Wireless surface electromyography nerve conduction test system

Abstract

The invention belongs to the technical field of medical electronic equipment, and particularly relates to a wireless surface electromyogram nerve conduction testing system which comprises a flexible wearable sensor module. A multi-mode signal processing module and a wireless transmission and edge calculation module; an intelligent analysis terminal; according to the wireless surface electromyogram nerve conduction test system, a wireless transmission mode is adopted, limitation of a cable to motion is avoided, motion artifacts caused by cable traction are reduced, acquired signals can reflect muscle activity and nerve conduction conditions more truly, the risk of missed diagnosis or misdiagnosis is reduced, and the test efficiency is improved. Secondly, high-precision synchronization of signals and effective noise reduction of the adaptive noise reduction unit are realized through hardware-level timestamps, useful electromyographic signals and nerve conduction signals can be more accurately separated and extracted, interference factors are eliminated, and doctors can more clearly analyze signal characteristics and accurately judge neuromuscular function states; and misjudgment caused by signal interference is avoided.

Description

Technical Field

[0001] This invention belongs to the field of medical electronic equipment technology, specifically a wireless surface electromyography nerve conduction testing system. Background Technology

[0002] Surface electromyography (sEMG) records electrical signals during muscle activity using electrodes on the skin surface to assess muscle function, such as muscle fatigue, activation level, or the presence of neuromuscular diseases. Neuroconduction testing (NCS), on the other hand, assesses nerve conduction function by stimulating nerves and recording their responses, such as nerve damage and conduction velocity. The surface electromyography neuroconduction testing system is a testing system used in biomedical engineering and neuroelectrophysiology. Its working principle is mainly to induce motor or sensory nerve potentials through external electrical stimulation, and to collect neuromuscular signals at specific distances along the nerve pathway using surface electrodes. These signals are then amplified, processed, and sent to a host computer to obtain an electromyogram (EMG).

[0003] Traditional electromyography (EMG) devices use wired connections, which are prone to motion artifacts, leading to missed or misdiagnosis during the diagnostic process. In addition, wired connections can easily lead to misjudgments of the degree of muscle strength and coordination recovery during rehabilitation.

[0004] Therefore, the present invention provides a wireless surface electromyography nerve conduction testing system. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: a wireless surface electromyography nerve conduction testing system according to this invention, comprising:

[0007] A flexible wearable sensor module is used to simultaneously acquire surface electromyography signals and nerve conduction signals;

[0008] The multimodal signal processing module includes an adaptive noise reduction unit and a synchronization triggering unit, which realizes the time synchronization of the surface electromyography signal and the nerve conduction signal through hardware-level timestamps, with a synchronization error ≤0.5ms;

[0009] The wireless transmission and edge computing module supports low-power Bluetooth 5.3 protocol transmission and energy recovery power supply, and integrates a neural network processor for real-time calculation of time-domain and frequency-domain indicators.

[0010] The intelligent analysis terminal generates neuromuscular function assessment reports and personalized rehabilitation plans based on AI models.

[0011] The flexible wearable sensor module integrates a liquid metal surface electromyography electrode and a microneedle nerve stimulation electrode, with a sampling rate of 1000Hz to 2000Hz.

[0012] The liquid metal surface electromyography electrode is printed with gallium indium alloy on a polydimethylsiloxane substrate with a thickness of 0.2 mm to 0.4 mm. The radius of curvature of the contact surface with the skin matches the curvature of the human body, and the radius of curvature ranges from 50 mm to 200 mm.

[0013] The microneedle nerve stimulation electrode adopts an array-type microneedle structure, with a single microneedle height of 150μm to 250μm and a diameter of 30μm to 70μm. Painless nerve stimulation is performed through electroporation technology, with a stimulation threshold ≤0.8mA.

