Subchin electrical stimulation combined upper airway muscle group training system and method based on bioelectric feedback

Abstract

The invention relates to a chin electrical stimulation combined upper airway muscle group training system and method based on bioelectric feedback, and belongs to the field of upper airway muscle group training. Comprising a pressure detection module used for detecting a pressure value in an upper respiratory tract muscle group training process; the myoelectricity acquisition module is used for detecting myoelectricity numerical values in the training process of the upper respiratory tract muscle group; the electrical stimulation output module is used for stimulating partial nerve excitation and muscle contraction of the muscle group of the upper respiratory tract of the trainee; the signal conditioning module is used for carrying out preprocessing and analog-to-digital conversion on signals transmitted among the modules; and the control module is used for controlling signal interaction in the whole training process. And the display interaction terminal is used for interacting with the trainee. Based on bioelectricity feedback, a new individualized electrical stimulation and upper airway muscle group training treatment mode is newly established, accurate electrical stimulation and training intensity are formulated, and fatigue and injury are avoided.

Description

Technical Field

[0001] The invention relates to a submental electrical stimulation combined with upper airway muscle group training system and method based on bioelectric feedback, belonging to the field of upper airway muscle group training. Background Art

[0002] At present, most upper airway muscle training models focus on improving muscle strength, namely tongue muscle-resistance strength training, which is a classic training model that gradually increases the contraction strength of targeted muscles to resist resistance. It is considered to be an effective method to increase the maximum contraction strength of muscles and corresponding neural plasticity. Previous clinical trials have found that although the acceptance of tongue muscle training is higher than that of CPAP and other treatments, it has a high time requirement, so long-term compliance is difficult to ensure, which affects the efficacy and limits the clinical application of tongue muscle training. If the training model can be optimized, the training time can be shortened, and the training efficiency can be improved, the compliance of trainees can be improved while ensuring the efficacy. Summary of the invention

[0003] The present invention is based on bioelectric feedback and creates a new individualized treatment model of submental electrical stimulation combined with upper airway muscle group training. Digital sensors, digital signal processing and other technologies are introduced into muscle function training devices to provide a submental electrical stimulation combined with upper airway muscle group training system and method based on bioelectric feedback, realizing daily digital visualization real-time feedback and intelligent interaction.

[0004] The technical solution adopted by the present invention to achieve the above-mentioned purpose is:

[0005] A submental electrical stimulation combined with upper airway muscle group training system based on bioelectric feedback, comprising:

[0006] Pressure detection module, used to detect the pressure value during the upper respiratory tract muscle training process;

[0007] The electromyography acquisition module is used to detect the electromyography values ​​during the training of the upper respiratory tract muscles;

[0008] The electrical stimulation output module is used to stimulate the nerve excitement and muscle contraction of the trainee's upper respiratory tract muscles;

[0009] Signal conditioning module, used for preprocessing and analog-to-digital conversion of signals transmitted between modules;

[0010] Control module, used to control the signal interaction during the entire training process;

[0011] Display interactive terminal for interaction with trainers.

[0012] The pressure detection module is an S-shaped tension pressure sensor, which is placed in front of the trainee's face, collects the pressure applied by the tongue on the effective area of ​​the sensor, and converts the resistance value change into the voltage value change through the signal conditioning module.

[0013] The myoelectric acquisition module is an electrode ball placed under the trainee's tongue. The electrode ball contains positive and negative poles connected to the signal conditioning module through silver wires to collect the potential difference signal generated by muscle contraction when the upper respiratory tract muscles move.

[0014] The electrical stimulation output module is a disposable medical electrode sheet, which is attached to the trainee's chin. The square wave pulses emitted by the control module generate electrical stimulation, causing nerve excitement in the corresponding part and contraction of the tongue root muscles.

[0015] A submental electrical stimulation combined with upper airway muscle group training method based on bioelectric feedback comprises the following steps:

[0016] 1) Before training, personalize the basic information of the trainee;

[0017] 2) Conduct submental electrical stimulation combined with upper airway muscle training on the trainees;

[0018] 3) Predict the training effect by collecting the electromyography and pressure values ​​during the training process.

