Biofeedback electrical stimulation device and method for quadriceps femoris rehabilitation training

Abstract

The invention provides a biofeedback electrical stimulation device and method for quadriceps femoris rehabilitation training. The device comprises a host, an integrated electrode slice, a universal electrode swivel and a wireless myoelectricity module. The wireless myoelectricity module comprises a myoelectricity power supply module, a myoelectricity main control module, a posture sensing module, a wireless connection module and an electrode assembly, the electrode assembly comprises at least three surface myoelectricity collecting electrodes and an electrical stimulation output electrode, and the myoelectricity main control module wirelessly transmits collected myoelectricity data to the host; and after calculation of the host, stimulation parameters are wirelessly transmitted to the electrical stimulation output electrode, and wireless biofeedback electrical stimulation is completed. The NMES technology and the biological feedback technology are combined to treat the quadriceps femoris atrophy after the ligament operation, stimulation parameters are automatically adjusted, and a common patient can easily use the device. By means of the innovation, the use threshold is lowered, risks caused by misoperation are reduced, and meanwhile the rehabilitation speed of the patient is remarkably increased.

Description

Technical Field

[0001] The present invention relates to the technical field of sports rehabilitation, and particularly to a biofeedback electrical stimulation device and method for quadriceps femoris rehabilitation training. Background Art

[0002] The incidence of knee joint sports injuries is high. After knee joint injuries, quadriceps femoris atrophy is the most common complication, and it is difficult to restore the dimension of the quadriceps femoris, which has an important impact on the postoperative rehabilitation process of sports people.

[0003] Knee joint injuries are common types of sports injuries, mainly including anterior cruciate ligament (ACL) injuries, posterior cruciate ligament (PCL) injuries, medial collateral ligament (MCL) injuries, meniscus injuries, patellar dislocations, etc. With the popularization of the national fitness campaign in China, the incidence of knee joint sports injuries has increased year by year. Knee joint injuries may result from various injury mechanisms, such as high-energy impacts, low-energy sprains, car accidents, contact or non-contact injuries during sports, etc. Among them, non-contact injuries are the most common, especially occurring in actions such as rotation and jumping. Knee joint sports injuries can not only cause knee joint pain, swelling and joint instability, but also secondary traumatic osteoarthrosis, etc., thus seriously affecting the athlete's professional career.

[0004] At present, the treatment options for knee joint sports injuries mainly include surgical and non-surgical treatments. The specific choice needs to be determined by combining multiple factors such as the patient's combined injuries, risk factors, as well as the patient's age, weight, activity level and expected goals. Research shows that compared with non-surgical treatment, during the 1-year, 2-year, 5-year, and 10-year follow-ups after ACL reconstruction, the patients' symptoms have been significantly improved, and the knee joint function and quality of life have been significantly improved. PCL injuries require brace fixation in the extended position in the first 2-4 weeks to prevent excessive posterior displacement of the tibia. For patients with high-intensity sports needs, especially those who hope to return to sports involving sudden stops and turns, or those who have experienced knee joint instability multiple times, combined with early osteoarthritis or cartilage injuries, meniscus injuries, it is recommended to receive surgical treatment.

[0005] Common complications after knee joint surgery or long-term brace fixation include quadriceps atrophy, limited range of motion of the knee joint, knee joint infection, and deep vein thrombosis. Among them, infections, deep vein thrombosis, etc. can be routinely prevented, and limited range of motion of the knee joint can be intervened by means of early functional exercise, manual release, or arthroscopic release when necessary. In contrast, it is difficult for the strength and muscle dimension of the patient's quadriceps to fully recover within 6 months after surgery or before returning to sports, which has a negative impact on the postoperative rehabilitation process. Epidemiological statistics show that quadriceps weakness may still be observed 10 to 20 years after anterior cruciate ligament reconstruction; for PCL patients treated conservatively, the recovery of quadriceps strength also plays an important role in the recovery of knee joint stability. Therefore, although surgery can reconstruct the structural stability of the knee joint, it is not sufficient to completely restore its functional stability, and lower limb muscle atrophy and muscle strength decline may still exist for a long time after surgery. Therefore, the final outcome of the treatment of knee joint sports injuries depends not only on the success of the surgery but also on scientific and systematic postoperative rehabilitation.

[0006] Muscles play an important role in maintaining the dynamic stability of joints. If muscle strength is insufficient, it will affect basic activities such as standing, walking, and running. The quadriceps is composed of the rectus femoris, vastus intermedius, vastus medialis, and vastus lateralis, and is the main power source during knee joint extension. Compared with the knee-flexing muscle groups, the strength of the quadriceps usually decreases more significantly after knee joint surgery. In the early postoperative period, due to factors such as joint pain, swelling, and other soft tissue injuries of the affected limb, it often leads to arthrogenic muscle inhibition (AMI), further aggravating quadriceps atrophy, resulting in a decline in lower limb mechanical properties, and even inducing knee osteoarthritis (KOA).

[0007] Muscle activation is one of the important indicators for evaluating muscle function and activity level. The failure of quadriceps activation will reduce its ability of voluntary muscle contraction, thus showing muscle weakness. After knee joint surgery, there are often differences in insufficient activation of the quadriceps in different regions. For example, during activities such as squatting and going downstairs, the activation of the vastus medialis may be significantly lower than that of the vastus lateralis, which may be one of the important factors leading to postoperative knee joint pain. Therefore, if the rehabilitation plan is only designed for the surgical area, it may not be sufficient. Starting from the overall force generation pattern changes of the quadriceps and carrying out targeted rehabilitation interventions on its atrophied parts are of great significance for improving the quality of knee joint postoperative rehabilitation.

[0008] Closed-chain exercises are effective methods commonly used in early rehabilitation, which can strengthen the hamstrings and quadriceps simultaneously. Straight leg raise training, isometric contraction of the quadriceps, and ankle pump exercises can be started immediately after surgery as they do not impose excessive stress on the graft. However, if the stress load on the patellofemoral joint is too high during training, it may trigger patellofemoral pain syndrome (PFJ). Meanwhile, the compensatory effect of lower limb biomechanics will also affect the rehabilitation process. Therefore, if a treatment strategy can be found that can efficiently improve quadriceps strength and muscle dimension, reduce the risk of graft injury, and avoid secondary problems such as PFJ, it will help patients achieve better postoperative rehabilitation outcomes.

[0009] Existing studies have shown that neuromuscular electrical stimulation can enhance muscle strength, exogenously activate the quadriceps, and effectively improve the rehabilitation effect after knee surgery.

[0010] Muscles are the power source of movement, and the generation of movements depends on the coordinated action of the neuromuscular system. During muscle contraction, a low-frequency, low-amplitude bioelectrical signal with a frequency of 20 - 2000 Hz and an amplitude of approximately 50 μV - 20 mV is generated, and this signal is called electromyogram (EMG). As the muscle movement state changes, the EMG signal also changes accordingly. By analyzing the EMG signal, we can infer the muscle activity information and further evaluate the muscle recovery level.

[0011] In clinical practice, to obtain accurate EMG signals, traditional methods usually require inserting needle electrodes directly into the muscle. However, this invasive measure is not only complex to operate but also highly traumatic, bringing pain and infection risks to patients; for cases that require continuous monitoring, it may also increase the discomfort and operational difficulty of patients. Therefore, the practical application scenario of this method is relatively limited. In contrast, the acquisition method of surface electromyography (sEMG) is more widely adopted due to its advantages such as simplicity, non-invasiveness, and painlessness. Just placing surface electrodes on the body surface can collect the sum of all EMG signals within its coverage area. Through post-processing of the signals, the EMG signals of specific muscles can also be extracted to meet the needs of targeted analysis.

[0012] It should be noted that when muscles contract actively, electromyographic signals are generated. Similarly, muscle contractions can also be induced by externally applying current, namely neuromuscular electrical stimulation (NMES). NMES can directly depolarize motor axons, causing involuntary muscle contractions. In the early postoperative period of knee joint surgery or during the brace fixation stage, patients usually experience the phenomenon of failed quadriceps activation. At this time, exogenous stimulation through NMES can effectively activate the quadriceps and overcome this problem. Previous studies have shown that compared with simple exercise training, the comprehensive therapy of NMES combined with exercise has more significant advantages in improving muscle strength. Using NMES in the early postoperative period helps reduce the atrophy of MHC type II skeletal muscle fibers and maintain the contractile force of MHC type I fibers. Therefore, exogenous activation of the quadriceps through NMES can not only enhance muscle strength without affecting graft remodeling, but also further improve the rehabilitation effect after knee joint surgery.

[0013] Biofeedback technology can provide real-time feedback to patients, improve the pertinence and accuracy of rehabilitation training, and enhance the patient's ability to control muscles.

[0014] Biofeedback technology is a modern technology that integrates physical and mental therapies, and can real-time feedback the physiological or pathological information in the body to patients through physiological science instruments. After training, patients can adjust their own states through conscious mental control and psychological regulation, so as to relieve or eliminate pathological processes. The electromyogram biofeedback system can measure electromyographic activity signals and provide real-time feedback to patients, reducing or enhancing muscle activity according to the diagnosis results. In the past decade, a large number of studies have supported the application of biofeedback technology in the rehabilitation of the nervous system and musculoskeletal system. Combining biofeedback therapy with traditional quadriceps training can effectively relieve the symptoms of patellofemoral joint pain and enhance muscle strength. For patients with limited knee joint movement after ACLR, biofeedback therapy also has significant curative effects.