[0014] The electroporation stimulation waveform of the microneedle array is a biphasic pulse, and the pulse width is calculated using the following formula:

[0015]

[0016] Among them, C m =1μF / cm 2 I represents the nerve cell membrane capacitance, ΔV is the perforation threshold voltage, and ΔV ranges from 0.5V to 1V. stim The stimulation current is selected from 0.5mA to 0.8mA.

[0017] The liquid metal surface electromyography electrode and the microneedle nerve stimulation electrode are arranged in a coplanar layout, with a spacing of 2mm to 10mm between them. The flexible wearable sensor module array covers the target muscle group and the corresponding nerve pathway, with a coverage area ≤50cm². 2 .

[0018] The adaptive noise reduction unit of the multimodal signal processing module includes:

[0019] Wavelet transform filtering subunit is used to separate baseline drift of electromyographic signals;

[0020] Independent component analysis subunit, used to eliminate ECG interference, with a signal-to-noise ratio ≥35dB after noise reduction. The formula for calculating the signal-to-noise ratio is:

[0021]

[0022] Among them, P signal (f) is the power spectrum of the electromyographic signal, P noise (f) represents the noise power spectrum;

[0023] The synchronous triggering unit generates microsecond-level timestamps through the FPGA chip to synchronously trigger the acquisition timing of surface electromyography signals and nerve conduction signals.

[0024] The wireless transmission and edge computing unit includes:

[0025] The energy recovery subunit integrates piezoelectric materials and Seebeck effect thermoelectric modules to convert the mechanical energy of human movement and the heat energy of the body surface into electrical energy, with a power supply duration of ≥12 hours.

[0026] The edge computing subunit calculates time-domain and frequency-domain metrics in real time through the NPU, with a calculation latency of ≤50ms. The time-domain metrics include integrated electromyography values ​​and root mean square values, while the frequency-domain metrics include median frequency and average power frequency.

[0027] The formula for calculating the integral electromyographic signal is as follows:

[0028]

[0029] Wherein, EMG(t) is the instantaneous amplitude of the electromyographic signal, and t1 and t2 are the analysis time windows;

[0030] The formula for calculating the median frequency is as follows:

[0031]

[0032] Where P(f) is the power spectral density of the electromyographic signal, f max =500Hz is the highest frequency of the signal.

[0033] The piezoelectric material of the energy recovery subunit is polyvinylidene fluoride, and its piezoelectric constant d 33 ≥25pC / N; the thermoelectric module uses bismuth telluride-based material with a conversion efficiency ≥8%.

[0034] The intelligent analysis terminal includes:

[0035] The dynamic visualization module generates muscle activation time-series diagrams and 3D reconstruction models of nerve conduction pathways, supporting AR / VR interactive display.

[0036] The AI-assisted diagnostic module is based on a deep learning model with a Transformer architecture. The input data includes time-domain indicators, frequency-domain indicators, conduction parameters, sensory conduction velocity, and motor conduction velocity. The output is the type of neuropathy or the level of muscle fatigue.

[0037] Sensory conduction velocity was obtained by stimulating sensory nerve trunks and recording it at the distal end of the fingers or toes.

[0038] Motor conduction velocity is obtained by stimulating motor nerves and recording compound muscle action potentials on the muscles innervated by the nerves, based on the distance between the stimulation point and the recording point and the latency.

[0039] The intelligent analysis terminal also includes:

[0040] The rehabilitation plan generation module recommends electrical stimulation parameters and exercise training loads based on the assessment results.

[0041] The formula for the exercise training load is:

[0042]

[0043] Where k is the adjustment coefficient, IEMG target For the target electromyography integral value, IEMG current This is a real-time monitoring value.

[0044] The beneficial effects of this invention are as follows:

[0045] 1. The wireless surface electromyography (EMG) nerve conduction testing system of the present invention adopts a wireless transmission method, which avoids the restriction of movement by cables, reduces motion artifacts caused by cable traction, and makes the collected signals more realistically reflect muscle activity and nerve conduction, thereby reducing the risk of missed diagnosis or misdiagnosis. Secondly, the high-precision synchronization of signals is achieved through hardware-level timestamps and the effective noise reduction of the adaptive noise reduction unit, which can more accurately separate and extract useful EMG signals and nerve conduction signals, eliminate interference factors, and enable doctors to analyze signal characteristics more clearly, accurately judge the neuromuscular functional state, and avoid misjudgment caused by signal interference.