[0019] The step 1) comprises the following steps:

[0020] 1.1) Individual measurement of electrical stimulation signal: A disposable medical electrode is attached to the trainee's chin, and an electrical stimulation signal of a certain frequency and amplitude is given. The amplitude of the signal is gradually adjusted. When the trainee feels a noticeable pulling sensation at the root of the tongue or difficulty in swallowing, the amplitude U at this time is recorded as the appropriate electrical stimulation signal for the trainee.

[0021] 1.2) Individual force sensor measurement: Keep the electrical stimulation pulse output continuously, the trainee opens his mouth and bites the connecting rod of the pressure sensor, and ensures that the distance between the tongue and the pressure sensor remains constant. Use the maximum force to press the force sensor in front of the tongue, collect the pressure value, and process it to obtain the maximum value of the tongue extension force F. MAX ;

[0022] 1.3) Personal electromyographic signal measurement: turn off the electrical stimulation pulse signal, wear the electromyographic signal sensor, i.e., the electrode ball, press the force sensor in front of the tongue with maximum force, collect the pressure value, and process it using the method in step 1.2) to obtain the personal electromyographic value E MAX ;

[0023] 1.4) Save the electrical stimulation amplitude U and the maximum tongue extension force F MAX And the electromyographic value E MAXUsed for subsequent training parameter setting and training effect evaluation.

[0024] The specific processing of the pressure value is as follows:

[0025] The force sensor is used to continuously collect data for 5 times, each time for the same time and interval, and the maximum value of pressure collected for each action is recorded as F t1 ,F t2 ,F t3 ,F t4 ,F t5 , sort the pressure values ​​from small to large, and subtract the two adjacent items in turn to get the pressure difference D F1 ,D F2 ,D F3 ,D F4 , the two adjacent pressure difference values ​​are added in sequence, and the pressure value corresponding to the minimum value of the addition result is recorded as the data F with the minimum pressure difference among the three pressure values ​​in the five test data. t ′ 1 , F t ′ 2 , F t ′ 3 The average of the three pressure values ​​is taken as the maximum tongue extension force of the trainee. If the deviation condition is met The maximum value of tongue extension force is F MAX If the above conditions are not met, check the calibration and test again until the deviation conditions are met.

[0026] The step 2) comprises the following steps:

[0027] 2.1) The electrical stimulation output module applies electrical stimulation according to the calibrated electrical stimulation amplitude U;

[0028] 2.2) Perform upper respiratory tract muscle training according to the guidance of the interactive display terminal;

[0029] 2.3) Evaluate the training results after completing a single training exercise and after completing all training exercises.

[0030] The step 2.2) comprises the following steps:

[0031] 2.2.1) Pre-set the training pressure window, pressure window duration and rest time, and set the pressure window range size to [F MAX *30%, F MAX *50%];

[0032] 2.2.2) After a single training action is completed, fatigue judgment is performed, and a duration of 5 seconds is set, and the range is [FMAX *70%, F MAX *100%] pressure window, and score the completion. If the completion meets the threshold, continue training for the next action. If the completion is less than the threshold for two consecutive times, it is considered that the trainee has entered a fatigue state and is prompted to terminate the training.

[0033] The step 2.3) comprises the following steps:

[0034] 2.3.1) After a single training action is completed, the completion score is scored. Within 0.5s before and after the pressure window, the completion score λ of the i-th tongue extension pressure action is calculated. i ,

[0035]

[0036] in, is the pressure curve of the i-th tongue-out pressure action, k is the k-th point in the i-th tongue-out pressure action curve, G is the judgment function, if If the pressure is within the set pressure window, 1 point will be recorded, otherwise 0 point will be recorded;

[0037] 2.3.2) After all training movements are completed, analyze and evaluate the training results and calculate the overall score SC of this training result:

[0038]

[0039] Among them, there are m training actions in total, and the score is recorded as The actual continuous training time is recorded as T last , the set continuous training duration is T 0 .

