Walking simulation electrical stimulation system for bedridden lower limb arteriosclerosis obliterans patient

Abstract

According to the walking simulation electrical stimulation system for the bedridden lower limb arteriosclerosis obliterans patient, after the control host receives a walking simulation electrical stimulation instruction, the multi-channel electrical stimulation generation unit is controlled to generate an electrical stimulation signal matched with the target walking simulation stimulation level, and the electrical stimulation signal is applied to the lower limbs of the patient through the multi-channel stimulation electrode; controlling pause when the target stimulation time is reached; the multi-channel myoelectricity acquisition unit acquires myoelectricity signals of the lower limbs of the patient through a multi-channel myoelectricity electrode; the control host judges that the multi-channel electromyographic electrodes are not in contact with the patient well according to the multi-channel electromyographic signals within the first set time from the beginning, gives an alarm, and obtains the multi-channel electromyographic signals within the second set time from the end of the target stimulation time after judging that the contact is good; performing preprocessing and feature extraction on the electromyographic signals of each channel within the time to obtain multi-channel features, and inputting the multi-channel features into a target neural network recognition model to recognize and obtain a lower limb arteriosclerosis obliterans recognition result of the patient.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical stimulation systems, and in particular to a walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans. The invention technology originates from the project: National Key R&D Program: "Research on Pan-vascular Disease Screening, System Evaluation and Prevention and Treatment System", No.: 2021YFC2500500. Background Art

[0002] Lower extremity arteriosclerosis obliterans is a common vascular disease, and patients often suffer from lower extremity pain, numbness, and dysfunction. Walking exercises are of great significance for improving blood circulation in patients' lower extremities and relieving symptoms. However, some patients are unable to perform normal walking exercises due to their severe condition. Although the electrical stimulation devices in the prior art can help patients exercise to a certain extent, they cannot meet the demand for simulated walking exercises while lying in bed. Summary of the invention

[0003] In view of the problems and shortcomings of the prior art, the present invention provides a walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans, which enables the patients to perform simulated walking exercises while lying in bed, thereby improving blood circulation in the lower limbs.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] The present invention provides a walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans, which is characterized in that it comprises a lower limb fixing device and a control box, the lower limb fixing device comprises an L-shaped support frame, the L-shaped support frame is composed of a horizontal base and a vertical support column, the front and rear ends of the horizontal base are both fixed with oblique support frames, the front and rear ends of the oblique support frames are fixed with lower limb arc placement frames matching the shapes of human lower limbs and feet for placing human lower limbs and feet thereon, and the vertical support column is fixed with a pendant for hanging on the guardrail at the end of the bed;

[0006] The control box comprises a host shell, in which a multi-channel electrical stimulation generating unit, a multi-channel electromyography acquisition unit and a control host are arranged, the host shell extends a multi-channel stimulation electrode connected to each channel in a one-to-one correspondence and a multi-channel electromyography electrode connected to each channel in a one-to-one correspondence, the control host pre-stores a plurality of one-to-one correspondences between walking simulation stimulation levels and electrical stimulation signals, and an alarm is embedded in the host shell;

[0007] The control host is used to control the multi-channel electrical stimulation generation unit to generate an electrical stimulation signal matching the target walking simulation stimulation level and apply the electrical stimulation signal to the lower limbs of the target patient through the multi-channel stimulation electrodes after receiving the walking simulation electrical stimulation instruction containing the target stimulation time and the target walking simulation stimulation level, and control the multi-channel electrical stimulation generation unit to pause when the target stimulation time is reached;

[0008] The multi-channel electromyography acquisition unit is used to collect electromyography signals of the lower limbs of the target patient through multi-channel electromyography electrodes in a set acquisition cycle and transmit them to the control host;

[0009] The control host is also used to determine the multi-channel electromyographic signal within the first set time from the beginning, and if the electromyographic signal of any channel within N consecutive acquisition cycles has an amplitude that is no longer within the set amplitude range, it is determined that the multi-channel electromyographic electrode is not in good contact with the target patient, and the alarm is controlled to issue a warning message to remind reinstallation, where N is a positive integer;

