A method for electrical stimulation tactile perception

Abstract

The present invention discloses a method for tactile perception through electrical stimulation; the present invention uses a high-density electrode array composed of functional electrical stimulation electrodes as a single electrical stimulation module, and realizes motor rehabilitation of fine hand movements and tactile feedback of different contact points of the hand through the combination of electrical stimulation electrodes. The method of the present invention directly stimulates the nerves and muscles related to hand movements and touch. Both motor rehabilitation and tactile feedback are direct effects of electrical stimulation, which conforms to human's natural intuition and greatly reduces the training cost of users. The present invention improves the spatial resolution of the electrical stimulation system by using a high-density electrode array, which greatly compensates for the shortcomings of the existing electrical stimulation system, such as insufficient spatial resolution and insufficiently fine stimulation function. The present invention can quickly and accurately search for electrode combinations corresponding to different fine movements and different fine position touches.

Description

Technical Field

[0001] The present invention relates to an electrical stimulation tactile perception method, belonging to the technical field of artificial intelligence medical equipment. Background Art

[0002] The functional electrical stimulation system is a technology used for motor rehabilitation and tactile feedback induction [1]. By artificially applying electric current to specific nerve sites, it can drive patients with movement disorders to voluntarily contract the corresponding muscles and produce limb movements. At the same time, by electrically stimulating the patient's sensory nerves, the patient's corresponding hand tactile perception is artificially induced, thereby helping the patient gradually recover the tactile feedback function. Through the motor function rehabilitation of "patient-electrical stimulation-exercise" and the tactile feedback of "exercise-electrical stimulation-patient", efficient two-way rehabilitation training can be achieved.

[0003] Most existing electrical stimulation systems only achieve one of the above-mentioned motor rehabilitation and tactile feedback. The system that can currently achieve bidirectional rehabilitation training is the lower limb gait rehabilitation system proposed by Aurelie Selfslagh et al. in 2019 [2]. The system uses 16 electrode channels to stimulate the target muscles of the left and right legs of the lower limbs: gluteus maximus, gluteus medius, proximal rectus femoris, hamstrings, vastus lateralis, tibialis anterior, gastrocnemius, and soleus (1 electrode channel for each of the 8 muscles in the left and right legs, a total of 16 channels) to enable the muscles of both legs to produce target gait movements. To further provide tactile feedback to patients, the system further uses a tactile feedback system called "tactile shirt" [3] to provide tactile feedback from the direction of movement to the patient. The tactile feedback system is embedded in the sleeves of the shirt, and each sleeve contains three small coin-shaped vibrators arranged in sequence along the distal-proximal axis of the forearm. The system uses the concepts of "sensory substitution" and "sensory remapping" to provide tactile feedback to the lower limbs by generating vibration information from the forearm. “Sensory substitution” refers to using a different modality to transmit sensory information from another modality. For example, Bach-y-Rita et al. have achieved the use of touch to provide visual information [4]. “Sensory remapping” refers to providing sensory information from one part of the body to another different part of the body. Through this “tactile sleeve” module, the system can achieve two-way lower limb rehabilitation assistance.

[0004] Existing sports rehabilitation assistance systems have a common defect, that is, they can only enable patients with movement disorders to achieve relatively rough motor functions (such as movements of the lower limbs, shoulders or elbows) through electrical stimulation, but the movements of the hands are more complex than those of the above parts, and the distribution of related muscle parts in the anatomical structure is close to each other, and the functions are also more complex. At present, rehabilitation training for fine hand movements is still a difficulty in related fields. This defect is mainly limited by the limited spatial resolution of existing electrical stimulation systems, which makes it only possible to stimulate large muscle groups that are easy to locate to produce relatively fixed and rough movements in specific joints.

[0005] Existing rehabilitation assistance systems usually implement motor rehabilitation and tactile feedback through two separate modules.

[0006] Existing tactile feedback induction systems based on functional electrical stimulation mostly focus on using electrical stimulation to produce different types of tactile feedback (such as the hardness of the contact surface material and the magnitude of the contact force) at a single contact point. However, for human hand tactile perception, accurately identifying the tactile sensations generated by different hand positions is one of the key factors for the dexterity of human hands in daily life. Therefore, how to induce a tactile feedback system that can accurately identify different hand contact points through functional electrical stimulation systems remains a research challenge.

[0007] At the same time, improving the spatial resolution of the functional electrical stimulation system requires more precise and accurate quantification of the stimulation points and possible combinations of points. This depends on the anatomical structure of the arm muscles, which varies from person to person. Existing systems and research cannot solve the problem of quickly and accurately searching for the optimal electrode array that can produce different hand movements and tactile positions for each individual. Summary of the Invention

[0008] The present invention aims to improve the spatial resolution of the electrical stimulation system by using a (high-density) electrode array as a single electrical stimulation module covering the forearm, and to achieve motor rehabilitation of fine hand movements and tactile feedback at different contact points of the hand through a precise combination of electrical stimulation electrodes. The present invention uses the same surface electrical stimulation to simultaneously achieve motor rehabilitation in the "patient-electrical stimulation-motor" direction and tactile feedback in the "motor-electrical stimulation-patient" direction, which can greatly reduce the complexity of the system and improve the system's configuration and wearability.