[0040] The present invention has the following beneficial effects and advantages:

[0041] 1. Based on bioelectric feedback, the present invention creates a new treatment model of individualized electrical stimulation combined with upper airway muscle training, formulates precise electrical stimulation and training intensity, avoids fatigue and injury, and maximizes and lasts longer the therapeutic effect.

[0042] 2. The training model proposed in the present invention takes into account both strength training and functional training of the upper airway muscles, which can more effectively improve the stability of the upper airway and induce changes in the neuroplasticity of the genioglossus muscle, while improving the fatigue resistance of the upper airway muscles.

[0043] 3. The present invention adopts submental electrical stimulation combined with upper airway muscle training. Submental electrical stimulation enables the trainee to accurately sense the position of the submental upper airway muscle group during training, thereby strengthening the coordination for force training. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1The structure of the training system of the present invention is shown in FIG. Figure 1 ;

[0045] Figure 2 The structure of the training system of the present invention is shown in FIG. Figure 2 ;

[0046] Figure 3 The sensor installation and physical diagram of the training system of the present invention;

[0047] Figure 4 It is a schematic diagram of the process of the personalized calibration method before training of the present invention;

[0048] Figure 5 This is a schematic diagram of the submental electrical stimulation combined with upper airway muscle training process of the present invention. DETAILED DESCRIPTION

[0049] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0050] like Figure 1 and Figure 2 As shown, it is a schematic diagram of the structure of the measurement training system and the sensor installation of the present invention, including: a control module 1, a signal conditioning module 2, a pressure detection module 3, an electromyography acquisition module 4, an electrical stimulation output module 5, a display interactive terminal 6, a high-precision S-type tension and pressure sensor 7, a customized electrode ball 8, and a disposable medical electrode sheet 9, wherein:

[0051] The control module 1 is used to control the transmission and reception of input and output signals during the entire training process, such as collecting pressure signals and electromyographic signals to understand the current training status; outputting electrical stimulation signals to assist in training. At the same time, it interacts with the display interactive terminal 6 to save and input the parameters set at the beginning of the training into the control system to control the training mode and training sequence, etc.

[0052] The signal conditioning module 2 is used for circuit conversion between digital signals and analog signals. For example, the pressure signal and electromyographic signal generated by the pressure detection module 3 and the electromyographic acquisition module 4 are amplified and filtered by the signal conditioning module 2, and then sampled to form a digital signal for the control module 1; the control module 1 sends a square wave pulse digital signal containing a certain frequency, which is shaped, amplified and processed by the signal conditioning module 2, and then sent to the electrical stimulation output module 5.

[0053] The pressure detection module 3 is used to detect the pressure value during the training of the upper respiratory tract muscles. The implementation method is to place a high-precision S-shaped tension pressure sensor 7 in front of the trainer's face. The tension pressure sensor is made of S-shaped stainless steel, with a pressure sensing range of 0-20N, an accuracy of 0.5%, and a sampling rate of 20Hz. The measurement principle is that the pressure applied by the tongue on the effective area of ​​the sensor increases, the sensor output resistance decreases, and the resistance value change is converted into a voltage value change through the circuit in the signal conditioning module 2. The tension pressure sensor is sealed with a 1mm edible silicone seal for waterproofing. Before use, the relationship between the sensor and the voltage change is calibrated using a calibration experiment.

[0054] The electromyography acquisition module 4 is used to detect the electromyography value during the training of the upper respiratory tract muscles. The implementation method is to place a customized electrode ball 8 under the tongue of the trainee. The customized electrode ball is made of food-grade pure silver, and contains positive and negative electrodes led out by silver wire. The silver wire is wrapped with a food-grade film to prevent the silver wire from being too thin to scratch the lips or tongue. The measurement principle is that when the upper respiratory tract muscles move, the muscle contraction generates a potential difference. The electrical signal generated by the contraction can be captured by the customized electrode ball and transmitted to the control module 1 through the signal conditioning module 2 for analysis and measurement.