[0010] The control host is also used to obtain the multi-channel electromyographic signal within a second set time from the end of the target stimulation time when it is determined that the multi-channel electromyographic electrode is in good contact with the target patient, perform signal preprocessing and signal feature extraction on the electromyographic signal of each channel within this time, obtain the single electromyographic time domain feature and the single electromyographic frequency domain feature of each channel and form a multi-channel feature, input the multi-channel feature into the target neural network recognition model for recognition to obtain the recognition result of lower limb arteriosclerosis obliterans of the target patient, and output the recognition result of lower limb arteriosclerosis obliterans and the target stimulation time and target walking simulation stimulation level of the next electrical stimulation when it is recognized that the target patient has lower limb arteriosclerosis obliterans, and output the recognition result of lower limb arteriosclerosis obliterans when it is recognized that the target patient does not have lower limb arteriosclerosis obliterans, and the recognition result of lower limb arteriosclerosis obliterans includes no lower limb arteriosclerosis obliterans, mild lower limb arteriosclerosis obliterans, moderate lower limb arteriosclerosis obliterans and severe lower limb arteriosclerosis obliterans.

[0011] The positive and progressive effects of the present invention are:

[0012] The present invention is suitable for patients with lower limb arteriosclerosis obliterans, who can perform simulated walking exercises while lying in bed to improve blood circulation in the lower limbs; the system of the present invention is easy to operate, and the patient can adjust stimulation parameters according to their own needs; the lower limb fixation device of the present invention enables the patient's lower limbs to maintain a suitable position, thereby improving the electrical stimulation effect; the present invention uses electrical stimulation signals for simulated walking exercises, which are closer to actual walking exercises and are helpful for the patient's recovery. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a control principle diagram of a walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans according to a preferred embodiment of the present invention.

[0014] Figure 2-3 Schematic diagram of the structure of a lower limb fixation device according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0016] like Figure 1-3 As shown, the present embodiment provides a bed-ridden lower limb arteriosclerosis obliterans walking simulation electrical stimulation system, which includes a lower limb fixation device 10, the lower limb fixation device 10 includes an L-shaped support frame, the L-shaped support frame is composed of a horizontal base 11 and a vertical support column 12, the horizontal base 11 is formed by splicing two horizontal rods 111 arranged on the left and right and two longitudinal rods 112 arranged on the front and back, and the shape formed by the splicing is a rectangle, the front and rear longitudinal rods 112 of the horizontal base 11 are both fixed with an oblique support frame 13, and the front and rear end oblique support frames 13 are fixed The support is provided with a lower limb arc placement frame 14 that matches the shape of the lower limbs and feet of the human body, which is used for placing the lower limbs and feet of the human body. The vertical support column 12 is composed of two vertical columns 121, and the two vertical columns 121 are respectively fixed at the front left end and the front right end of the horizontal base 11. A connecting rod 122 is fixed between the two vertical columns 121. A hanger 15 for hanging on the guardrail at the end of the bed is fixed on the vertical support column 12. The hanger 15 includes two hooks 151, and the two hooks 151 are respectively inserted and fixed in the corresponding vertical columns 121.

[0017] In this embodiment, the lower limb fixation device 10 is used to place the patient's lower limbs. The lower limb fixation device 10 is hung on the guardrail at the end of the bed through the hanger 15, so that the patient's lower limbs can be kept in a suitable position on the bed for simulated walking exercises.