[0046] 2. The wireless surface electromyography nerve conduction testing system described in this invention calculates time-domain and frequency-domain indicators in real time through an edge computing subunit. Combined with the comprehensive analysis of multiple indicators by the AI-assisted diagnostic module, it can timely and accurately assess the degree of recovery of muscle strength and coordination. At the same time, the rehabilitation plan generation module recommends personalized electrical stimulation parameters and exercise training loads based on the assessment results, making rehabilitation training more targeted and scientific, and thus avoiding misjudgments caused by inaccurate assessments in traditional rehabilitation processes. Attached Figure Description

[0047] The invention will now be further described with reference to the accompanying drawings.

[0048] Figure 1 This is a system architecture diagram of the present invention;

[0049] Figure 2 This is a cross-sectional view of the flexible electrode structure;

[0050] Figure 3 This is a flowchart of the signal processing in this invention;

[0051] Figure 4 This is a schematic diagram of the energy recovery circuit in this invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Please see Figure 1-4 ,

[0054] This embodiment provides: a wireless surface electromyography (EMG) nerve conduction testing system, comprising:

[0055] A flexible wearable sensor module is used to simultaneously acquire surface electromyography signals and nerve conduction signals;

[0056] The flexible wearable sensor module integrates liquid metal surface electromyography electrodes and microneedle nerve stimulation electrodes, with a sampling rate of 1000Hz to 2000Hz.

[0057] The liquid metal surface electromyography electrode is printed with gallium indium alloy on a polydimethylsiloxane substrate with a thickness of 0.2mm to 0.4mm. The curvature radius of the contact surface with the skin matches the curvature of the human body, and the curvature radius ranges from 50mm to 200mm.

[0058] The microneedle nerve stimulation electrode adopts an array-type microneedle structure, with a single microneedle height of 150μm~250μm and a diameter of 30μm~70μm. Painless nerve stimulation is performed through electroporation technology, with a stimulation threshold ≤0.8mA.

[0059] The electroporation stimulation waveform of the microneedle array is a biphasic pulse, and the pulse width is calculated using the following formula:

[0060]

[0061] Among them, C m =1μF / cm 2 I represents the nerve cell membrane capacitance, ΔV is the perforation threshold voltage, and ΔV ranges from 0.5V to 1V. stim The stimulation current is selected from 0.5mA to 0.8mA.

[0062] The liquid metal surface electromyography electrode and the microneedle nerve stimulation electrode are arranged in a coplanar layout, with a spacing of 2mm to 10mm between them. The flexible wearable sensor module array covers the target muscle group and the corresponding nerve pathway, with a coverage area ≤50cm². 2 .

[0063] The impedance matching formula for a coplanar layout is:

[0064]

[0065] Among them, R skin The skin contact resistance ranges from 50Ω to 500Ω, R electrode X represents the impedance of the electrode material (≤1Ω for liquid metal electrodes). skin With X electrode These are the capacitive reactances of the skin and the electrodes, respectively.

[0066] It should be noted that the liquid metal surface electromyography electrode uses a gallium indium alloy (GaIn). 20 Printed on a polydimethylsiloxane (PDMS) substrate, PDMS has good flexibility and biocompatibility, which can better conform to the curvature of human skin. Compared with rigid electrode materials, flexible materials can deform with the slight deformation of the skin, reducing the relative movement between the electrode and the skin, thereby reducing the contact impedance fluctuation caused by the poor contact between the electrode and the skin.

[0067] The liquid metal surface electromyography (EMG) electrodes have a skin contact surface with a curvature radius matching the human body surface, ranging from 50mm to 200mm. This design allows the electrodes to fit more closely to the skin at different locations, further ensuring good contact between the electrode and the skin, reducing variations in contact impedance, and improving the stability of signal acquisition.