[0055] The electrical stimulation output module 5 is used to stimulate the nerve excitement and muscle contraction of the upper respiratory tract muscle group of the trainee. The implementation method is to attach a disposable medical electrode sheet 9 under the chin of the trainee and output an electrical stimulation signal according to pre-set parameters, such as a bidirectional symmetrical square wave with a frequency of 20KHz, and the amplitude can be adjusted within the range of -90V to 90V according to the tolerance of the trainee. The working principle is to cause nerve excitement and tongue root muscle contraction in the corresponding part through transcutaneous electrical stimulation of the electrode sheet attached to the trainee's chin, which can help the trainee find the force-generating part more accurately.

[0056] The display interactive terminal 6 is used for the trainer to interact with the muscle function training device. On the one hand, key parameters such as electrical stimulation frequency, amplitude, upper respiratory tract muscle group strength threshold, etc. are set before training; on the other hand, the current training rhythm and training force value change information are displayed in real time in the form of image animation during training, and evaluation is given during and after the training. The specific implementation method is to use a Raspberry Pi with a touch display function as a terminal. Each trainer enters personal user information and calls in personal training parameters before training. After the training, the personal training information is stored in the database for the convenience of trainers and trainers to query.

[0057] like Figure 3 As shown, the physical schematic diagram of the training system of the present invention includes a control system host 10, an adjustable force sensor bracket 11, a mandibular bracket 12 for fixing the head, and a system base 13, wherein:

[0058] The control system host 10 is installed on a bracket with adjustable up and down and pitch angles, which is convenient for the trainee to operate and observe during the training process. The host includes a control circuit, a signal conditioning circuit, an interactive display screen, and multiple sensor plug interfaces.

[0059] The adjustable force sensor bracket 11 is located between the control system host 10 and the mandibular bracket 12, and has the functions of being adjustable up and down and front and back.

[0060] The mandibular support 12 for fixing the head is used to fix the posture during training. The trainee places the chin on the mandibular support tray and the forehead is placed on the mandibular support to fix the trainee's head.

[0061] The system base 13 is a movable base plate, which is used to fix the control system host 10 , the adjustable force sensor bracket 11 and the mandibular bracket 12 .

[0062] like Figure 4 As shown, it is a schematic diagram of the process of the personalized calibration method before training of the present invention.

[0063] Step 1. Since each trainee has his or her own training history data and training parameters, personal information needs to be entered and confirmed before each calibration and training. If you are a new user, you need to enter more detailed personal health information according to the prompts, such as age, height, weight, sleep AHI index, etc., and save and create a personal profile.

[0064] Step 2, personal measurement of electrical stimulation signal. A disposable medical electrode sheet 9 is pasted under the chin of the trainee. First, an electrical stimulation signal of a certain frequency and a small amplitude is given. In this example, a bidirectional symmetrical square wave with a frequency of 20KHz is used. The amplitude adjustment range is -100V to +100V. The training intensity is changed by adjusting the amplitude. When the trainee feels the electrical stimulation, the intensity is adjusted to a level where there is an obvious pulling sensation at the root of the tongue or difficulty in swallowing. The amplitude U is recorded as a suitable electrical stimulation signal for the trainee.

[0065] Step 3: Individual force sensor test. After the electrical stimulation signal is adjusted, the trainee should open his mouth and bite the connecting rod of the pressure sensor according to the trainer's instructions while the electrical stimulation pulse is continuously output. The distance between the tongue and the pressure sensor should be kept constant. The trainee should use the maximum force to press the force sensor in front of him. This should be repeated 5 times. Each action lasts for 3 seconds with an interval of 30 seconds. The maximum pressure is recorded as F. t1 ,F t2 ,F t3 ,F t4 ,F t5 Sort the pressure values ​​from small to large, and subtract the two adjacent items in turn to get the pressure difference D F1 ,D F2 ,D F3,D F4 , the two adjacent pressure difference values ​​are added in sequence, and the pressure value corresponding to the minimum value is recorded as the data F with the smallest pressure difference among the three pressure values ​​in the five test data. t ′ 1 , F t ′ 2 , F t ′ 3 , take the average of these three pressure values ​​as the trainee's maximum pressure If the deviation condition is met The trainer's maximum pressure F MAX Valid. If the above conditions are not met, it means that if the 5 pressure values ​​fluctuate greatly, the trainee's movements are not standard or there is a problem with the sensor. Check the calibration and try again.