[0018] The electrical stimulation system also includes a control box, which includes a mainframe housing, in which a multi-channel electrical stimulation generation unit, a multi-channel electromyography acquisition unit and a control mainframe 20 are arranged, the multi-channel electrical stimulation generation unit includes a stimulation signal generator 21 and a multi-channel power amplifier 22, the mainframe housing extends a multi-channel stimulation electrode 23 connected to each channel of the multi-channel power amplifier 22 in a one-to-one correspondence, the multi-channel electromyography acquisition unit includes a multi-channel preamplifier 24 and a multi-channel analog-to-digital converter 25, the mainframe housing extends a multi-channel electromyography electrode 26 connected to each channel of the multi-channel preamplifier 24 in a one-to-one correspondence, an alarm 27 is embedded in the mainframe housing, and a power supply unit is also arranged in the mainframe housing, which is a rechargeable lithium battery to provide a stable power supply for the electrical stimulation system. The multi-channel stimulation electrode 23 and the multi-channel electromyography electrode 26 are made of wearable flexible materials with good conductivity and comfort.

[0019] The control host 20 pre-stores a one-to-one correspondence between multiple walking simulation stimulation levels (stimulation intensity, stimulation frequency, stimulation waveform, etc.) and electrical stimulation signals, and also pre-stores a one-to-one correspondence between the symptom degree of lower limb arteriosclerosis obliterans and the target stimulation time of electrical stimulation and its target walking simulation stimulation level.

[0020] During use, the target patient lies on the bed, attaches the multi-channel stimulation electrode 23 and the multi-channel electromyography electrode 26 to the calf muscles, hangs the lower limb fixation device 10 on the bed end guardrail through the pendant 15, and places the target patient's calf on the lower limb arc placement frame 14. The electrical stimulation system is turned on and starts to work, helping the patient to perform simulated walking exercises and improve blood circulation in the lower limbs.

[0021] The control host 20 is used to control the stimulation signal generator 21 to generate an electrical stimulation signal matching the target walking simulation stimulation level after receiving a walking simulation electrical stimulation instruction containing a target stimulation time and a target walking simulation stimulation level. The multi-channel power amplifier 22 is used to amplify the power of the electrical stimulation signal and apply it to the lower limb muscles of the target patient through the multi-channel stimulation electrodes 23, and control the stimulation signal generator 21 to pause when the target stimulation time is reached.

[0022] The multi-channel preamplifier 24 is used to collect the electromyographic signals of the lower limbs of the target patient through the multi-channel electromyographic electrodes 26 in a set collection cycle and amplify the signals. The multi-channel analog-to-digital converter 25 is used to convert the electromyographic signals after amplification of each channel signal into analog-to-digital signals and transmit them to the control host 20.

[0023] The control host 20 is also used to judge the multi-channel electromyographic signal within the first set time from the start. If it is determined that the amplitude of the electromyographic signal of any channel within N consecutive acquisition cycles is no longer within the set amplitude range, it is determined that the multi-channel electromyographic electrode is not in good contact with the target patient, and the control alarm 27 sends a warning message to remind reinstallation, where N is a positive integer.

[0024] The control host 20 is also used to obtain the multi-channel electromyographic signal within a second set time from the end of the target stimulation time after determining that the multi-channel electromyographic electrode 26 is in good contact with the target patient (the end time point of the second set time is the end time point of the target stimulation time), perform signal preprocessing and signal feature extraction on the electromyographic signal of each channel within this time, obtain the single electromyographic time domain feature and the single electromyographic frequency domain feature of each channel and form a multi-channel feature, input the multi-channel feature into the target neural network recognition model for recognition to obtain the lower limb arteriosclerosis obliterans recognition result of the target patient, and output the lower limb arteriosclerosis obliterans recognition result and the target stimulation time and target walking simulation stimulation level for the next electrical stimulation when it is identified that the target patient does not have lower limb arteriosclerosis obliterans, output the lower limb arteriosclerosis obliterans recognition result when it is identified that the target patient does not have lower limb arteriosclerosis obliterans, and the lower limb arteriosclerosis obliterans recognition result includes no lower limb arteriosclerosis obliterans, mild lower limb arteriosclerosis obliterans, moderate lower limb arteriosclerosis obliterans and severe lower limb arteriosclerosis obliterans.

[0025] Furthermore, when it is identified that the target patient has lower limb arteriosclerosis obliterans and specifically mild lower limb arteriosclerosis obliterans, the identification result of lower limb arteriosclerosis obliterans is mild lower limb arteriosclerosis obliterans, as well as the target stimulation time and target walking simulation stimulation level of the next electrical stimulation corresponding to mild lower limb arteriosclerosis obliterans are output and stored.