[0015] After knee joint surgery, due to limb swelling, pain, and weakened joint receptor activity, the activation of joint muscle motor units will be negatively affected. To restore quadriceps function, patients usually need to perform exercises such as straight leg raises. However, such exercises may cause patients to compensate with hip muscles due to pressure on the postoperative wound, thus neglecting the effective contraction of the knee extensor muscles in the lower limb. In addition to muscle compensation caused by pain, the surgical incision may also affect the physiological activities of the joint. Krebs et al. believe that the tissue trauma caused by the open anterior approach during ACLR will lead to temporary dysregulation of joint capsule receptor activity, which will last for three weeks or even longer. These receptor activities play an important role in maintaining the normal function of the quadriceps, and their dysregulation may lead to abnormal perception of muscle control by patients, thereby weakening the effect of rehabilitation training.

[0016] Therefore, if patients can intuitively feel the results of muscle contraction or relaxation in an auditory or visual way, they can accurately improve their muscle control ability, adjust the muscle force state according to the real-time feedback results, and thus enhance the effect of rehabilitation training. This is exactly the strength of biofeedback technology. Moreover, muscles will develop tolerance to continuous electrical stimulation of the same frequency. Biofeedback technology can detect the muscle fatigue state and feedback it to the stimulation electrode in real time, change the stimulation frequency, and complete the electrical stimulation rehabilitation treatment more efficiently.

[0017] The existing biofeedback electrical stimulation therapy instruments are expensive, large in size, low in intelligence, single in function, and difficult to operate. It is urgent to develop wearable intelligent devices.

[0018] Currently, the biofeedback electrical stimulation therapy instruments on the market are mainly divided into two types: full-size desktop and portable. Hospital rehabilitation departments or rehabilitation centers usually equip large full-size terminals. Such devices usually consist of a main unit, a display, etc., and are installed on a movable equipment cart with universal wheels for flexible movement within a small range. The full-size terminal has advantages such as rich functions, complete interfaces, and good user experience, but its disadvantages are also very obvious:

[0019] 1. The high price makes it difficult for even professional rehabilitation institutions to equip them on a large scale, and at the same time, the long payback period further increases the cost of single treatment;

[0020] 2. The operation is complex, the parameter setting is professional, and it needs to be operated by professional physicians or rehabilitation therapists. Ordinary patients are difficult to use by themselves;

[0021] 3. Patients need to go to professional institutions for rehabilitation treatment, and the single rehabilitation time is short. For patients who are far from the rehabilitation institution or unable to go regularly due to physical reasons, this mode has poor popularization.

[0022] The main target markets of portable biofeedback electrical stimulation treatment devices are outpatient rehabilitation consulting rooms and home environments. Their design focuses more on portability rather than comprehensive functionality. Compared with full-sized terminals, the volume and price of such devices have been significantly reduced, but they still do not reach the level of being completely "portable at any time". Although the selling price has dropped to an acceptable range for home users, it is still not considered "affordable for the common people". Currently, the portable biofeedback electrical stimulation treatment devices on the market are mostly used for diseases such as stroke hemiplegia, cerebral infarction, or urinary and fecal incontinence. Whether their electrical stimulation mode is applicable to the treatment of quadriceps atrophy after knee surgery still requires further clinical verification.

[0023] In addition, there are already many pure NMES treatment devices, sEMG monitors, limb posture monitors, etc. on the market. These devices are small in size and low in price, and can even be worn without being noticed. However, they have few built-in presets, usually only have a single function, and have a low degree of intelligence, making it difficult to meet the wide-ranging needs of different patients. More importantly, the stimulation pulse parameters of these NMES treatment devices are fixed, and the muscles are prone to fatigue because they quickly adapt to the rhythm; the sEMG monitor can only detect sEMG signals, and the results are difficult for ordinary users to interpret; the limb posture monitor is widely used in fields such as VR games and motion capture and has not been popularized in the field of rehabilitation treatment. Therefore, the development of a multi-functional portable biofeedback combined electrical stimulation treatment device suitable for quadriceps atrophy after knee surgery is extremely urgent. Summary of the Invention

[0024] In order to solve the problems of the prior art, the present invention provides a biofeedback electrical stimulation device and method for quadriceps rehabilitation training

[0025] The present invention provides a biofeedback electrical stimulation device for quadriceps rehabilitation training, including a main unit, an integrated electrode patch, a universal electrode adapter, and a wireless EMG module; the integrated electrode patch includes a housing and an electrical stimulation output electrode, a surface EMG acquisition electrode, and a transmission module arranged in the housing. A skin attachment surface is provided on the housing, and the skin attachment surface is adhered to the clean skin through a conductive hydrogel. The surface EMG signals collected by the electrical stimulation output electrode and the surface EMG acquisition electrode are transmitted to the main unit through the transmission module; the wireless EMG module includes an EMG power module, an EMG main control module, a posture sensing module, a wireless connection module, and an electrode assembly. The electrode assembly includes at least three surface EMG acquisition electrodes and one electrical stimulation output electrode. The EMG main control module wirelessly transmits the collected EMG data to the main unit, and the main unit wirelessly transmits the stimulation parameters to the electrical stimulation output electrode after calculation to complete wireless biofeedback electrical stimulation.

[0026] The integrated electrode and the universal adapter are connected to the host through physical interfaces, and only one of them can be connected at the same time. The physical interfaces include a pair of full-function USB Type-C interfaces (the host side is a USB Type-C female port, and the integrated electrode and the universal adapter module are USB Type-C male ports) and a pair of edge quick-release buckles. When connecting, press the quick-release buckles at both ends of the integrated electrode or the universal adapter with both hands until the buckles open, insert the USB Type-C interfaces into the top, and then release both hands to lock them; when disassembling, press the quick-release buckles at both ends of the integrated electrode or the universal adapter with both hands until the buckles open, and then pull out the integrated electrode or the universal adapter to complete the disassembly. The wireless EMG module is wirelessly connected to the host. The device pairing can be completed through the button on the edge of the wireless EMG module. One host can be connected to multiple wireless modules, and it can also be used simultaneously with the integrated electrode and the universal adapter module. The wireless module can be placed at any position and can be networked with the connected device to cooperate in electrical stimulation or data acquisition.

[0027] For further improvement, the host includes a main control module, a host power module, a wireless interconnection module, an attitude sensing module, a surface EMG signal preprocessing module, an electrical stimulation output module, and a signal output module connected to the main control module. Among them, the wireless interconnection module is wirelessly connected to the wireless EMG module, and the surface EMG signal preprocessing module and the electrical stimulation output module are connected to the integrated electrode patch or the universal electrode head.

[0028] For further improvement, the universal electrode head is configured with a number of ordinary electrode interfaces to customize the surface EMG acquisition and electrical stimulation output channels.

[0029] For further improvement, the signal output module includes a display output module and a speaker module, and the speaker module is provided with a vibration motor.

[0030] The present invention also provides a biofeedback electrical stimulation method for quadriceps rehabilitation training, including the following steps:

[0031] 1) Set the electrical stimulation parameters;

[0032] 2) Acquisition period, surface EMG signal acquisition and processing;

[0033] 2.1) Surface EMG signal acquisition: Pre-acquire the surface EMG signals sEMG of the patient's affected limb in the relaxed state and the maximum voluntary contraction MVC state, and record the root mean square value RMS of the EMG r , RMS m ; At the same time, determine the following stimulation parameters:

[0034] Minimum stimulation parameters: The minimum stimulation amplitude A min and the minimum stimulation pulse width W min ;

[0035] Maximum stimulation parameter: maximum stimulation amplitude A without causing pain max and maximum stimulation pulse width W max ;

[0036] After determining the stimulation parameters, the stimulation pulse amplitude range is set at A min -A max , and the stimulation pulse width range is W min -W max ;

[0037] 2.2) Surface electromyogram signal processing;

[0038] 3) Stimulation cycle, biofeedback electrical stimulation: When performing neuromuscular electrical stimulation, the stimulation parameters are adjusted in real time to optimize the passive contraction effect of the quadriceps femoris, and the electrical stimulation is stopped in a timely manner when muscle fatigue is detected. The specific process is as follows: Calculate the fatigue characteristic Q value by combining non-linear characteristic parameters and time-domain characteristic parameters for real-time detection of muscle fatigue status:

[0039]

[0040] where u s is the sample entropy value of the electromyogram signal at the stimulation site, and u i is the integral value of the electromyogram signal at the stimulation site;

[0041] The sample entropy is used to describe the complexity of the time series. The smaller the value, the lower the complexity of the time series; when the muscle is fatigued, the muscle change rate slows down, and the activities of motor units tend to be consistent, resulting in a decrease in the sample entropy. At the same time, as the fatigue level deepens, the muscle gradually tenses, and more motor units will be recruited at the fatigued site, increasing the discharge frequency and synchronization degree, thus increasing the integrated electromyogram value; the characteristic Q value in the fatigue state will be significantly smaller than the initial state and continue to decrease as the fatigue level deepens. When it is detected that the Q value is lower than a certain specific threshold, it can be determined that the muscle enters the fatigue state and the electrical stimulation output is stopped.