[0066] Step 4: Personal determination of electromyographic signal. After the personal test of the force sensor is completed, turn off the electrical stimulation signal, rest for more than 5 minutes, wear the electromyographic signal sensor, and press the force sensor in front of the tongue with maximum force according to the method in step 3. Do it 5 times in a row, each action lasts for 3 seconds, with an interval of 30 seconds. The peak value of the electromyographic signal is recorded as E t1 ,E t2 ,E t3 ,E t4 ,E t5 The method in step 3 is used to obtain the individual electromyographic signal E MAX .

[0067] Step 5: After the personal parameter test is completed, the system displays the current suitable electrical stimulation amplitude U and the maximum tongue extension force F for the trainee. MAX , EMG value E MAX , the trainer confirms and saves it for subsequent training parameter setting and training effect evaluation.

[0068] like Figure 5 As shown, this is a schematic diagram of the submental electrical stimulation combined with upper airway muscle training process of the present invention.

[0069] Step 1: Log in before training and confirm your personal information and training parameters. If you are a new user, you need to enter more detailed personal information according to the prompts. After completing the personalized calibration before training, save and create a personal profile. At the same time, confirm that the electrical stimulation sensor and force sensor are installed in place.

[0070] Step 2: The electrical stimulation signal is output according to the personalized calibrated electrical stimulation amplitude U, and the intensity is adjusted to a level where there is an obvious pulling sensation at the root of the tongue or difficulty in swallowing. If the trainee feels that the stimulation is too large or too small, the sensor installation position can be checked or the parameters can be adjusted.

[0071] Step 3. Click the Start Training button on the human-computer interaction interface to start upper respiratory muscle training. The training is achieved through an interesting game. The interface guides the trainee to extend the tongue and press the front sensor. The pressure size and pressure interval are set as follows:

[0072] ①Preset the pressure window, and the trainee should try to control the tongue pressure within this pressure window. The range of the pressure window is [F MAX *30%, F MAX *50%], in order to allow trainees to concentrate, through interesting games, the game sets the duration of the pressure window to a dynamic value, ranging from 3 to 5 seconds; the rest time is a dynamic value, ranging from 10 to 20 seconds;

[0073] ② Fatigue judgment: The training time can be set before training. Generally, the setting time is 15-30 minutes. Every 1 minute, a range of [F MAX *70%, F MAX *100%] pressure window, which lasts for 5 seconds. If the trainee's completion score meets 80 points (see step 4 for scoring details), the trainee is considered to be not in a fatigue state and can continue training. If the completion score is less than 80 points for two consecutive times, the trainee is considered to be in a fatigue state and is prompted to terminate the training.

[0074] Step 4: After a single training action is completed, the host evaluates the training result using the degree of completion. Within 0.5s before and after the pressure window, is the pressure curve of the i-th tongue-extending pressure action, k is the k-th point in the i-th tongue-extending pressure action curve, λ i is the completion score of the i-th tongue-extending pressure action, and the scoring rule is:

[0075]

[0076] Among them, G is the judgment function, if If the pressure is within the set pressure window, 1 point will be recorded; otherwise, 0 point will be recorded.

[0077] Step 5: After all training movements are completed, analyze and evaluate the training results.