[0068] Microneedles can penetrate the stratum corneum of the skin and directly contact the tissue under the skin, reducing the influence of skin surface conditions (such as sweat, stratum corneum thickness, etc.) on the contact between the electrode and the tissue, thereby reducing the fluctuation of skin contact impedance;

[0069] Painless nerve stimulation can be achieved through electroporation, with a stimulation threshold of ≤0.8mA. Electroporation can create temporary pores in the cell membrane, making it easier for drugs or currents to act on nerve cells. At the same time, this technology can achieve effective nerve stimulation at a lower current, reducing skin irritation and damage, avoiding significant pain for patients, and helping to maintain a stable contact between the electrode and the skin.

[0070] Coplanar Layout: The liquid metal surface electromyography electrodes and microneedle nerve stimulation electrodes are coplanarly arranged with reasonable spacing, and the sensor module array covers the target muscle group and the corresponding nerve pathway, with a coverage area ≤50cm². 2 This ensures the synchronization and accuracy of signal acquisition; for example, when detecting arm muscle and nerve function, it can fully cover the relevant areas and obtain complete signal information.

[0071] The multimodal signal processing module, including an adaptive noise reduction unit and a synchronization triggering unit, achieves time synchronization between surface electromyography signals and nerve conduction signals through hardware-level timestamps, with a synchronization error ≤0.5ms;

[0072] The adaptive noise reduction unit of the multimodal signal processing module includes:

[0073] The wavelet transform filtering subunit uses the Daubechies 4 (db4) wavelet basis to perform a 5-level decomposition on the surface electromyography (sEMG) signal, removing low-frequency interference from the signal and effectively separating the baseline drift of the electromyography signal; for example, when monitoring electromyography signals for a long time, the influence of baseline drift on signal analysis can be avoided.

[0074] It should be noted that these signal processing techniques can effectively remove noise and interference from signals, improve signal quality, and ensure that the acquired signals have high usability even when there are certain fluctuations in skin contact impedance.

[0075] Independent component analysis subunit is used to eliminate ECG interference. ECG interference is eliminated by FastICA algorithm. The signal-to-noise ratio after noise reduction is ≥35dB. The formula for calculating the signal-to-noise ratio is:

[0076]

[0077] Among them, P signal (f) is the power spectrum of the electromyographic signal, P noise (f) represents the noise power spectrum;

[0078] The synchronization triggering unit generates microsecond-level timestamps through the FPGA chip, synchronously triggering the acquisition timing of surface electromyography signals and nerve conduction signals. The timestamp accuracy formula is:

[0079]

[0080] Among them, t clock The FPGA clock cycle ranges from 1μs to 2μs.

[0081] The wireless transmission and edge computing module supports low-power Bluetooth 5.3 protocol transmission and energy recovery power supply, and integrates a neural network processor for real-time calculation of time-domain and frequency-domain indicators.

[0082] It is worth noting that the microsecond-level timestamps generated by the FPGA chip enable the synchronous acquisition of surface electromyography (EMG) signals and nerve conduction signals. This hardware-level time synchronization method avoids the timing asynchrony problem caused by separate measurements. Furthermore, when measuring motor and sensory conduction velocities (MCVs), it can accurately record the time relationship between stimulus and response, thus accurately calculating nerve conduction velocities. For example, when measuring MCV, after stimulating the motor nerve, the synchronous triggering unit ensures that the time point for recording the compound muscle action potential on the muscle innervated by the nerve precisely corresponds to the stimulation time. The MCV obtained based on the distance between the stimulation point and the recording point, as well as the latency, can truly reflect the actual speed of nerve conduction, reducing errors introduced by timing asynchrony and providing accurate data support for diagnosing neuropathy. Secondly, the flexible wearable sensor module integrates liquid metal surface EMG electrodes and microneedle nerve stimulation electrodes, enabling the synchronous acquisition of surface EMG signals and nerve conduction signals. This integrated design allows the measurement process to be performed on the same device and in the same measurement environment, reducing errors caused by different devices, different measurement environments, and separate operations, ensuring the consistency and accuracy of the measurement results.