[0042] For further improvement, to avoid the interference of the stimulation electrode on the surface electromyogram signal acquisition, the acquisition cycle in step 2) and the stimulation cycle in step 3) run alternately;

[0043] During the acquisition cycle, collect the surface electromyogram signal and calculate its characteristic values RMS(k), the pulse amplitude A(k + 1), pulse width W(k + 1) and fatigue characteristic Q(k) of the next cycle. When Q(k) is greater than the fatigue threshold, the electrical stimulation module outputs electrical stimulation according to the values of A(k + 1) and W(k + 1) during the stimulation cycle;

[0044] The calculation formula for the stimulation parameters is:

[0045]

[0046] Wherein, Q(0): initial Q value; b: muscle fatigue threshold.

[0047] For further improvement, the muscle fatigue threshold is 50%, that is, when Q(k) < q(0) / 2, the system determines muscle fatigue and stops electrical stimulation.

[0048] For further improvement, when RMS(k) does not exceed the maximum tolerable stimulation value RMS m the stimulation parameters increase with the increase of the electromyogram value RMS(k); when RMS(k) > RMS m the stimulation parameters are fixed at the maximum stimulation parameters; when RMS(k) < RMS r the stimulation parameters are fixed at the minimum stimulation parameters.

[0049] For further improvement, the electrical stimulation parameters in step 1) include electrical stimulation waveform, pulse width, pulse frequency, pulse amplitude, on-off ratio. The electrical stimulation waveform adopts a constant current bipolar symmetric square wave, the pulse width is 0.1 - 0.5 ms, the pulse frequency is 1 - 10 Hz, and the pulse amplitude is 5 - 30 mA; different types of muscle fibers are suitable for different on-off ratios: type I muscle fibers adopt an on-off ratio of 1:1, type IIA muscle fibers adopt an on-off ratio of 1:2, and type IIB muscle fibers adopt an on-off ratio of 1:3.

[0050] For further improvement, the specific process of surface electromyogram signal processing in step 2.2) is as follows: use a pre - differential amplifier circuit to amplify the signal, and at the same time, use a band - pass filter circuit to process the signal to filter out noise components outside the range of 10 - 500 Hz; use a post - adjustable amplifier circuit to automatically adjust the amplification factor of the signal, and use an analog - to - digital conversion module ADC to convert the electrical signal.

[0051] The beneficial effects of the present invention are as follows:

[0052] 1. It has the dual functions of both NMES (neuromuscular electrical stimulation) and biofeedback technology. It can not only activate the quadriceps femoris and induce its contraction through exogenous electrical stimulation, but also enable the patient to intuitively perceive the contraction and relaxation states of the quadriceps femoris through biofeedback technology. It can realize the adaptive adjustment of electrical stimulation parameters according to the collected electromyogram signals and automatically stop electrical stimulation when muscle fatigue is detected, thus effectively avoiding muscle injury. Whether in the early or late stage of postoperative rehabilitation, the rehabilitation needs at different stages can be met by the portable biofeedback electrical stimulation therapeutic apparatus, thus helping the patient to return to the sports field earlier.

[0053] 2. Innovatively, the volume of the biofeedback electrical stimulation device has been reduced to the wearable range, without affecting the daily life of patients, providing great convenience for the rehabilitation of patients. It enables patients to perform rehabilitation training anytime and anywhere, significantly extending the absolute time of rehabilitation.

[0054] 3. Biofeedback electrical stimulation, human motion recognition, and wireless electromyogram monitoring are integrated into a portable device, developing a new category of rehabilitation equipment. The addition of intelligent functions allows patients to adjust device parameters according to their own conditions and generate personalized rehabilitation prescriptions, making the rehabilitation process more professional and precise.

[0055] 4. It has a wide range of applications. It can not only be used for the rehabilitation of quadriceps atrophy after knee surgery, but also be popularized and applied to the prevention and treatment of disuse muscle atrophy, the relief of pain after knee joint trauma and the activation of the quadriceps, the maintenance of muscle dimension in conservative treatment patients, and standardized motion monitoring, etc.

[0056] 5. For professional athletes, this device can also be used for auxiliary applications in sports training through the professional mode. Wearing the device during training can not only perform biofeedback electrical stimulation to increase the training intensity, but also provide a muscle fatigue warning function, providing a new solution for the intelligence and scientific nature of sports training rehabilitation. Description of the Drawings

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0058] Figure 1 It is a diagram of action potential and excitatory process;

[0059] Figure 2 It is a schematic diagram of a three-axis accelerometer;

[0060] Figure 3 It is a schematic diagram of Coriolis force;

[0061] Figure 4 It is a diagram of electromagnetic vector decomposition;

[0062] Figure 5 It is a schematic diagram of the overall structure of the main unit;

[0063] Figure 6 It is a front view of the main unit;

[0064] Figure 7 It is a left and right side view of the main unit;

[0065] Figure 8 View on the host

[0066] Figure 9 Schematic diagram of the integrated electrode sheet

[0067] Figure 10 Universal electrode adapter

[0068] Figure 11 Front view of the wireless EMG module

[0069] Figure 12 Back view of the wireless EMG module

[0070] Figure 13 Schematic diagram of the host GUI

[0071] Figure 14 Schematic diagram of the design of the host main board module

[0072] Figure 15 Physical reference diagram Detailed implementation mode

[0073] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0074] The theoretical basis of the present invention is as follows:

[0075] 1. Nerve:

[0076] A neuron, also known as a nerve cell, is a highly differentiated cell that constitutes the structure of the nervous system and performs its functional activities. A neuron consists of a cell body and cell processes and has the ability to sense stimuli, integrate information, and conduct nerve impulses. Its receptive zone is located in parts of the dendrites and the cell body, where potential changes occur, manifested as graded electrogenesis. The more synaptic inputs the dendrites receive, the more significant the impact on the membrane potential of the cell body, and vice versa. The junction between the axon and the cell body is called the trigger zone, which is the key starting point for determining whether a nerve impulse is generated. The conducting zone is part of the axon, and when an action potential (AP) is generated, the conducting zone can transmit nerve impulses according to the "all or none" law. Finally, the nerve impulse is transmitted to the output zone through synapses, where neurotransmitters or electrical signals are released by the nerve endings to complete signal transmission.

[0077] 2. Muscle:

[0078] Muscle cells, also known as muscle fibers, are the basic components of muscle tissue. These cells are long and tubular and are differentiated from myoblasts. After myogenesis, muscle cells are specialized into several types, including skeletal muscle, smooth muscle, and cardiac muscle. These muscles each have their own functions. Skeletal muscle is responsible for human movement; cardiac muscle is mainly responsible for the beating of the heart and the circulation of blood throughout the body. Smooth muscle is related to human intestinal activities. Different from involuntary muscles such as cardiac muscle and smooth muscle, skeletal muscle is a voluntary muscle, and its contraction and relaxation are controlled by an individual's consciousness and are achieved through the stimulation of somatic nerves.

[0079] Skeletal muscle can be further divided into two types. One is slow-twitch muscle (Type I), which is rich in microvessels, myoglobin, and mitochondria, appears red, and can carry more oxygen to support aerobic exercise. The other is fast-twitch muscle (Type II), which can be divided into 3 types according to the contraction speed from slow to fast: 1. Type IIa: Similar to slow-twitch muscle, rich in mitochondria and microvessels, appears red. 2. Type IIx (or IId), containing fewer mitochondria and myoglobin, is the fastest muscle type in the human body. It contracts faster and is more powerful than aerobic muscles, but can only maintain for a short time and performs anaerobic exercise before the muscle becomes painful. 3. Type IIb, a "white" muscle mainly relying on anaerobic metabolism, contains even fewer mitochondria and myoglobin and mainly obtains energy through glycolysis.

[0080] Different types of exercise also vary in their use of muscles. Aerobic exercise refers to long-duration, low-intensity muscle activity that mainly relies on the oxygen system, primarily uses type I slow-twitch muscle fibers, and uses lipids, proteins, and carbohydrates as energy sources, producing a small amount of lactic acid. Anaerobic exercise, on the other hand, is short-duration, high-intensity muscle activity that mainly uses type II fast-twitch muscle fibers, is powered by ATP or glucose, consumes a small amount of oxygen, lipids, and proteins, but produces a large amount of lactic acid and cannot sustain long-term exercise.

[0081] Lactic acid in the muscle inhibits the production of ATP. Although it does not directly cause fatigue, a high concentration of lactic acid can inhibit or even stop muscle activity. However, long-term training can promote the formation of new blood vessels in the muscle, improve the ability to excrete metabolic waste, and thus enhance the contractile ability of the muscle. Once lactic acid is removed from the muscle, when its high concentration is stored in the myofibrils, it can serve as an energy source for other muscles or tissues, or be transferred to the liver for metabolism into pyruvate. In addition, intense exercise also causes an increase in the potassium ion concentration in the interstitial fluid near the muscle fibers. The acidification effect of lactic acid helps with strength recovery. Therefore, lactic acid is not only not the root cause of fatigue, but can, to a certain extent, delay the onset of fatigue.

