact that the hardware equipment is affected by the sine wave with a frequency of about 50Hz from the power grid system during work, resulting in the collected signal being mixed with interference signals with a frequency of about 50Hz. Even if the hardware system is independently powered and not directly connected to the power grid system, it will be affected by electromagnetic waves. At the same time, this interference is limited in the collection process of bioelectric signals. Due to the energy loss during the rectification and conduction process of the 50Hz power frequency through the circuit system, the intensity of the 50Hz power frequency noise is very limited compared with the collected effective signal, and the noise signal intensity will also be suppressed when there is no shedding. However, when the collection electrode falls off, the intensity of the 50Hz power frequency noise is quite obvious among the many disordered noise signals. The present invention is achieved by using this basic phenomenon and by calculating the signal-to-noise ratio.

[0044] like Figures 2 to 7 As shown, they are schematic diagrams of effective signal when the electrode is not detached, power frequency noise signal when the electrode is not detached, signal-to-noise ratio when the electrode is not detached, effective signal when the electrode is detached, power frequency noise signal when the electrode is detached, and signal-to-noise ratio when the electrode is detached. Through these schematic diagrams, it can be intuitively and clearly found that the signal-to-noise ratio when the electrode is detached is significantly lower than the signal-to-noise ratio when the electrode is not detached, which has a good distinction effect.

[0045] exist Figure 7 In the calculation, according to the principle of signal-to-noise ratio, the intensity ratio of all effective signals in the original signal to the intensity ratio of all power frequency signal noise needs to be calculated. However, in the specific calculation, it is impossible to take an infinitely long time series for calculation. Therefore, in practical applications, the effective signal within a period of time and the noise signal within the same period of time will be intercepted to calculate the signal-to-noise ratio. Figure 7 In the figure, before time zero, a certain number (N, 200 in this example) of valid signals and noise signal values ​​have actually been stored in the cache. After time zero, the cache is updated in a first-in-first-out (queue) manner. At the same time, the signal-to-noise ratio is calculated. In the figure, X represents the time when the system starts working, in seconds. Y represents the signal-to-noise ratio of the current and previous N valid signals to the noise signal, in dB. The length of both signals is N+1.

[0046] In summary, the electrode detachment detection method provided by the present invention utilizes the strength advantage and inhibitory effect of bioelectric signals relative to noise signals, and uses the signal-to-noise ratio of bioelectric signals as a characteristic parameter to perform threshold judgment; the intensity of effective signals and noise signals is used as the calculation basis, and the algorithm is simple; this method does not require additional hardware circuit-assisted software detachment detection algorithms, and has higher versatility and flexibility; the operation is simple, and only a simple filter is required to obtain a clear and obvious threshold dividing boundary.

[0047] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting electrode detachment, characterized in that: The following steps are involved: Step 1: Collect the original signal of a biological signal of an organism; Step 2: Use a digital filter to filter out the 50Hz power frequency interference and other noise of the original signal to obtain a valid signal; Step 3: Use a digital filter to obtain the 50Hz power frequency noise of the original signal; Step 4: Calculate the signal-to-noise ratio: First, use formula 1 to calculate the signal strength of the effective signal and the signal strength of the 50Hz power frequency noise respectively; secondly, use formula 2 to calculate the signal-to-noise ratio of the effective signal and the power frequency noise of this section of the original signal; compare the signal-to-noise ratio of the original signal with the preset threshold, and then determine whether the electrode has fallen off; Formula 1: Formula 2: Among them, P ∞ represents the signal strength of the current signal, n is the serial number of the discrete signal, x[n] represents the sequence value of the nth sample point, 2N+1 represents the effective sample length of a single discrete signal sequence, and N represents the maximum number of samples on both sides of the vertical axis of the rectangular coordinate system; SNR is the signal-to-noise ratio, P S is the effective signal strength, P N is the noise signal strength.

2. The electrode detachment detection method according to claim 1, characterized in that: In step one, the original biological signal of the organism is collected by an electrocardiogram detector, a dynamic electroencephalograph or a portable electroencephalograph.

3. The electrode detachment detection method according to claim 1, characterized in that: In step 2 and step 3, the 50 Hz power frequency noise and the effective signal are obtained by filtering the same section of the original signal, and the signal length remains consistent.

4. The electrode detachment detection method according to claim 1, characterized in that: In step 4, the preset threshold is set based on different collected signals and on the basis of effective experiments.

Images