[0026] When it is identified that the target patient has lower limb arteriosclerosis obliterans, specifically moderate lower limb arteriosclerosis obliterans, the lower limb arteriosclerosis obliterans identification result is moderate lower limb arteriosclerosis obliterans, as well as the target stimulation time and target walking simulation stimulation level of the next electrical stimulation corresponding to the moderate lower limb arteriosclerosis obliterans.

[0027] When it is identified that the target patient has lower limb arteriosclerosis obliterans, specifically severe lower limb arteriosclerosis obliterans, the lower limb arteriosclerosis obliterans identification result is severe lower limb arteriosclerosis obliterans, as well as the target stimulation time and target walking simulation stimulation level of the next electrical stimulation corresponding to severe lower limb arteriosclerosis obliterans are output and stored.

[0028] In this embodiment, the specific implementation method of constructing the target neural network recognition model is as follows: the control host 20 is also used to control the multi-channel electrical stimulation generation unit to generate an electrical stimulation signal matching the walking simulation stimulation level corresponding to the symptom degree of each historical patient and apply it to the lower limbs of the historical patient through the multi-channel stimulation electrode 23. The symptom degree of the historical patient includes no lower limb arteriosclerosis obliterans, mild lower limb arteriosclerosis obliterans, moderate lower limb arteriosclerosis obliterans and severe lower limb arteriosclerosis obliterans. The collected historical patients include historical patients without lower limb arteriosclerosis obliterans, historical patients with mild lower limb arteriosclerosis obliterans, historical patients with moderate lower limb arteriosclerosis obliterans and historical patients with severe lower limb arteriosclerosis obliterans.

[0029] The control host 20 is also used to collect the electromyographic signals of the lower limbs of each historical patient through a multi-channel electromyographic acquisition unit and a multi-channel electromyographic electrode 26 .

[0030] The control host 20 is also used to judge the multi-channel electromyographic signal within the first set time from the beginning. If the amplitude of the electromyographic signal of any channel in N consecutive acquisition cycles is no longer within the set amplitude range, it is determined that the multi-channel electromyographic electrode 26 is not in good contact with the historical patient, and the control alarm 27 sends a warning message to remind reinstallation.

[0031] The control host 20 is also used to obtain multi-channel electromyographic signals within a period of time starting from the end of the first set time after determining that the multi-channel electromyographic electrode 26 is in good contact with the historical patient, perform signal preprocessing and signal feature extraction on the electromyographic signals of each channel, obtain the single electromyographic time domain features and single electromyographic frequency domain features of each channel of the historical patient and form multi-channel features, thereby obtaining the multi-channel features of each historical patient.

[0032] A feature set is constructed for each historical patient, the feature set being composed of multi-channel features and corresponding symptom levels, and a training sample set is constructed, the training sample set including the feature set of each historical patient.

[0033] A neural network recognition model is constructed based on the convolutional neural network algorithm. The particle swarm + genetic algorithm is used to optimize the hyperparameters of the neural network recognition model. The model is trained using a training sample set. Through continuous iterative optimization, the optimal hyperparameters are obtained, and the target neural network recognition model is constructed using the optimal hyperparameters.

[0034] Specifically, a neural network recognition model is constructed based on the convolutional neural network algorithm, and the model hyperparameters are initialized. Each particle in the particle swarm represents a set of hyperparameters. The number of particles L, the maximum number of iterations T, and the particle search space dimension D corresponding to the hyperparameters are set, and the speed and position of each particle are initialized.

[0035] A corresponding neural network recognition model is constructed based on each particle. The neural network recognition model corresponding to the particle is trained using the training sample set, and the corresponding fitness is calculated using the fitness function. The fitness function is a function that is inversely proportional to the mean square error of the model deviation. The greater the fitness, the better the corresponding model training. Similarly, the fitness of each particle can be calculated. The maximum fitness of each particle is taken as the optimal fitness, the current position of each particle is taken as the individual optimal position of each particle, and the current position of the particle corresponding to the optimal fitness is taken as the group optimal position of the particle swarm.