[0009] The technical solution of the present invention is specifically described as follows.

[0010] The present invention provides an electrical stimulation tactile perception method, which uses an electrode array composed of functional electrical stimulation electrodes as a single electrical stimulation module, and realizes motor rehabilitation of fine hand movements and tactile feedback at different contact points of the hand through the combination of electrical stimulation electrodes.

[0011] Preferably, the electrode array adopts a high-density electrode array; preferably, the electrode spacing in the high-density electrode array is 0.8-1.2 cm, and the coverage area is 300 cm 2 When the number of electrodes is between 280 and 320.

[0012] In the present invention, an electrode array is placed over the forearm area, and motor rehabilitation of fine hand movements is achieved through the combination of electrical stimulation electrodes. The specific method is as follows:

[0013] Step 1: Search for electrical stimulation electrode combinations that produce different types of hand fine movements using the optimal electrode array rapid search method;

[0014] Step 2: Based on the hand movement type generated by the electrode electrical stimulation, the patient independently performs corresponding flexion and extension movements, and the muscle electrical signals generated are detected. If the patient's movement disorder is more severe, the method of mirror movement of both hands is used to detect the electromyography of the healthy side, and then deep learning and machine learning algorithms are used to identify different movements;

[0015] Step 3: Using the patient's subjective movement intention as input, identify the corresponding rehabilitation movement that generates myoelectricity. After receiving the identification signal, activate the corresponding electrode combination. The electrical stimulation electrode array stimulates the designated muscle group, giving the patient appropriate assistance. Gradually increase the current intensity to increase the range of motion until the required standard of rehabilitation training is achieved and the patient's intended movement is completed; wherein:

[0016] In step 1, the optimal electrode array rapid search method uses a multi-round automatic search strategy to quickly find the electrode combination that induces different hand fine movement patterns to achieve motor rehabilitation. The specific steps are as follows:

[0017] a) In the first round of search, a single electrode is selected and connected to the reference electrode. The current gradually increases from the initial value of 15mA. The test of this electrode is completed when the following two conditions are met:

[0018] 1) The bending angle of a certain joint reaches the required value for rehabilitation training;

[0019] 2) The user experiences pain;

[0020] The current at this point is recorded as the maximum current allowed for that electrode. When a certain electrode stimulates a movement, the movement is classified using an unsupervised learning classification method. If the movement cannot be classified into an existing similar category, it is considered a new category. The first round of search requires traversing the test conditions of all single electrodes and the ground electrode. In addition, if the user feels pain when a certain electrode is connected to the initial 15mA, this electrode is eliminated.

[0021] b) In the second round of search, a current is connected between two electrodes to search for hand movements that can be produced. The current is increased from the initial value of 15mA. When the following two conditions are met, the test of this electrode combination is completed:

[0022] 1) The bending angle of a certain joint reaches the required value for rehabilitation training;

[0023] 2) The current amplitude reaches the maximum current allowed by a certain electrode recorded in step (a);

[0024] Whenever a two-electrode combination stimulates a motor action, an unsupervised learning classification method is used to classify the action. If the action cannot be classified into an existing similar category, a new action category is created for the action.

[0025] The second round of search testing ends when the following two conditions are met:

[0026] 1) The different fine hand movements stimulated have met the needs of rehabilitation training;

[0027] 2) All possible combinations of two electrodes have been traversed.

[0028] In the present invention, in step b), the two-dimensional hotspot distribution map generated by the finger muscles during a single finger movement obtained by quantifying the regional neural function is used as prior knowledge. First, two electrode combinations near the hotspot area are searched, and then the electrode combinations after the hotspot area are searched, so as to search for as many different hand movement actions as possible in the shortest time.

[0029] In the present invention, if the second round of search has traversed all two-electrode combinations and cannot meet the hand fine movement rehabilitation training requirements, a third round of search is performed based on the three-electrode combination.

[0030] The present invention also includes a fourth round of search, which combines the results obtained from the second and third rounds of search in pairs to produce more fine hand movements, or adds a certain delay between two combinations to observe the resulting hand movement results.

[0031] In the present invention, during exercise rehabilitation, electrical stimulation uses a 10kHz high-frequency narrow pulse width current or a sine wave current.