[0082] 3. Electrophysiology

[0083] The selective permeability of the cell membrane results in a significant difference in ion concentration inside and outside the cell. The concentration of sodium ions (Na + ) and chloride ions (Cl - ) outside the cell is relatively high, while the concentration of potassium ions (K + ) inside the cell is even higher. Molecules and ions diffuse from high-concentration regions to low-concentration regions driven by the concentration gradient. In addition, ions are also affected by the electric field. Therefore, the ion distribution balance inside and outside the cell is established through the electrochemical gradient.

[0084] The resting membrane potential of skeletal muscle fibers is approximately -70 mV, and its absolute refractory period is shorter than the time when the Ca 2+ concentration increases, thus allowing action potentials (APs) to sum repeatedly and resulting in smooth and continuous muscle contraction. When the stimulation frequency is between 20 and 50 Hz, fused contraction is triggered.

[0085] The charge distribution across the cell membrane forms the resting transmembrane potential V m = V in - V out , where V in , V out represent the potentials inside and outside the membrane respectively. The average value of the resting transmembrane potential is approximately -60 mV. For a specific ion, its equilibrium potential (i.e., the transmembrane potential when the ion reaches equilibrium under the action of the electrochemical gradient) can be calculated by the following formula:

[0086]

[0087] where R is the gas constant, T is the absolute temperature, z is the valence of the ion, F is the Faraday constant, C o is the concentration outside the ion membrane, and C i is the concentration inside the ion membrane. Taking sodium ions as an example, the intracellular concentration is about 10 mM, the extracellular concentration is about 150 mM, and the corresponding equilibrium potential is about +65 mV; while for potassium ions, the intracellular concentration is about 140 mM, the extracellular concentration is about 5 mM, and its equilibrium potential is about -95 mV.

[0088] When current is injected into the cell membrane, the transient response of the membrane potential is determined by the time constant τ = R m ×C m where R m is the membrane resistance and C m is the membrane capacitance. An inward (positive) current will cause the membrane potential to move in the negative direction (hyperpolarization), while an outward (negative) current will cause the membrane potential to move in the positive direction (depolarization). When the membrane potential does not reach the activation threshold, no action potential (AP) will be generated; once it exceeds the threshold, the depolarization of the membrane will trigger the generation of an action potential.

[0089] An action potential is a rapid non-linear response of an excitable cell membrane to potential changes. When the change in membrane potential exceeds the threshold, it will trigger a rapid change in the transmembrane potential, usually causing the transmembrane potential to increase by 30 to 40 mV and then return to the resting potential. As Figure 1 shown, the time period corresponding to the peak of the action potential is called the absolute refractory period (ab segment), during which the sodium channels are closed and the cell cannot be excited even if an electrical stimulus is applied again. This period determines the shortest interval between two adjacent excitations.

[0090] As the action potential subsides, it enters the relative refractory period (bc segment). During this stage, some of the sodium channels are open, sodium ions flow in, and the excitability of the cell gradually recovers. Although the cell can be excited at this time, the required stimulus intensity must be higher than the original threshold intensity. After the relative refractory period, the cell enters the supernormal period (cd segment) and the subnormal period (de segment). In the supernormal period, a lower stimulus intensity is required to trigger a new action potential, while in the subnormal period, a higher stimulus intensity is required.

[0091] An action potential is the basic unit for most axons to transmit information. It propagates throughout the cell membrane without attenuation, and this process is called the conduction of the action potential. If the action potential occurs on a nerve fiber, it is called a nerve impulse. The goal of functional electrical stimulation is to target excitable cells through external stimulation and artificially induce or inhibit the generation of action potentials. This technology has a wide range of applications in the biomedical field, such as for the treatment of neuromuscular diseases or the restoration of muscle function.

[0092] 4. Principle of Accelerometer:

[0093] The main function of an accelerometer is to measure the acceleration of an object along three axes (x, y, and z axes). In a stationary state, the accelerometer can accurately measure the constant downward gravity because force is equal to the product of mass and acceleration. Triaxial accelerometers are usually applied to measure the physical orientation of a device relative to gravity, especially to detect attitude changes such as the tilt angle, pitch angle, and roll angle of the device.

[0094] As Figure 2 shown in (a) below, a typical triaxial accelerometer can measure the forces acting along the x, y, and z axes respectively. In a stationary state, gravity is described as a downward vector acting on the z-axis of the accelerometer, with an amplitude of g (acceleration due to gravity). When the z-axis of the accelerometer is perfectly aligned with the direction of gravity, the x-axis and y-axis are perfectly perpendicular to the direction of gravity. Therefore, in this state, the acceleration of the z-axis is equal to g, while the accelerations of the x-axis and y-axis are both zero.

[0095] Figure 2 Shown in (b) below is a biaxial view of the accelerometer. When the accelerometer rotates along the xz plane, the contribution of gravity to the z-axis decreases, while the contribution to the x-axis increases. As Figure 2 shown in (c) below, when the accelerometer rotates 60° relative to the horizon, the acceleration of the z-axis is g×sin60°, and the acceleration of the x-axis is g×sin60°. This relationship between the measurement direction of the accelerometer and the direction of gravity can be used to calculate the corresponding tilt angle. By integrating the triaxial accelerations (acceleration values of the x, y, and z axes), the tilt angle can be calculated using the following formula:

[0096]

[0097] Essentially, a triaxial accelerometer is an inertial sensor mainly used to measure the specific force of an object. The concept of specific force can be defined as the difference between the absolute acceleration of the carrier relative to inertial space and the gravitational acceleration. According to the above formula, in the absence of external forces, the accelerometer can calculate the pitch angle and roll angle of an object based on the measured acceleration values, and there will be no cumulative error in the calculation process.

[0098] 5. Principle of Gyroscope:

[0099] A gyroscope is used to measure the angular velocity of an object around its own three axes (i.e., the body coordinate system), which is based on the angular velocity generated by the Coriolis force effect. As Figure 3 shown below.

[0100] When an object moves along a straight line, if a rotation is applied to the coordinate system in which it is located, the object will simultaneously exhibit two velocity components: the radial velocity v1 and the tangential velocity v2. Using the Coriolis force formula, the angular velocity can be calculated. The Coriolis force formula is as follows:

[0101] F = -2mω×v

[0102] where m is the mass of the object, ω is the angular velocity of the rotating coordinate system, and v is the velocity of the object relative to the rotating reference frame, i.e., v = v1 - v2. Combining the acceleration a measured by the accelerometer and the above formula, the angular velocity ω of the gyroscope can be obtained:

[0103]

[0104] 6. Principle of magnetometer:

[0105] In an Inertial Measurement Unit (IMU), the magnetometer is responsible for measuring the magnetic fields received by the device on the x, y, and z axes, and processing this data through a microcontroller to calculate the heading angle related to the magnetic north pole, thereby realizing the detection of geographical orientation. The magnetometer consists of three mutually perpendicular magnetoresistive sensors. Each sensor in a specific direction detects the intensity of the geomagnetic field in that axis, thus accurately identifying the direction and intensity of the external magnetic field of the object.

[0106] The magnetometer can not only compensate for the tilt errors in the accelerometer and gyroscope data, but also perform yaw correction on the three-axis acceleration data and three-axis angular velocity data. As Figure 4 shown, the projections of the geomagnetic intensity T have the following characteristics: The projection on the x-axis is T X , and the corresponding direction is north; the projection on the y-axis is T Y , and the corresponding direction is east; the projection on the z-axis is T z , and the corresponding direction is downwards; the projection on the xy plane is T XY .

[0107] In addition, the magnetometer can detect two types of angles:

[0108] Magnetic declination (α): It is defined as the angle between T XY and T X in the horizontal component T Y , and is expressed as:

[0109]

[0110] Magnetic dip (β): It is defined as the angle between the geomagnetic intensity T and the horizontal plane, and is calculated from the vertical component T z and the horizontal component T XY , and is expressed as:

[0111]

[0112] Meanwhile, the total value of the geomagnetic intensity T can be expressed as:

[0113]

[0114] Since the signal of the magnetometer cannot be directly read, it needs to be converted into an electrical signal for reading. When the magnetometer detects a magnetic field shift, the residual magnetic field inside the sensor can be cleared by setting and resetting pulses, restoring it to the normal state, thereby effectively eliminating the drift phenomenon caused by the magnetic field shift.

[0115] The implementation of the biofeedback algorithm is as follows:

[0116] 1. Electrical stimulation parameters:

[0117] Electrical stimulation parameters generally refer to electrical stimulation waveforms, frequencies, pulse widths, intensities, duty cycles, etc. Different electrical stimulation parameters determine the effect of electrical stimulation.

[0118] (1) Electrical stimulation waveform: Usually, the stimulation waveforms commonly used in functional electrical stimulation include bidirectional sine waves, bidirectional rectangular waves, and bidirectional triangular waves, etc. When evaluating which stimulation waveform is the most ideal in clinical medicine and research, the following three key factors need to be considered comprehensively: the average stimulation current required to achieve the desired muscle contraction tension, the subjective comfort during the stimulation process, and the physiological response of the body to electrical stimulation, which is an important objective criterion for measuring comfort and potential tissue damage.

[0119] Existing research shows that the comfort level of sine waves is relatively high, followed by rectangular waves. Due to the sudden change characteristics of its current, the rectangular wave can achieve more efficient muscle contraction. In clinical applications, rectangular waves are usually preferred for neuromuscular electrical stimulation, and by adjusting parameters such as frequency and pulse width, targeted selective stimulation of different muscles or symptoms can be achieved. Triangular waves and other waveforms are mostly used for electrical stimulation treatment of muscles without nerve innervation.