[0036] Update the velocity and position of each particle:

[0037]

[0038] X ld (t+1)=X ld (t)+V ld (t+1)

[0039]

[0040] Among them, V ld (t) and X ld (t) represent the velocity and position of the dth dimension of the lth particle in the tth iteration, V ld (t+1) and X ld (t+1) represent the velocity and position of the dth dimension of the lth particle in the t+1th iteration, C1 and C2 represent the acceleration factors, W(t) represents the adaptive inertia weight, and W min represents the minimum inertia weight, W max Indicates the maximum inertia weight, Xpb ld Indicates the optimal position of the historical individual, and the corresponding fitness is the optimal fitness of the historical individual, Xgb d It represents the optimal position of the historical group, and the corresponding fitness is the optimal fitness of the historical group.

[0041] Update the historical individual optimal position and the historical group optimal position: use the fitness function to calculate the fitness of each particle after the speed and position are updated, compare the updated fitness of each particle with the corresponding historical individual optimal fitness, if the updated fitness of a particle is greater than the corresponding historical individual optimal fitness, then update the historical individual optimal fitness of the particle = the updated fitness of the particle, and update the historical individual optimal position of the particle = the updated position of the particle, if the updated fitness of the particle is not greater than the corresponding historical individual optimal fitness, then do not update; then, take the maximum value of the historical individual optimal fitness of each particle as the current optimal fitness, if the current optimal fitness is greater than the historical group optimal fitness, then update the historical group optimal fitness = the current optimal fitness, and update the historical group optimal position = the current position of the particle corresponding to the current optimal fitness value, if the current optimal fitness is not greater than the historical group optimal fitness, then do not update.

[0042] Determine whether the optimal fitness of the historical group reaches the set fitness or whether the number of iterations reaches the maximum number of iterations T. If yes, obtain the optimal hyperparameters based on the particles in the optimal position of the historical group, and use the optimal hyperparameters to build the target neural network recognition model. Otherwise, determine whether the current number of iterations t is less than the set threshold T1. If yes, use the genetic algorithm to perform hybridization operations on the updated particles, and update the speed and position of each particle in the particle group formed after hybridization. If no, directly update the speed and position again, T / 2≤T1≤T.

[0043] Among them, a genetic algorithm is used to perform a hybridization operation on the updated particles, and the speed and position of each particle in the particle population formed after hybridization are updated again. The specific implementation method is: the current historical individual optimal fitness of each particle is arranged in descending order, and L1 particles with the lowest order are eliminated. The particle with the optimal position in the historical group among the non-eliminated particles is used as the parent generation, and L1 particles are randomly selected from the remaining non-eliminated particles as the parent generation. Each selected parent generation is hybridized with the parent generation to generate L1 particles. The current positions of the L1 particles generated by hybridization are used as the positions of the corresponding particles after t iterations. The current speed of each particle in the L1 particles generated by hybridization is the average speed of the particles on both sides of the particle, so as to obtain the particle population formed after hybridization, and the speed and position of each particle in the particle population formed after hybridization are updated again. L1 is a positive integer and 1≤L1≤L / 2, t=t+1.

[0044] In this embodiment, hybridization operations are used in the early and middle stages of iteration to prevent the particle swarm from falling into a local optimal solution, and in the late stages of iteration, hybridization operations are reduced to ensure the stability and convergence of the solution. This embodiment combines the advantages of both the particle swarm algorithm and the genetic algorithm to construct an optimized neural network recognition model.