[0078] After the training, there are a total of m training actions, and the score is recorded as The actual continuous training time is recorded as T last , the set continuous training duration is T 0 , the overall score of this training result is recorded as SC

[0079]

[0080] By collecting the electromyographic and pressure values ​​during the training process, the training effect is predicted, specifically:

[0081] Step 1: Integrate and normalize the collected pressure data, efficacy data, and electromyographic activity data during the training process. Segment the input data by treatment course to create labels for predicting the efficacy of different treatment courses. The polysomnography results are used as the treatment status label of the current treatment course, and the expert's adjustment of parameters during the treatment process is used as the feedback label to obtain the efficacy prediction model data set.

[0082] Step 2: Use a multimodal feature extractor based on residual convolutional neural network and long short-term memory network to extract data set features.

[0083] A unimodal feature extractor is used to fuse an adaptive residual convolutional network and a multi-head long short-term memory network to extract local features and global dependencies of the data. The multimodal feature extractor constructed includes three channels, each of which is composed of a unimodal feature extractor and an adaptive attention mechanism. The unimodal feature extractor fuses an adaptive residual convolutional network and a multi-head long short-term memory network to extract local features and global dependencies of the data, and outputs the output through an adaptive attention mechanism, and calculates the attention score through a multi-layer perceptron, thereby enhancing the model's ability to express multimodal input data;

[0084] After extracting the features, a graph neural network is used to construct a modal feature graph for information propagation and fusion. The extracted feature information of different modalities is used as input, and the feature information of different modalities is fused through the topological structure of the graph. The position information attention mechanism is introduced to capture the relevance and importance of different modal features, aggregate the feature information of different modalities, capture direction and position sensitive information, and enhance the representation ability of the model.

[0085] By setting different loss functions, calculating the losses of the three channels and the graph neural network fusion loss, the model is trained, and the trained model is used to predict the training effect.

Claims

1. A submental electrical stimulation combined with upper airway muscle training system based on bioelectric feedback, characterized in that: include: Pressure detection module, used to detect the pressure value during the upper respiratory tract muscle training process; The electromyography acquisition module is used to detect the electromyography values ​​during the training of the upper respiratory tract muscles; The electrical stimulation output module is used to stimulate the nerve excitement and muscle contraction of the trainee's upper respiratory tract muscles; Signal conditioning module, used for preprocessing and analog-to-digital conversion of signals transmitted between modules; Control module, used to control the signal interaction during the entire training process; Display interactive terminal for interaction with trainers.

2. A submental electrical stimulation combined with upper airway muscle training system based on bioelectric feedback according to claim 1, characterized in that: The pressure detection module is an S-shaped tension pressure sensor, which is placed in front of the trainee's face, collects the pressure applied by the tongue on the effective area of ​​the sensor, and converts the resistance value change into the voltage value change through the signal conditioning module.

3. The submental electrical stimulation combined with upper airway muscle training system based on bioelectric feedback according to claim 1, characterized in that: The myoelectric acquisition module is an electrode ball placed under the trainee's tongue. The electrode ball contains positive and negative poles connected to the signal conditioning module through silver wires to collect the potential difference signal generated by muscle contraction when the upper respiratory tract muscles move.

4. The submental electrical stimulation combined with upper airway muscle training system based on bioelectric feedback according to claim 1, characterized in that: The electrical stimulation output module is a disposable medical electrode sheet, which is attached to the trainee's chin. The square wave pulses emitted by the control module generate electrical stimulation, causing nerve excitement in the corresponding part and contraction of the tongue root muscles.

5. A submental electrical stimulation combined with upper airway muscle training method based on bioelectric feedback, characterized in that: The following steps are involved: 1) Before training, personalize the basic information of the trainee; 2) Conduct submental electrical stimulation combined with upper airway muscle training on the trainees; 3) Predict the training effect by collecting the electromyography and pressure values ​​during the training process.