[0120] (2) Electrical stimulation frequency: The electrical stimulation frequency refers to the number of stimulation pulses per unit time. In neuromuscular electrical stimulation, different frequencies will trigger different muscle contraction responses. For human muscles, electrical stimulation at 1 - 10 Hz will cause single contractions; electrical stimulation at 20 - 30 Hz will make the muscle enter an incomplete tetanic contraction state; while a higher frequency of 50 Hz will cause a complete tetanic contraction. For sensory nerves, a frequency of around 50 Hz will produce an obvious tremor sensation, while a frequency of around 100 Hz has an analgesic effect.

[0121] (3) Pulse width of electrical stimulation: The pulse width of electrical stimulation refers to the duration of a single pulse. Neurons can respond to a current stimulation with a pulse width of 30 μs, while muscles require a longer pulse width to be effectively activated. In clinical applications, a stimulation current with a pulse width of 300 μs is more comfortable and less likely to cause pain compared to 50 μs or 1000 μs, thus having more advantages in actual operations.

[0122] (4) Amplitude of electrical stimulation: The amplitude of electrical stimulation refers to the intensity of electrical stimulation. The higher the intensity, the greater the contraction intensity of the muscle. In clinical applications, it is necessary to reasonably set the amplitude of electrical stimulation by comprehensively considering the size of the electrode patch and the subjective feelings of the patient. Usually, the electrical stimulation intensity is generally set to 20%-30% of the maximum tetanic contraction to improve the comfort of the patient while ensuring the effect.

[0123] (5) Duty cycle: The duty cycle of electrical stimulation refers to the ratio of the duration of a series of pulses to the rest time. In clinical applications, due to the influence of electrophysiological characteristics, the longer the rest time, the less likely the muscle is to fatigue, but the total time of electrical stimulation required will also increase accordingly. Research shows that starting from the characteristics of muscle fibers, different types of muscle fibers are suitable for different duty cycles: type I muscle fibers are suitable for a duty cycle of 1:1, type IIA muscle fibers are suitable for a duty cycle of 1:2, and type IIB muscle fibers are suitable for a duty cycle of 1:3.

[0124] (6) Waveform: A symmetric biphasic wave is more comfortable and safer than a unidirectional wave. And the current can quickly rise to the amplitude, which can avoid the adaptation phenomenon of neuromuscular and quickly reach the threshold intensity to cause action potential.

[0125] The biphasic symmetric pulse waveform can effectively reduce the accumulation of charge at the stimulation site compared to the monophasic waveform, thereby reducing the possible damage to human tissues. At the same time, using a constant current output can keep the current magnitude constant when the human impedance changes, ensuring the stability of the stimulation effect. Therefore, this design selects a constant current biphasic square wave as the stimulation waveform, and its pulse parameters are shown in the following table:

[0126]

[0127] 2. Surface electromyogram signal acquisition:

[0128] 2.1 Characteristics of surface electromyogram signals:

[0129] Surface electromyogram (sEMG) signal is a low-frequency signal with bipolar characteristics. Its signal spectrum is mainly distributed between 20 Hz and 500 Hz, and the frequency range of 50 Hz to 200 Hz is the main concentration area of spectral energy. The sEMG signal is very weak, and its voltage amplitude is usually between 5 μV and 1 mV. In addition, there are significant individual differences in sEMG signals, which are significantly affected by factors such as gender, age, and muscle size. Even for the same person, changes in muscle position and fatigue level can also have a great impact on the measurement results.

[0130] 2.2 Bioelectrodes:

[0131] Muscle Fiber Action Potential (MFAP) is the result of the propagation of action potential along muscle fibers and is the most basic detection unit of sEMG signal. During a movement, a large number of muscle fibers are divided into multiple groups, each group is called a motor unit, and the human muscle is on average composed of 20 to 50 motor units. Therefore, the electrophysiological signal acquisition device cannot detect the motor units of individual muscle fibers but can only detect the sum of Motor Unit Action Potentials (MUAP). Multiple MUAPs are generated during the repeated relaxation and contraction of motor units, and the collection of these MUAPs forms a motor unit action potential sequence, each of which is unique. As the muscle contracts, the number of motor units increases, and the superposition of the signals of each motor unit finally forms the required sEMG signal.

[0132] How to efficiently and accurately collect sEMG signals is crucial, and bioelectrodes are the key link in this process. Bioelectrodes need to meet the following basic requirements:

[0133] (1) Low interface impedance: There should be a low interface impedance between the electrode and the skin to ensure that the collected electrophysiological signal has a high signal-to-noise ratio, thereby improving the signal quality;

[0134] (2) Stable contact: The electrode needs to be in close contact with the skin to ensure stable contact and reduce the interference of motion artifacts on signal acquisition;

[0135] (3) Good biocompatibility: The electrode should be harmless to the human body and not cause adverse reactions to ensure safety and comfort.

[0136] According to the acquisition and measurement principles, bioelectric electrodes can be classified into wet electrodes, dry electrodes, semi-dry electrodes, hydrogel electrodes, etc. Wet electrodes are the most commonly used electrode type in current clinical applications, among which silver / silver chloride electrodes (Ag / AgCl) coated with conductive gel are the most common. This type of electrode is a disposable product that can collect signals with relatively high resolution, but its inherent defects limit further applications. First, the skin needs to be pre-treated before using wet electrodes, such as cleaning the skin to reduce the high impedance caused by oil. Second, the conductive gel may irritate the skin, causing allergic reactions and having poor air permeability. In addition, during use, the conductive gel will gradually dry out, resulting in an increase in the interface impedance between the electrode and the skin, thus reducing the signal quality. Therefore, wet electrodes are more suitable for short-term signal monitoring due to their low cost and relatively small interface impedance. Dry electrodes do not require the use of conductive gel, so there is no obvious performance degradation during long-term monitoring, providing the possibility for continuous recording of human physiological signals. However, the interface impedance between dry electrodes and the skin is relatively large, resulting in less than ideal signal quality and low comfort. Semi-dry electrodes reduce the impedance by continuously releasing electrolyte solution to the skin-electrode interface during use, but they still cannot reach the standard of wet electrodes. To achieve long-term, stable and non-invasive acquisition, this project adopts conductive hydrogel electrodes with high biocompatibility, high conductivity and good adhesion.

[0137] Collecting electromyography signals usually requires multiple electrodes to work together, including reference electrodes, working electrodes, etc. For a single-channel electromyography signal acquisition system, only three electrodes are needed to form an array: an active electrode, a ground electrode, and a reference electrode. Among them, the active electrode needs to be placed on the skin surface of the target to be monitored for signal acquisition; the reference electrode needs to be placed at a location where the bioelectric potential fluctuates relatively gently to form a potential difference with the active electrode to achieve effective signal acquisition; the ground electrode is mainly used to shield external noise interference to ensure the accuracy of signal acquisition.

[0138] For a multi-channel electromyography signal acquisition system, more electrodes are required, and the design of the electrode array used needs to be optimized according to the distribution area and size of the target muscle to accurately collect the surface electromyography signals of the target muscle. Such systems usually adopt an integrated electrode array, and its design incorporates flexible, ultra-thin and highly adaptable electronic technologies. Factors such as the shape, size, structure of the electrode array and the distance between electrodes will have a significant impact on the signal acquisition effect.

[0139] The shape of the electrode is generally designed as square, rectangular or circular, and these shapes can better cover the target muscle area. The diameter of the electrode is usually selected within the range of 1 mm to 5 mm. If the diameter exceeds 5 mm, it will change the characteristics of the surface motor unit action potential (MUAP), interfere with the acquisition of surface electromyogram signals, and thus have a negative impact on the measurement results. The distance between electrodes refers to the center distance between the conductive centers of two adjacent electrodes. Since the sizes of target muscles vary greatly and the electrode spacing affects the signal crosstalk effect, the electrode spacing is not a fixed design. Usually, 5 mm or 10 mm is selected as a relatively common spacing configuration.

[0140] The electrode configuration of surface electromyogram signals is usually monopolar configuration or bipolar configuration. The bipolar configuration uses two detection electrodes and one ground electrode. The signals collected by the detection electrodes will be transmitted to a differential amplifier, and the signal is obtained by recording the potential difference between the two detection electrodes. At the same time, the bipolar configuration effectively suppresses environmental noise through the amplifier, thereby providing a higher signal-to-noise ratio (SNR) and improving the signal quality.

[0141] The monopolar configuration consists of one detection electrode and one ground electrode, and the ground electrode also serves as another detection reference electrode at the same time. Compared with the bipolar configuration, the monopolar configuration has a lower signal frequency response and a relatively weaker signal-to-noise ratio.

[0142] The multi-channel electrode array can be divided into four types according to the configuration characteristics: linear array, two-dimensional electrode array, high spatial resolution array and high density array. The linear array is composed of multiple electrodes for signal measurement connected in a straight line arrangement, and is used to record electromyogram signals in a specific direction. The two-dimensional electrode array is composed of a combination of multiple surface electromyogram (sEMG) electrodes arranged to cover the skin surface of one or more muscles to improve the coverage range of signal acquisition. Both the high spatial resolution array and the high density array are further derived and developed based on the two-dimensional electrode array, and have higher spatial resolution and dense electrode layouts, which are suitable for more refined signal acquisition requirements.