[0083] The wireless transmission and edge computing unit includes:

[0084] The energy recovery subunit integrates piezoelectric materials and a Seebeck effect thermoelectric module to convert the mechanical energy of human movement and the heat energy of the body surface into electrical energy, providing a power supply time of ≥12 hours. This solves the power supply problem of the device, improving its portability and usage time. For example, in outdoor rehabilitation monitoring scenarios, it eliminates the need for frequent battery replacements and can operate continuously for extended periods.

[0085] The edge computing subunit calculates time-domain and frequency-domain metrics in real time through the NPU, with a calculation latency of ≤50ms. The time-domain metrics include integrated electromyography values ​​and root mean square values, while the frequency-domain metrics include median frequency and average power frequency.

[0086] The formula for calculating the integrated electromyographic signal is:

[0087]

[0088] Wherein, EMG(t) is the instantaneous amplitude of the electromyographic signal, and t1 and t2 are the analysis time windows;

[0089] The formula for calculating the median frequency is:

[0090]

[0091] Where P(f) is the power spectral density of the electromyographic signal, f max =500Hz is the highest frequency of the signal.

[0092] The piezoelectric material of the energy recovery subunit is polyvinylidene fluoride, and its piezoelectric constant d 33 ≥25pC / N; The thermoelectric module uses bismuth telluride-based materials with a conversion efficiency of ≥8%.

[0093] The intelligent analysis terminal generates neuromuscular function assessment reports and personalized rehabilitation plans based on AI models.

[0094] The intelligent analysis terminal includes:

[0095] The dynamic visualization module generates muscle activation time sequence diagrams and 3D reconstruction models of nerve conduction pathways, supporting AR / VR interactive display, enabling doctors and patients to more intuitively understand the state of neuromuscular function.

[0096] For example, during rehabilitation treatment, patients can clearly see the activation status of their muscles and the nerve conduction pathways through AR / VR devices, which enhances their understanding and confidence in the treatment.

[0097] The AI-assisted diagnostic module is based on a deep learning model with a Transformer architecture. The input data includes time-domain indicators, frequency-domain indicators, conduction parameters, sensory conduction velocity, and motor conduction velocity. The output is the type of neuropathy or the level of muscle fatigue, with high diagnostic accuracy.

[0098] For example, during rehabilitation treatment, patients can clearly see the activation status of their muscles and the nerve conduction pathways through AR / VR devices, which enhances their understanding and confidence in the treatment.

[0099] Sensory conduction velocity (SCV) is obtained by stimulating sensory nerve trunks and recording at the distal end of the fingers or toes. It includes anterograde and retrograde methods. The anterograde method involves stimulating the distal end of the sensory nerve and recording sensory action potentials (SNAPs) near the spinal cord; the retrograde method involves stimulating the sensory nerve trunk and recording at the distal end, such as the fingers or toes. The sensory conduction velocity is calculated based on the distance between the stimulation point and the recording point, as well as the latency.

[0100] Motor conduction velocity is obtained by stimulating motor nerves and recording compound muscle action potentials on the muscles innervated by the nerves. It is based on the distance between the stimulation point and the recording point and the latency (the time from stimulation to the occurrence of an action potential). The formula is: Motor conduction velocity, i.e., MCV = distance / latency difference (the distance between stimulation points divided by the difference in latency between two different stimulation points that elicit action potentials).

[0101] It is important to note that abnormalities in the median ventricle (MCV) may indicate motor neuron disease, such as peripheral nerve injury or motor neuron disease. For example, in carpal tunnel syndrome, compression of the median nerve at the wrist can lead to a slowing of its motor conduction velocity.