[0045] Example 2

[0046] On the basis of Example 1, the present embodiment further optimizes the technology as follows: the control host 20 is used to identify the identity of the target patient after receiving the walking simulation electrical stimulation instruction containing the identity information of the target patient, and when it is identified that the target patient is using the electrical stimulation system for the first time, determine whether the walking simulation electrical stimulation instruction contains the target stimulation time and the target walking simulation stimulation level, and when they are contained, control the stimulation signal generator 21 to generate an electrical stimulation signal matching the target walking simulation stimulation level, the multi-channel power amplifier 22 amplifies the electrical stimulation signal and applies it to the lower limb muscles of the target patient through the multi-channel stimulation electrodes 23, and controls the stimulation signal generator 21 to pause when the target stimulation time is reached, and when they are not contained, use the preset default stimulation time and the default walking simulation stimulation level as the target stimulation time and the target walking simulation stimulation level, control the stimulation signal generator 21 to generate an electrical stimulation signal matching the target walking simulation stimulation level, the multi-channel power amplifier 22 amplifies the electrical stimulation signal and applies it to the lower limb muscles of the target patient through the multi-channel stimulation electrodes, and controls the multi-channel electrical stimulation generation unit to pause when the target stimulation time is reached.

[0047] When it is identified that this is not the first time for the target patient to use the electrical stimulation system, the multi-channel stimulation signal generator 21 is controlled to generate an electrical stimulation signal matching the target walking simulation stimulation level based on the target stimulation time and target walking simulation stimulation level stored last time, the multi-channel power amplifier 22 amplifies the power of the electrical stimulation signal and applies it to the lower limb muscles of the target patient through the multi-channel stimulation electrodes, and controls the multi-channel electrical stimulation generation unit to pause when the target stimulation time is reached.

[0048] Another implementation scheme when it is identified that this is the first time for the target patient to use this electrical stimulation system and it is determined that the walking simulation electrical stimulation instruction does not include the target stimulation time and the target walking simulation stimulation level is: output a custom electrical stimulation mode and a default electrical stimulation mode for the target patient to choose, and when the custom electrical stimulation mode is selected, electrical stimulation is performed according to the input target stimulation time and target walking simulation stimulation level; when the default electrical stimulation mode is selected, electrical stimulation is performed according to the default stimulation time and default walking simulation stimulation level.

[0049] Although the specific embodiments of the present invention are described above, those skilled in the art should understand that these are only examples, and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims

1. A walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans, characterized in that: It includes a lower limb fixing device and a control box, the lower limb fixing device includes an L-shaped support frame, the L-shaped support frame is composed of a horizontal base and a vertical support column, the front and rear ends of the horizontal base are fixed with oblique support frames, the front and rear ends of the oblique support frames are fixed with lower limb arc placement frames matching the shapes of human lower limbs and feet for placing human lower limbs and feet thereon, and the vertical support column is fixed with a pendant for hanging on the guardrail at the end of the bed; The control box comprises a host shell, in which a multi-channel electrical stimulation generating unit, a multi-channel electromyography acquisition unit and a control host are arranged, the host shell extends a multi-channel stimulation electrode connected to each channel in a one-to-one correspondence and a multi-channel electromyography electrode connected to each channel in a one-to-one correspondence, the control host pre-stores a plurality of one-to-one correspondences between walking simulation stimulation levels and electrical stimulation signals, and an alarm is embedded in the host shell; The control host is used to control the multi-channel electrical stimulation generation unit to generate an electrical stimulation signal matching the target walking simulation stimulation level and apply the electrical stimulation signal to the lower limbs of the target patient through the multi-channel stimulation electrodes after receiving the walking simulation electrical stimulation instruction containing the target stimulation time and the target walking simulation stimulation level, and control the multi-channel electrical stimulation generation unit to pause when the target stimulation time is reached; The multi-channel electromyography acquisition unit is used to collect electromyography signals of the lower limbs of the target patient through multi-channel electromyography electrodes in a set acquisition cycle and transmit them to the control host; The control host is also used to determine the multi-channel electromyographic signal within the first set time from the beginning, and if the electromyographic signal of any channel within N consecutive acquisition cycles has an amplitude that is no longer within the set amplitude range, it is determined that the multi-channel electromyographic electrode is not in good contact with the target patient, and the alarm is controlled to issue a warning message to remind reinstallation, where N is a positive integer; The control host is also used to obtain the multi-channel electromyographic signal within a second set time from the end of the target stimulation time after determining that the multi-channel electromyographic electrode is in good contact with the target patient, perform signal preprocessing and signal feature extraction on the electromyographic signal of each channel within this time, obtain the single electromyographic time domain feature and the single electromyographic frequency domain feature of each channel and form a multi-channel feature, input the multi-channel feature into the target neural network recognition model for recognition to obtain the recognition result of lower limb arteriosclerosis obliterans of the target patient, and output the recognition result of lower limb arteriosclerosis obliterans and the target stimulation time and target walking simulation stimulation level of the next electrical stimulation when it is recognized that the target patient has lower limb arteriosclerosis obliterans, and output the recognition result of lower limb arteriosclerosis obliterans when it is recognized that the target patient does not have lower limb arteriosclerosis obliterans, and the recognition result of lower limb arteriosclerosis obliterans includes no lower limb arteriosclerosis obliterans, mild lower limb arteriosclerosis obliterans, moderate lower limb arteriosclerosis obliterans and severe lower limb arteriosclerosis obliterans.

2. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 1, characterized in that: The control host is also used for outputting and storing, when it is identified that the target patient has lower limb arteriosclerosis obliterans and specifically mild lower limb arteriosclerosis obliterans, the identification result of lower limb arteriosclerosis obliterans is mild lower limb arteriosclerosis obliterans, as well as the target stimulation time and target walking simulation stimulation level of the next electrical stimulation corresponding to the mild lower limb arteriosclerosis obliterans; When it is identified that the target patient has lower limb arteriosclerosis obliterans, and specifically moderate lower limb arteriosclerosis obliterans, the identification result of lower limb arteriosclerosis obliterans is moderate lower limb arteriosclerosis obliterans, and the target stimulation time and target walking simulation stimulation level of the next electrical stimulation corresponding to the moderate lower limb arteriosclerosis obliterans are output and stored; When it is identified that the target patient has lower limb arteriosclerosis obliterans, specifically severe lower limb arteriosclerosis obliterans, the lower limb arteriosclerosis obliterans identification result is severe lower limb arteriosclerosis obliterans, as well as the target stimulation time and target walking simulation stimulation level of the next electrical stimulation corresponding to severe lower limb arteriosclerosis obliterans are output and stored.

3. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 2, characterized in that: The control host is used to identify the identity of the target patient after receiving the walking simulation electrical stimulation instruction containing the identity information of the target patient, and when it is identified that the target patient is using this electrical stimulation system for the first time, determine whether the walking simulation electrical stimulation instruction contains the target stimulation time and the target walking simulation stimulation level, and if so, control the multi-channel electrical stimulation generation unit to generate an electrical stimulation signal matching the target walking simulation stimulation level and apply it to the lower limbs of the target patient through the multi-channel stimulation electrodes, and control the multi-channel electrical stimulation generation unit to pause when the target stimulation time is reached, and if not, use the preset default stimulation time and the default walking simulation stimulation level as the target stimulation time and the target walking simulation stimulation level, control the multi-channel electrical stimulation generation unit to generate an electrical stimulation signal matching the target walking simulation stimulation level and apply it to the lower limbs of the target patient through the multi-channel stimulation electrodes, and control the multi-channel electrical stimulation generation unit to pause when the target stimulation time is reached; When it is identified that this is not the first time that the target patient uses the electrical stimulation system, the multi-channel electrical stimulation generation unit is controlled to generate an electrical stimulation signal matching the target walking simulation stimulation level and apply it to the lower limbs of the target patient through the multi-channel stimulation electrodes based on the target stimulation time and target walking simulation stimulation level stored last time, and the multi-channel electrical stimulation generation unit is controlled to pause when the target stimulation time is reached.

4. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 3, characterized in that: The control host is used to output a custom electrical stimulation mode and a default electrical stimulation mode for the target patient to select when it is identified that the target patient is using the electrical stimulation system for the first time and it is determined that the walking simulation electrical stimulation instruction does not include the target stimulation time and the target walking simulation stimulation level. When the custom electrical stimulation mode is selected, electrical stimulation is performed according to the input target stimulation time and the target walking simulation stimulation level; when the default electrical stimulation mode is selected, electrical stimulation is performed according to the default stimulation time and the default walking simulation stimulation level.

5. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 1, characterized in that: The control host is also used to construct a target neural network recognition model: control the multi-channel electrical stimulation generation unit to generate an electrical stimulation signal matching the walking simulation stimulation level corresponding to the symptom degree of each historical patient and apply the electrical stimulation signal to the lower limbs of the historical patient through the multi-channel stimulation electrode, wherein the symptom degree of the historical patient includes no lower limb arteriosclerosis obliterans, mild lower limb arteriosclerosis obliterans, moderate lower limb arteriosclerosis obliterans and severe lower limb arteriosclerosis obliterans; The control host is also used to collect the electromyographic signals of the lower limbs of each historical patient through a multi-channel electromyographic acquisition unit and a multi-channel electromyographic electrode; The control host is also used to determine the multi-channel electromyographic signal within the first set time from the beginning, and if the electromyographic signal of any channel within N consecutive acquisition cycles has an amplitude that is no longer within the set amplitude range, it is determined that the multi-channel electromyographic electrode is not in good contact with the historical patient, and the alarm is controlled to issue a warning message to remind reinstallation; The control host is also used to obtain the multi-channel electromyographic signals within a period of time starting from the end of the first set time after determining that the multi-channel electromyographic electrodes are in good contact with the historical patient, perform signal preprocessing and signal feature extraction on the electromyographic signals of each channel, obtain the single electromyographic time domain features and single electromyographic frequency domain features of each channel of the historical patient and form multi-channel features, thereby obtaining the multi-channel features of each historical patient; Construct a feature set for each historical patient, the feature set consists of multi-channel features and corresponding symptom levels, and construct a training sample set, the training sample set includes the feature set of each historical patient; A neural network recognition model is constructed based on the convolutional neural network algorithm. The particle swarm + genetic algorithm is used to optimize the hyperparameters of the neural network recognition model. The model is trained using a training sample set. The optimal hyperparameters are obtained through continuous iteration, and the target neural network recognition model is constructed using the optimal hyperparameters.

6. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 1, characterized in that: A power supply unit is also provided in the host housing, and the power supply unit is used to provide power for the electrical stimulation system.

7. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 1, characterized in that: The transverse base is in a rectangular shape formed by splicing two transverse rods arranged on the left and right and two longitudinal rods arranged on the front and back. The front and rear longitudinal rods of the transverse base are both fixed with oblique support frames.

8. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 7, characterized in that: The vertical support column is composed of two vertical columns, and the two vertical columns are respectively fixed to the front left end and the front right end of the horizontal base, and a connecting rod is fixed between the two vertical columns.

9. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 8, characterized in that: The hanging component comprises two hooks, and the two hooks are respectively inserted and fixed in corresponding vertical columns.

10. The walking simulation electrical stimulation system for bedridden patients with lower limb arteriosclerosis obliterans as claimed in claim 1, characterized in that: The multi-channel electrical stimulation generating unit comprises a stimulation signal generator and a multi-channel power amplifier, wherein the stimulation signal generator is used to generate an electrical stimulation signal matching the target walking simulation stimulation level, and the multi-channel power amplifier is used to amplify the power of the electrical stimulation signal and apply it to the lower limb muscles of the target patient through the multi-channel stimulation electrodes; The multi-channel electromyography acquisition unit includes a multi-channel preamplifier and a multi-channel analog-to-digital converter. The multi-channel preamplifier is used to collect the electromyography signals of the lower limbs of the target patient through multi-channel electromyography electrodes with a set acquisition period and amplify the signals. The multi-channel analog-to-digital converter is used to convert the electromyography signals of each channel after signal amplification into analog-to-digital signals and transmit them to the control host.

Images