[0143] 2.3 Processing of surface electromyogram signals:

[0144] The surface electromyogram signal (sEMG) is a weak electrical signal. In order to suppress the common-mode noise interference in the signal, a preamplifier differential circuit is used and the signal is amplified by 20 times. At the same time, a band-pass filter circuit is used to process the signal to effectively filter out the noise components outside the range of 10 - 500 Hz.

[0145] In the subsequent circuit design, a tunable amplifier circuit is added, which can automatically adjust the amplification factor of the signal according to the specific situation of the myoelectric signal, thereby improving the acquisition effect and adaptability. In terms of signal digitization, a 16-bit bipolar analog-to-digital conversion module (ADC) is used to perform high-precision conversion on the electrical signal, and the processed data is transmitted to the main control module. The sampling frequency is set to 2000Hz to ensure that sufficient signal details are captured and high-quality signal acquisition and processing are achieved.

[0146] 2.4 Biofeedback Electrical Stimulation Algorithm:

[0147] The biofeedback electrical stimulation algorithm of this product is mainly applied to optimize the passive contraction effect of the quadriceps femoris by adjusting the stimulation parameters in real time during neuromuscular electrical stimulation (NMES), and stop the electrical stimulation in time when muscle fatigue is detected.

[0148] Before performing NMES, it is necessary to pre-collect the surface electromyogram (sEMG) of the patient's affected limb in the relaxed state and the maximum voluntary contraction (MVC) state, and record the root mean square value RMS of the myoelectricity. r , RMS m ; At the same time, determine the following stimulation parameters:

[0149] Minimum stimulation parameter: The minimum stimulation amplitude A min and the minimum stimulation pulse width W min .

[0150] Maximum stimulation parameter: The maximum stimulation amplitude A max and the maximum stimulation pulse width W max .

[0151] To avoid muscle fatigue caused by long-term stimulation, the algorithm combines non-linear characteristic parameters (sample entropy) and time-domain characteristic parameters (integrated electromyogram value, iEMG) to calculate the fatigue characteristic Q value for real-time detection of muscle fatigue state:

[0152]

[0153] where, u s is the sample entropy value of the myoelectric signal at the stimulation site, and u i is the integrated value (iEMG) of the myoelectric signal at the stimulation site.

[0154] Sample entropy is used to describe the complexity of a time series. A smaller value indicates a lower complexity of the time series. When the muscle is fatigued, the rate of muscle change slows down, and the activities of motor units tend to be consistent, resulting in a decrease in sample entropy. At the same time, as the degree of fatigue deepens, the muscle gradually tenses, and more motor units will be recruited in the fatigued area, increasing the discharge frequency and synchronization degree, thus increasing the integrated electromyogram value (iEMG). Therefore, the characteristic Q value in the fatigued state will be significantly smaller than that in the initial state and continue to decrease as the degree of fatigue deepens. When it is detected that the Q value is lower than a specific threshold, it can be determined that the muscle has entered the fatigued state, and the electrical stimulation output is stopped.

[0155] To avoid the interference of the stimulating electrode on the sEMG signal acquisition, the system is designed to operate alternately between the acquisition period and the stimulation period. During the acquisition period, the sEMG signal is acquired and its characteristic values RMS(k), the pulse amplitude A(k + 1), pulse width W(k + 1) of the next period, and the fatigue characteristic Q(k) are calculated. When Q(k) is greater than the fatigue threshold, the electrical stimulation module outputs electrical stimulation according to the values of A(k + 1) and W(k + 1) during the stimulation period.

[0156] The calculation formulas for the stimulation parameters are as follows:

[0157]

[0158]

[0159] Q(0): Initial Q value; b: Muscle fatigue threshold

[0160] According to relevant research, when the Q value drops to 50% of the initial value, muscle fatigue begins to occur. In this design, the tentative fatigue threshold b is 50%, that is, when Q(k) < Q(0) / 2, the system determines that the muscle is fatigued and stops the electrical stimulation.

[0161] In addition, the muscle will gradually develop tolerance when receiving external electrical stimulation. Long-term use of the same stimulation intensity may cause the stimulation effect to decrease over time. Therefore, it is stipulated in the design that: when RMS(k) does not exceed the maximum tolerated stimulation value RMS m , the stimulation parameters increase with the increase of the electromyogram value RMS(k); when RMS(k) > RMS m , the stimulation parameters are fixed at the maximum stimulation parameters; when RMS(k) < RMS r , the stimulation parameters are fixed at the minimum stimulation parameters.

[0162] 2.5 Wireless transmission protocol:

[0163] A stable wireless connection transmission protocol is an important guarantee for the connection reliability between devices. To balance the data transmission rate, communication distance, power consumption, and design cost, several wireless communication protocols are selected as follows:

[0164]

[0165] The transmission rates of LoRa and Zigbee are relatively low, and there may be a certain degree of spectrum interference in the practical application of LoRa technology. Although Bluetooth technology limits the maximum number of devices that a master device can connect to to 7, there are not many wireless nodes in this project, which can not only meet the requirements, but also has the advantages of low power consumption, low radiation and low cost.

[0166] In contrast, although WiFi has a higher transmission rate, it has higher energy consumption, higher cost, and requires additional devices such as a separate network card. Considering the advantages and disadvantages of each technology above, Bluetooth is finally selected as the wireless communication method.

[0167] 2.6 Human motion recognition based on inertial sensing signals:

[0168] Due to the widespread problem of activation failure after quadriceps femoris surgery, and patients often have difficulty judging whether their movements are standard based on their own experience when starting rehabilitation exercises. During the rehabilitation process, if there is a compensatory phenomenon in other muscles, it will not only significantly reduce the exercise efficiency of the quadriceps femoris, but may also lead to secondary injuries. Therefore, it is crucial to use inertial sensors to sense the limb posture signals, present the real-time feedback to the patients in the form of sound or image, and combine with electromyography signals to provide accurate movement guidance for the patients. In addition, by connecting a wireless monitoring module, a joint system of the entire lower limb can be constructed to achieve more accurate data collection and evaluation, thereby optimizing the rehabilitation effect.

[0169] Compared with the traditional optical-based motion capture system, although inertial sensors have problems such as data offset, low accuracy, and need to be calibrated before use, the optical motion capture system is costly, requiring not only the installation of high-speed cameras with ultra-high frame rates in a professional venue, but also the testers to wear special marker suits. Such an operation process is unacceptable for postoperative patients and medical sites. Therefore, it is a more reasonable choice to adopt an inertial sensor system with lower cost and more convenient use.

[0170] The hardware structure is as Figure 5 shown:

[0171] 1. Host design:

[0172] The host is designed as a high-performance mobile terminal with a screen, a speaker and physical buttons, and its main external components are as Figures 6 - 8 shown.

[0173] Housing 1: Adopt aluminum alloy CNC process, with a variety of color options. Among them, white uses electrophoresis process, and silver and dark gray use 180-mesh sandblasting anodizing process.

[0174] Power-on button 2: A toggle switch with a lock is adopted, and physical locking can be achieved in the on / off state to prevent accidental touch.

[0175] Volume adjustment button 3: An integrated metal push button is adopted.

[0176] Power indicator 4: An multi-color LED is adopted. When the power is turned on, it shows white and stays on constantly; when the power is less than 30%, it shows orange and stays on constantly; when the power is less than 20%, it shows red and stays on constantly; when the power is less than 5%, it shows red and blinks; when charging, it shows white and breathes; when using with external power supply, it shows green and stays on constantly.

[0177] The screen 5 selects a 3.2-inch AMOLED screen with a resolution of 360*640 and 16.7M display colors.

[0178] Speaker 6: A high-fidelity micro speaker is adopted and equipped with an X-axis linear motor.

[0179] Electrical stimulation intensity adjustment knob 7: An infinitely variable knob is adopted, and the electrical stimulation intensity can be adjusted linearly and infinitely.

[0180] Mode selection knob 8: A knob with gears is adopted, and the output mode of the host can be adjusted.

[0181] Start button 9: A push button switch is adopted. Pressing it once starts the host to output electrical stimulation.

[0182] Pause / Resume switch 10: A push button switch is adopted. Pressing it once pauses the output of the host, and pressing it again resumes the output of the host.

[0183] End button 11: A push button switch is adopted. Pressing it once ends the output of electrical stimulation by the host.

[0184] USB Type-C 12: As a data transmission interface for connecting electrodes, it cooperates with (13) to fix the integrated electrode or the universal electrode adapter module to the host.

[0185] Quick-release bayonet 13: Interlocks with the quick-release button of the integrated electrode or the universal electrode adapter module to fix the integrated electrode or the universal electrode adapter module.

[0186] Multi-functional strap port 14: When connecting the integrated electrode, this fixing port is connected to the fixing bayonet on the back of the integrated electrode patch to fix the integrated electrode. When connecting the universal electrode adapter module, this fixing port can fix the strap.

[0187] USB Type-C 15: When connecting the power supply, it charges the built-in lithium battery of the fuselage. After the battery is fully charged, it automatically switches to the external power supply mode and directly powers the host through this interface. When connecting the parameter adjustment software, this interface can transmit the host data to the computer.