[0102] Changes in SCV are commonly seen in sensory neuropathy, such as diabetic peripheral neuropathy. Patients often first experience abnormal sensory conduction velocity, manifesting as sensory abnormalities such as numbness and tingling in the hands and feet.

[0103] Specifically, the ability to simultaneously measure and accurately calculate nerve conduction velocity provides precise data for assessing an athlete's neuromuscular control ability. For example, by accurately measuring MCV and SCV, we can understand whether nerve conduction function is normal, determine whether an athlete has a potential risk of nerve injury, and provide effective support for sports training and injury prevention.

[0104] The intelligent analysis terminal also includes:

[0105] The rehabilitation plan generation module recommends electrical stimulation parameters and exercise training loads based on the assessment results.

[0106] The formula for exercise training load is:

[0107]

[0108] Where k is the adjustment coefficient, IEMG target For the target electromyography integral value, IEMG current To monitor values ​​in real time and provide personalized rehabilitation plans for patients based on these values;

[0109] For example, different training loads can be designed for patients with different degrees of muscle damage to improve rehabilitation outcomes.

[0110] It's worth noting that the rehabilitation plan generation module of the intelligent analysis terminal recommends electrical stimulation parameters and exercise training loads based on the assessment results. In sports medicine, this allows for the development of personalized training plans based on an athlete's neurological function, helping to improve neuromuscular control and prevent sports injuries. Simultaneously, it can also provide targeted rehabilitation programs to promote the recovery of nerve function when athletes experience nerve damage.

[0111] Workflow:

[0112] First, the flexible wearable sensor module integrates liquid metal surface electromyography (EMG) electrodes and microneedle nerve stimulation electrodes. The liquid metal surface EMG electrodes are printed with gallium indium alloy onto a polydimethylsiloxane (PDMS) substrate. The radius of curvature of the surface contacting the skin matches the human body's surface curvature, ranging from 50mm to 200mm. This design allows for close contact with different areas of the skin, reducing contact impedance variations and improving signal acquisition stability. Simultaneously, the excellent flexibility and biocompatibility of PDMS allow the electrodes to deform with slight skin deformations, reducing contact impedance fluctuations caused by relative movement between the electrodes and skin. The microneedle nerve stimulation electrodes employ an array of microneedles. The height of a single microneedle can penetrate the stratum corneum and directly contact subcutaneous tissue, reducing the influence of skin surface conditions on electrode-tissue contact and thus lowering skin contact impedance fluctuations. The liquid metal surface EMG electrodes and microneedle nerve stimulation electrodes are coplanarly arranged, and the sensor module array covers the target muscle group and corresponding nerve pathways, with a coverage area ≤50cm². 2 This ensures the synchronization and accuracy of signal acquisition;

[0113] Next, the microneedle nerve stimulation electrode uses electroporation technology to provide painless nerve stimulation with a stimulation threshold ≤0.8mA. Electroporation technology can create temporary pores in the cell membrane, making it easier for the current to act on nerve cells, achieving effective nerve stimulation at a lower current, reducing skin irritation and damage, avoiding significant pain for the patient, and helping to maintain a stable contact between the electrode and the skin. Simultaneously, a flexible wearable sensor module synchronously acquires surface electromyography signals and nerve conduction signals at a sampling rate of 1000Hz-2000Hz.

[0114] Subsequently, the acquired signals enter the multimodal signal processing module, where the adaptive noise reduction unit begins to function. The wavelet transform filtering subunit uses the Daubechies 4 (db4) wavelet basis to perform a 5-level decomposition of the surface electromyography (sEMG) signal, removing low-frequency interference and effectively separating baseline drift, thus avoiding the impact of baseline drift on signal analysis during long-term monitoring. The independent component analysis subunit eliminates electrocardiogram (ECG) interference using the FastICA algorithm, achieving a signal-to-noise ratio ≥35dB after noise reduction, improving signal quality and ensuring high usability of the acquired signals even with fluctuations in skin contact impedance. The synchronization triggering unit generates microsecond-level timestamps using an FPGA chip to synchronously trigger the acquisition timing of surface electromyography and nerve conduction signals, with a synchronization error ≤0.5ms. This achieves hardware-level time synchronization, avoiding timing asynchrony issues caused by multiple measurements. When measuring motor conduction velocity (MCV) and sensory conduction velocity (SCV), it can accurately record the time relationship between stimulus and response, thereby accurately calculating nerve conduction velocity.