6. A method for submental electrical stimulation combined with upper airway muscle training based on bioelectric feedback according to claim 5, characterized in that: The step 1) comprises the following steps: 1.1) Individual measurement of electrical stimulation signal: A disposable medical electrode is attached to the trainee's chin, and an electrical stimulation signal of a certain frequency and amplitude is given. The amplitude of the signal is gradually adjusted. When the trainee feels a noticeable pulling sensation at the root of the tongue or difficulty in swallowing, the amplitude U at this time is recorded as the appropriate electrical stimulation signal for the trainee. 1.2) Individual force sensor measurement: Keep the electrical stimulation pulse output continuously, the trainee opens his mouth and bites the connecting rod of the pressure sensor, and ensures that the distance between the tongue and the pressure sensor remains constant. Use the maximum force to press the force sensor in front of the tongue, collect the pressure value, and process it to obtain the maximum value of the tongue extension force F. MAX ; 1.3) Personal electromyographic signal measurement: turn off the electrical stimulation pulse signal, wear the electromyographic signal sensor, i.e., the electrode ball, press the force sensor in front of the tongue with maximum force, collect the pressure value, and process it using the method in step 1.2) to obtain the personal electromyographic value E MAX ; 1.4) Save the electrical stimulation amplitude U and the maximum tongue extension force F MAX And the electromyographic value E MAX Used for subsequent training parameter setting and training effect evaluation.

7. A method for submental electrical stimulation combined with upper airway muscle training based on bioelectric feedback according to claim 6, characterized in that: The specific processing of the pressure value is as follows: The force sensor is used to continuously collect data for 5 times, each time for the same time and interval, and the maximum value of pressure collected for each action is recorded as F t1 ,F t2 ,F t3 ,F t4 ,F t5 , sort the pressure values ​​from small to large, and subtract the two adjacent items in turn to get the pressure difference D F1 ,D 22 ,D F3, D F4 , the two adjacent pressure difference values ​​are added in sequence, and the pressure value corresponding to the minimum value of the addition result is recorded as the data F with the minimum pressure difference among the three pressure values ​​in the five test data. t ′ 1, F t ′ 2, F t ′ 3. Take the average of the three pressure values ​​as the maximum tongue extension force of the trainee. If the deviation condition is met The maximum value of tongue extension force is F MAX If the above conditions are not met, check the calibration and test again until the deviation conditions are met.

8. The method of submental electrical stimulation combined with upper airway muscle group training based on bioelectric feedback according to claim 5, characterized in that: The step 2) comprises the following steps: 2.1) The electrical stimulation output module applies electrical stimulation according to the calibrated electrical stimulation amplitude U; 2.2) Perform upper respiratory tract muscle training according to the guidance of the interactive display terminal; 2.3) Evaluate the training results after completing a single training exercise and after completing all training exercises.

9. The method of submental electrical stimulation combined with upper airway muscle training based on bioelectric feedback according to claim 8, characterized in that: Step 2.2) The following steps are involved: 2.2.1) Pre-set the training pressure window, pressure window duration and rest time, and set the pressure window range size to [F MAX *30%, F MAX *50%]; 2.2.2) After a single training action is completed, fatigue judgment is performed, and a duration of 5 seconds is set, and the range is [F MAX *70%, F MAX *100%] pressure window, and score the completion. If the completion meets the threshold, continue training for the next action. If the completion is less than the threshold for two consecutive times, it is considered that the trainee has entered a fatigue state and is prompted to terminate the training.

10. The method of submental electrical stimulation combined with upper airway muscle training based on bioelectric feedback according to claim 8, characterized in that: The step 2.3) comprises the following steps: 2.3.1) After a single training action is completed, the completion score is scored. Within 0.5s before and after the pressure window, the completion score λ of the i-th tongue extension pressure action is calculated. i , in, is the pressure curve of the i-th tongue-out pressure action, k is the k-th point in the i-th tongue-out pressure action curve, G is the judgment function, if If the pressure is within the set pressure window, 1 point will be recorded, otherwise 0 point will be recorded; 2.3.2) After all training movements are completed, analyze and evaluate the training results and calculate the overall score SC of this training result: Among them, there are m training actions in total, and the score is recorded as The actual continuous training time is recorded as T last , the set continuous training duration is T 0 .

Images