[0188] 2. Integrated electrode patch design, such as Figure 9 shown below:

[0189] The integrated electrode patch integrates the electrical stimulation output electrode and the surface electromyogram acquisition electrode into one, adheres to the clean skin through a conductive hydrogel, and transmits the collected surface electromyogram data to the host through a USB Type-C interface.

[0190] 3. Universal electrode adapter design, such as Figure 10 shown below:

[0191] For professionals who need to collect specific electromyogram signals and perform electrical stimulation on specific muscles, or patients who need to use more inexpensive universal electrodes, a universal electrode adapter can be used to connect to the host. The adapter is equipped with 8 3.5mm electrode interfaces, which can fully customize the surface electromyogram acquisition / electrical stimulation output channels, and can record up to 8-channel surface electromyogram simultaneously or output 8 groups of electrical stimulation.

[0192] 4. Wireless electromyogram module design, such as Figure 11 and 12 shown below:

[0193] The wireless electromyogram module consists of a rechargeable lithium battery, a main control module, an attitude sensing module, a wireless connection module and four electrodes. Among them, 3 electrodes are surface electromyogram acquisition electrodes, and 1 electrode is an electrical stimulation output electrode. When the wireless electromyogram module works in cooperation with the host, the main control module wirelessly transmits the collected electromyogram data to the host, and the host calculates and wirelessly transmits the stimulation parameters to the stimulation electrode to complete wireless biofeedback electrical stimulation. When the wireless electromyogram module is used as a joint node, it can be networked with the host and up to 5 other wireless electromyogram modules, and a limb can be constructed to complete human motion recognition, and the limb movement state can be displayed on the host and guidance can be provided.

[0194] 5. GUI design, such as Figure 13 shown below:

[0195] The GUI design follows the principle of simplicity and usability, adopts a black background, highlights the content and saves power in combination with the self-luminous characteristics of the OLED. The GUI is mainly divided into three areas. The left area mainly displays functions such as the current time, the host battery power, the host mode, the running time, and the connected nodes; the middle area visually displays the attitude sensor data; the right area visually displays the electromyogram signal.

[0196] 6. Host internal design, such as Figure 14 shown below:

[0197] The main components inside the host are a rechargeable lithium battery and a main board. The main board includes a main control module, an electrical stimulation output module, a surface electromyogram signal preprocessing module, a wireless interconnection module, an attitude perception module, a power management module, a display output module, a speaker module, and a vibration motor module.

[0198] Clinical trial verification:

[0199] 1. Research subjects:

[0200] 1.1 Source of research subjects:

[0201] Patients who underwent knee surgery at Jiangsu Provincial Hospital of Traditional Chinese Medicine from March 2025 to March 2026 were selected as research subjects. It is planned to enroll 100 people, including 80 in the experimental group and 20 in the control group (see the specific grouping below). In addition, another 10 healthy controls were recruited to estimate the effectiveness of the intervention measures on the recovery of quadriceps femoris function and health level.

[0202] 1.2 Inclusion / exclusion criteria:

[0203] 1.2.1 Inclusion criteria:

[0204] (1) Aged 18 - 60 years old;

[0205] (2) Unilateral arthroscopic anterior cruciate ligament reconstruction / Unilateral arthroscopic posterior cruciate ligament reconstruction / Unilateral arthroscopic medial patellofemoral ligament reconstruction / Unilateral arthroscopic meniscus suture / Unilateral medial collateral ligament repair of the knee / Unilateral tibial plateau fracture internal fixation / Unilateral unicompartmental knee surface replacement / Unilateral total knee replacement;

[0206] (3) No tibial shaft fracture or femoral fracture;

[0207] (4) No peripheral nerve injury;

[0208] (5) No postoperative infection;

[0209] (6) No cognitive dysfunction;

[0210] (7) Cooperate with the experimental process of this study and sign the informed consent form.

[0211] 1.2.2 Exclusion criteria:

[0212] (1) Multiple structural injuries of the knee joint and underwent multiple surgeries such as ACL reconstruction combined with PCL reconstruction, etc.;

[0213] (2) Suffering from severe lumbar disc herniation;

[0214] (3) Severe cardiovascular and cerebrovascular diseases, etc.;

[0215] (4) Allergic to gel dressing;

[0216] (5) Those who cannot tolerate rehabilitation training;

[0217] (6) Pregnant women.

[0218] 1.2.3 Criteria and procedures for stopping the trial / treatment:

[0219] Situations for stopping the trial / treatment:

[0220] (1) Stopping due to the subject withdrawing informed consent and being unable to follow up as required.

[0221] (2) Stopping decided by the researcher and the medical institution due to serious safety issues found during the trial.

[0222] (3) Stopping required by the national drug regulatory department.

[0223] (4) Stopping required by the sponsor.

[0224] Procedures for stopping the trial / treatment:

[0225] (1) All withdrawn subjects should retain all source data and source documents. After the subject withdraws, the researcher should contact the subject at least twice on different days in various forms such as by phone, email, or registered mail, and try to record the reasons for the subject's withdrawal.

[0226] (2) For subjects who have received the trial product / trial treatment, the sponsor should fulfill its responsibilities to the subjects during the trial in accordance with the content of GCP and the informed consent form.

[0227] 1.3 Research ethics statement

[0228] This study strictly adheres to the Declaration of Helsinki and relevant Chinese regulations on clinical trial research. The clinical research project was carried out after obtaining the approval of the Ethics Committee of Jiangsu Provincial Hospital of Traditional Chinese Medicine, and all research subjects signed the informed consent form.

[0229] 2 Research procedures:

[0230] 2.1 Collection of basic patient information:

[0231] Collect various information of patients meeting the inclusion and exclusion criteria, including gender, age, height, weight, occupation, surgery time, graft situation, Lyshom score, Tegner score, IKDC score, VAS score, RPE score, Beighton score, KTSSD, etc., and establish a patient database.

[0232] 2.2 Equipment preparation:

[0233] In this study, the above wearable biofeedback electrical stimulation device was used. Its acquisition module can achieve multi-channel electromyogram detection and output data, with a signal sampling frequency of 2000 Hz. The attitude perception module can output the relative spatial position of the affected limb, and the main control module can output data such as electrical stimulation signal data, pulse amplitude, pulse width, and Q value.

[0234] 2.3 Clinical trial verification of the rehabilitation effect of the portable biofeedback combined with electrical stimulation therapeutic apparatus

[0235] 2.3.1 Subject preparation:

[0236] For electromyogram testing, the bilateral vastus medialis, rectus femoris, and vastus intermedius of the subjects were selected. Before the test started, the subjects' skin was treated by removing hair and wiping with 75% medical alcohol. The acquisition electrodes were attached to the muscle bellies: for the vastus medialis, it was placed along the muscle fibers at 2 finger widths proximal to the upper medial edge of the patella; for the rectus femoris, it was at the midpoint of the line connecting the upper edge of the patella and the anterior superior iliac spine; for the vastus lateralis, it was placed along the muscle fibers at 4 finger widths proximal to the upper lateral edge of the patella.

[0237] 2.3.2 Baseline data detection:

[0238] Before the operation, the patients needed to complete baseline data detection, including relaxed state sEMG, sitting knee extension test, maximum voluntary contraction (MVC) test, and peak torque (PT) of the quadriceps femoris. Healthy participants only participated in data collection once to evaluate the activation and strength of the quadriceps femoris.

[0239] (1) Sitting knee extension test: The subject was made to sit with the popliteal fossa closely against the chair to fix the thigh. The knee joint was kept at 90°, and the subject was instructed to lift the lower leg at an appropriate speed until full knee extension. The entire knee extension movement was controlled within 2 s and then slowly lowered. The whole set of movements should be smooth and without pauses. After the patient was familiar with the movement essentials, a 5-minute rest was taken before the test. All patients performed 5 times, with a 10-s interval between each time, and the sEMG data of each muscle were recorded respectively.

[0240] (2) MVC: The patient sat with the knee flexed at 60°, and the lower leg was fixed with a strap during the test. The patient performed 3 full-force knee extensions, each maintaining a maximum voluntary isometric contraction for 5 s, with a 5-s interval between each time, to determine the MVC electromyogram data.

[0241] (3) PT: Calibrate the isokinetic dynamometer system. The patient sits on the test chair, crosses their arms in front of their chest, and the upper body and thighs are fixed with straps. Adjust the sitting position angle to approximately 110°, and align the knee joint axis with the axis of the power arm. The resistance pad at the end of the power arm is fixed at 3 cm above the upper edge of the medial malleolus of the ankle joint, and the joint range of motion is set to 0 - 90°. The concentric contraction protocol for the quadriceps femoris muscle is set as follows: flexion at 90(°) / s × 5, extension at 90(°) / s × 5; flexion at 360(°) / s × 5, extension at 360(°) / s × 5. The interval between each group is 2.5 minutes. Conduct 3 sets of practice before the test, and then conduct 5 formal tests.