[0115] The signal is then transmitted to the wireless transmission and edge computing module. This module's energy recovery subunit integrates piezoelectric materials and a Seebeck effect thermoelectric module. The thermoelectric module uses bismuth telluride-based materials with a conversion efficiency ≥8%. These materials convert the mechanical energy of human movement and surface heat into electrical energy, providing a power supply duration of ≥12 hours, thus solving the device's power supply problem and improving its portability and usage time. The edge computing subunit uses a neural network processor (NPU) to calculate time-domain and frequency-domain metrics in real time, with a calculation latency ≤50ms. Time-domain metrics include integrated electromyography (EMG) values ​​and root mean square (RMS) values, while frequency-domain metrics include median frequency and average power frequency.

[0116] Finally, the signal reaches the intelligent analysis terminal. The dynamic visualization module generates a muscle activation time sequence diagram and a 3D reconstruction model of the nerve conduction path, supporting AR / VR interactive display, allowing doctors and patients to more intuitively understand the neuromuscular functional state. The AI-assisted diagnosis module, based on a deep learning model with a Transformer architecture, takes into account time-domain indicators, frequency-domain indicators, conduction parameters, sensory conduction velocity, and motor conduction velocity, and outputs the type of neuropathy or the level of muscle fatigue, achieving high diagnostic accuracy. The rehabilitation plan generation module recommends electrical stimulation parameters and exercise training loads based on the assessment results. The exercise training load is calculated according to a formula, providing patients with personalized rehabilitation plans. In sports medicine, personalized training plans can be developed based on the athlete's neurological functional state to help improve neuromuscular control and prevent sports injuries. When athletes experience nerve damage, targeted rehabilitation plans can also be provided to promote the recovery of nerve function.

[0117] In the description of this invention, it should be understood that the terms "center", "longitudinal", "lateral", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of this invention.

[0118] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A wireless surface electromyography (EMG) nerve conduction testing system, characterized in that, include: A flexible wearable sensor module is used to simultaneously acquire surface electromyography signals and nerve conduction signals; The multimodal signal processing module includes an adaptive noise reduction unit and a synchronization triggering unit, which realizes the time synchronization of the surface electromyography signal and the nerve conduction signal through hardware-level timestamps, with a synchronization error ≤0.5ms; The wireless transmission and edge computing module supports low-power Bluetooth 5.3 protocol transmission and energy recovery power supply, and integrates a neural network processor for real-time calculation of time-domain and frequency-domain indicators. The intelligent analysis terminal generates neuromuscular function assessment reports and personalized rehabilitation plans based on AI models.

2. The wireless surface electromyography nerve conduction testing system according to claim 1, characterized in that: The flexible wearable sensor module integrates a liquid metal surface electromyography electrode and a microneedle nerve stimulation electrode, with a sampling rate of 1000Hz to 2000Hz.

3. The wireless surface electromyography nerve conduction testing system according to claim 2, characterized in that: The liquid metal surface electromyography electrode is printed with gallium indium alloy on a polydimethylsiloxane substrate with a thickness of 0.2 mm to 0.4 mm. The radius of curvature of the contact surface with the skin matches the curvature of the human body, and the radius of curvature ranges from 50 mm to 200 mm.

4. The wireless surface electromyography nerve conduction testing system according to claim 2, characterized in that: The microneedle nerve stimulation electrode adopts an array-type microneedle structure, with a single microneedle height of 150μm to 250μm and a diameter of 30μm to 70μm. Painless nerve stimulation is performed through electroporation technology, with a stimulation threshold ≤0.8mA. The electroporation stimulation waveform of the microneedle array is a biphasic pulse, and the pulse width is calculated using the following formula: Among them, C m =1μF / cm 2 I represents the capacitance of the nerve cell membrane, ΔV is the perforation threshold voltage, and ΔV ranges from 0.5V to 1V. stim The stimulation current is selected from 0.5mA to 0.8mA. The liquid metal surface electromyography electrode and the microneedle nerve stimulation electrode are arranged in a coplanar layout, with a spacing of 2mm to 10mm between them. The flexible wearable sensor module array covers the target muscle group and the corresponding nerve pathway, with a coverage area ≤ 50cm². 2 .

5. The wireless surface electromyography nerve conduction testing system according to claim 1, characterized in that: The adaptive noise reduction unit of the multimodal signal processing module includes: Wavelet transform filtering subunit is used to separate baseline drift of electromyographic signals; Independent component analysis subunit, used to eliminate ECG interference, with a signal-to-noise ratio ≥35dB after noise reduction. The formula for calculating the signal-to-noise ratio is: Among them, P signal (f) is the power spectrum of the electromyographic signal, P noise (f) represents the noise power spectrum.

6. The wireless surface electromyography nerve conduction testing system according to claim 1, characterized in that: The synchronous triggering unit generates microsecond-level timestamps through the FPGA chip to synchronously trigger the acquisition timing of surface electromyography signals and nerve conduction signals.

7. The wireless surface electromyography nerve conduction testing system according to claim 1, characterized in that: The wireless transmission and edge computing unit includes: The energy recovery subunit integrates piezoelectric materials and Seebeck effect thermoelectric modules to convert the mechanical energy of human movement and the heat energy of the body surface into electrical energy, with a power supply duration of ≥12 hours. The edge computing subunit calculates time-domain and frequency-domain metrics in real time through the NPU, with a calculation latency of ≤50ms. The time-domain metrics include integrated electromyography values ​​and root mean square values, while the frequency-domain metrics include median frequency and average power frequency. The formula for calculating the integral electromyographic signal is as follows: Wherein, EMG(t) is the instantaneous amplitude of the electromyographic signal, and t1 and t2 are the analysis time windows; The formula for calculating the median frequency is: Where P(f) is the power spectral density of the electromyographic signal, f max =500Hz is the highest frequency of the signal.

8. The wireless surface electromyography nerve conduction testing system according to claim 7, characterized in that: The piezoelectric material of the energy recovery subunit is polyvinylidene fluoride, and its piezoelectric constant d 33 ≥25pC / N; the thermoelectric module uses bismuth telluride-based material with a conversion efficiency ≥8%.

9. The wireless surface electromyography nerve conduction testing system according to claim 1, characterized in that: The intelligent analysis terminal includes: The dynamic visualization module generates muscle activation time-series diagrams and 3D reconstruction models of nerve conduction pathways, supporting AR / VR interactive display. The AI-assisted diagnostic module is based on a deep learning model with a Transformer architecture. The input data includes time-domain indicators, frequency-domain indicators, conduction parameters, sensory conduction velocity, and motor conduction velocity. The output is the type of neuropathy or the level of muscle fatigue. Sensory conduction velocity was obtained by stimulating sensory nerve trunks and recording it at the distal end of the fingers or toes. Motor conduction velocity is obtained by stimulating motor nerves and recording compound muscle action potentials on the muscles innervated by the nerves, based on the distance between the stimulation point and the recording point and the latency.

10. The wireless surface electromyography nerve conduction testing system according to claim 1, characterized in that: The intelligent analysis terminal also includes: The rehabilitation plan generation module recommends electrical stimulation parameters and exercise training loads based on the assessment results. The formula for the exercise training load is: Where k is the adjustment coefficient, IEMG target For the target electromyography integral value, IEMG current This is a real-time monitoring value.

Images