[0242] Relative peak torque = PT / BM

[0243] Gravity - corrected knee extension torque = (extension torque of the affected limb + torque generated by the weight of the affected limb) / (extension torque of the normal limb + torque generated by the weight of the limb) × 100

[0244] Gravity - corrected knee flexion torque = (flexion torque of the affected limb - torque generated by the weight of the affected limb) / (flexion torque of the normal limb - torque generated by the weight of the limb) × 100

[0245] 2.3.3 Patient grouping and rehabilitation program:

[0246] Use SPSS software to randomly divide 100 patients into 5 groups, with 20 patients in each group. The experimental grouping and the rehabilitation programs for each group are shown in the table:

[0247]

[0248]

[0249] 2.3.4 Observation indicators and detections:

[0250] Detect the sEMG of the quadriceps femoris muscle, MVC electromyography data, and PT of the patients at 4 weeks, 8 weeks, 12 weeks, and 6 months after surgery respectively.

[0251] 2.3.5 sEMG signal processing:

[0252] After the exported sEMG data is processed using MATLAB software, calculate the integrated electromyography value (iEMG) of each muscle. Calculate the average value of the iEMG values of repeated movements for analysis.

[0253] 2.3.6 Subjective fatigue and pain scores:

[0254] After the subject completes a set of test movements, evaluate using the PRE scale and VAS score

[0255] The PRE score starts from 6 (7, very easy; 9, relatively easy; 11, easy; 13, a bit nervous; 15, nervous; 17, very nervous; 19, extremely nervous; 20, exhausted).

[0256] VAS score: 0, no pain; 3, mild pain; 4 - 6, moderate pain, disturbing sleep but tolerable; 7 - 10, the pain gradually worsens, unbearable or even intolerable.

[0257] 2.3.7 Evaluation of thigh muscle size:

[0258] Use computed tomography (CT) to measure the cross - sectional area of thigh muscle tissue, and use statistical methods to evaluate the data.

[0259] 2.4 Statistical methods:

[0260] All data are processed using SPSS statistical software 28.0. Measurement data are expressed as "ˉx±s"; paired t - test is used for within - group comparison, Newman - keuls test for pairwise comparison of means among multiple samples for between - group comparison, and two - way ANONA test for correlation comparison between indicators. P < 0.05 is considered statistically significant.

[0261] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiments, the above - mentioned are only the preferred embodiments of the present invention. Since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the method embodiments. The above - mentioned are only the specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. For any person skilled in the art in the technical field disclosed by the present invention, any changes or substitutions that can be easily thought of by those of ordinary skill in the technical field, without departing from the principle of the present invention, should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A biofeedback electrical stimulation device for quadriceps rehabilitation training, characterized in that: It includes a host, an integrated electrode patch, a universal electrode adapter, and a wireless electromyogram module; the host is used in combination with the integrated electrode patch or connected to the universal electrode through the universal electrode adapter and then connected to the universal electrode for use; the integrated electrode patch includes a connection terminal, an electrode patch substrate, an electrical stimulation output electrode, and a surface electromyogram acquisition electrode. The electrical stimulation output electrode and the surface electromyogram acquisition electrode are provided with a skin contact surface, and the skin contact surface is adhered to the clean skin where the quadriceps femoris is located through a conductive hydrogel; the surface electromyogram signal collected by the surface electromyogram acquisition electrode is transmitted to the host through the connection terminal, and the electrical stimulation signal output by the host after calculation is transmitted to the electrical stimulation output electrode through the connection terminal; the wireless electromyogram module includes an electromyogram power module, an electromyogram main control module, an attitude sensing module, a wireless connection module, and an electrode assembly. The electrode assembly includes at least three surface electromyogram acquisition electrodes and one electrical stimulation output electrode. The electromyogram main control module wirelessly transmits the collected electromyogram data to the host, and the host wirelessly transmits the stimulation parameters to the electrical stimulation output electrode after operation to complete wireless biofeedback electrical stimulation.

2. The biofeedback electrical stimulation device for quadriceps rehabilitation training according to claim 1, wherein: The host includes a main control module, a host power module, a wireless interconnection module, an attitude sensing module, a surface electromyogram signal preprocessing module, an electrical stimulation output module, and a signal output module connected to the main control module. Among them, the wireless interconnection module is connected to the wireless electromyogram module through a wireless signal, and the surface electromyogram signal preprocessing module and the electrical stimulation output module are connected to the integrated electrode patch or the universal electrode adapter.

3. The biofeedback electrical stimulation device for quadriceps rehabilitation training according to claim 1 or 2, characterized in that: The universal electrode adapter is configured with a number of ordinary electrode interfaces to customize the surface electromyogram acquisition and electrical stimulation output channels.

4. The biofeedback electrical stimulation device for quadriceps rehabilitation training according to claim 2, wherein: The signal output module includes a display output module and a speaker module, and the speaker module is provided with a vibration motor.

5. A biofeedback electrical stimulation method for quadriceps rehabilitation training, characterized in that It includes the following steps: 1) Set the electrical stimulation parameters; 2) Acquisition cycle, surface electromyogram signal acquisition and processing; 2.1) Surface electromyogram signal acquisition: Pre-acquire the surface electromyogram signals sEMG of the patient's affected limb in the relaxed state and the maximum voluntary contraction MVC state, and record the root mean square value RMS of the electromyogram r , RMS m ; At the same time, determine the following stimulation parameters: Minimum stimulation parameters: the minimum stimulation amplitude A min and the minimum stimulation pulse width W min ; Maximum stimulation parameters: the maximum stimulation amplitude A without causing pain max and the maximum stimulation pulse width W max ; After determining the stimulation parameters, the stimulation pulse amplitude range is set at A min -A max , and the stimulation pulse width range is W min -W max ; 2.2) Surface electromyogram signal processing; 3) Stimulation cycle, biofeedback electrical stimulation: When performing neuromuscular electrical stimulation, the passive contraction effect of the quadriceps femoris is optimized by adjusting the stimulation parameters in real time, and the electrical stimulation is stopped in time when muscle fatigue is detected. The specific process is as follows: Calculate the fatigue characteristic Q value by combining non-linear characteristic parameters and time-domain characteristic parameters to detect the muscle fatigue state in real time: Among them, u s is the sample entropy value of the electromyogram signal at the stimulation site, and u i is the integral value of the electromyogram signal at the stimulation site; Sample entropy is used to describe the complexity of the time series. The smaller the value, the lower the complexity of the time series; when the muscle is fatigued, the muscle change rate slows down, and the activities of motor units tend to be consistent, resulting in a decrease in sample entropy. At the same time, as the fatigue degree deepens, the muscle gradually becomes tense, and more motor units will be recruited in the fatigued part, increasing the discharge frequency and synchronization degree, thus increasing the integrated electromyogram value; the characteristic Q value in the fatigued state will be significantly smaller than the initial state and continue to decrease as the fatigue degree deepens. When it is detected that the Q value is lower than a certain specific threshold, it can be determined that the muscle enters the fatigued state and the electrical stimulation output is stopped.

6. The biofeedback electrical stimulation method for quadriceps rehabilitation training according to claim 5, wherein: To avoid interference of the stimulation electrode on the surface electromyogram signal acquisition, the acquisition cycle in step 2) and the stimulation cycle in step 3) run alternately; During the acquisition period, surface electromyogram signals are acquired and their characteristic values RMS(k), the pulse amplitude A(k + 1), pulse width W(k + 1), and fatigue characteristic Q(k) of the next period are calculated. When Q(k) is greater than the fatigue threshold, the electrical stimulation module outputs electrical stimulation according to the values of A(k + 1) and W(k + 1) during the stimulation period; The calculation formula for the stimulation parameters is as follows: Where Q(0): the initial Q value; b: muscle fatigue threshold.

7. The biofeedback electrical stimulation method for quadriceps rehabilitation training according to claim 6, characterized in that: The muscle fatigue threshold is 50%, that is, when Q(k) < Q(0) / 2, the system determines muscle fatigue and stops electrical stimulation.

8. The biofeedback electrical stimulation method for quadriceps rehabilitation training according to claim 6, characterized in that: When RMS(k) does not exceed the maximum tolerable stimulus value RMS m the stimulation parameters increase as the EMG value RMS(k) increases; when RMS(k) > RMS m the stimulation parameters are fixed at the maximum stimulation parameters; when RMS(k) < RMS r the stimulation parameters are fixed at the minimum stimulation parameters.

9. The biofeedback electrical stimulation method for quadriceps rehabilitation training according to claim 5, characterized in that: The electrical stimulation parameters in step 1) include electrical stimulation waveform, pulse width, pulse frequency, pulse amplitude, and duty cycle. The electrical stimulation waveform uses a constant current bipolar symmetric square wave, the pulse width is 0.1 - 0.5 ms, the pulse frequency is 1 - 10 Hz, and the pulse amplitude is 5 - 30 mA; different types of muscle fibers are suitable for different duty cycles: type I muscle fibers use a 1:1 duty cycle, type IIA muscle fibers use a 1:2 duty cycle, and type IIB muscle fibers use a 1:3 duty cycle.

10. The biofeedback electrical stimulation method for quadriceps rehabilitation training according to claim 5, characterized in that: The specific process of surface electromyogram signal processing in step 2.2) is as follows: the signal is amplified using a pre - differential amplifier circuit, and at the same time, a band - pass filter circuit is used to process the signal to filter out noise components outside the range of 10 - 500 Hz; a post - adjustable amplifier circuit is used to automatically adjust the amplification factor of the signal, and an analog - to - digital conversion module ADC is used to convert the electrical signal.

Images