[0032] In the present invention, an electrode array is placed below the biceps and near the elbow of the upper arm to cover the median nerve, ulnar nerve and radial nerve, and tactile feedback is achieved at different contact points of the hand through the combination of electrical stimulation electrodes. The specific method is as follows: Step 1, the hand is divided into 11 areas: thumb, index finger, middle finger, ring finger, pinky finger, thenar eminence, hypothenar eminence + palm, index finger mound, middle finger mound, ring finger mound, pinky finger mound. If the patient can feel the touch in a certain area, it is considered that the area has received effective rehabilitation training;

[0033] Step 2: After connecting the current between the single electrode and the reference electrode, ask the user about their feelings. The current gradually increases from the initial value of 2mA. The electrode test is completed when the following three conditions are met:

[0034] 1) The user experiences pain;

[0035] 2) If the current value is too large, in addition to tactile perception, it will also stimulate the user's hand movement;

[0036] The hand area where tactile perception can be obtained and the corresponding single electrode are recorded. If the tactile areas produced are the same, the electrode that the patient feels more comfortable with is selected. If the user feels pain when an electrode is connected to an initial current of 2mA, this electrode is eliminated in subsequent searches.

[0037] Step 3: After connecting the current between the two electrodes, ask the user how they feel. The current increases from the initial value of 2mA. If the following two conditions are met, the test of this electrode combination is completed:

[0038] 1) The user experiences pain;

[0039] 2) In addition to tactile perception, stimulate the user's hand movements

[0040] Also record the hand area where tactile perception can be obtained and the corresponding two electrodes. If the tactile area produced is the same as a single electrode or other two-electrode combination, select the electrode that the patient feels more comfortable with.

[0041] The dual-electrode search is completed when the following two conditions are met:

[0042] 1) The evoked tactile perception area covers all fingers and most of the palm;

[0043] 2) All possible combinations of two electrodes have been traversed.

[0044] In the present invention, the tactile feedback training for patients includes the following two aspects:

[0045] First, using currents of varying amplitude as input, the patient's ability to perceive changes in the size of touch sensations in the hand is trained;

[0046] Secondly, different types of electric currents are used as input to see if patients can feel other detailed tactile sensations in addition to the size of the touch.

[0047] The present invention can apply precise current stimulation to specific neuromuscular sites in the forearms of patients with movement and tactile feedback disorders, causing the patient's arm muscles to contract autonomously, thereby driving the hand joints to achieve fine motor movements and enabling patients with tactile feedback disorders to produce tactile feedback. Compared with the existing technology, the present invention has the following beneficial effects:

[0048] 1. The present invention uses an electrode array as a single electrical stimulation module, which can simultaneously realize the bidirectional rehabilitation training functions of movement rehabilitation in the direction of "patient-electrical stimulation-movement" and tactile feedback in the direction of "movement-electrical stimulation-patient", thereby improving convenience.

[0049] 2. The present invention directly stimulates the nerves and muscles related to hand movement and touch. Movement rehabilitation and tactile feedback are both direct effects of electrical stimulation, which conforms to human natural intuition and greatly reduces the training cost of users.

[0050] 3. The present invention improves the spatial resolution of the electrical stimulation system by using a high-density electrode array, which greatly compensates for the shortcomings of the existing electrical stimulation system, such as insufficient spatial resolution and insufficiently refined stimulation function.

[0051] 4. This invention addresses the extremely high number of electrode combinations in high-density electrode arrays and, based on each subject's unique muscle anatomy, can quickly and accurately search for electrode combinations corresponding to different fine motor movements and fine-tactile sensations. Furthermore, the prior knowledge provided by regional neural function quantification further shortens the time required for electrode array search. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 : Flowchart of automatic intelligent search using a high-density electrical stimulation array during motor rehabilitation training.

[0053] Figure 2 : In the study, the corresponding two-dimensional spatial hot spot map and nerve innervation area of the forearm muscles when a single finger performs stretching exercises.

[0054] Figure 3 : Flowchart of collaborative work between subjective intention and flexible electrical stimulation robot (human-machine integration).

[0055] Figure 4 Diagram of the sensory nerves in the upper arm and the proposed locations for two high-density electrical stimulation arrays. Each red circle represents an electrical stimulation electrode. Note: The red squares represent the reference electrode locations for both arrays. This reference electrode is located on the back of this view and serves only as a reference electrode in the single-electrode configuration.

[0056] Figure 5 : Flowchart of intelligent search using a high-density electrical stimulation array during tactile perception functional rehabilitation training.

[0057] Figure 6 Figure 2 shows the areas of the hand that experienced tactile sensation when two electrodes of a high-density electrical stimulation array beneath the biceps muscle were activated in a pre-experimental study. For example, after activating the red electrode groups (row 3, column 3 and row 2, column 7), the red diagonal area of the hand experienced tactile feedback.

[0058] Figure 7 : Test diagram of the electrical stimulation tactile perception system in the embodiment. DETAILED DESCRIPTION

[0059] The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0060] (1) Implementation of functional electrical stimulation for fine motor rehabilitation of the hands:

[0061] In order to solve the problem that the existing functional electrical stimulation motor rehabilitation system lacks the rehabilitation of fine motor skills of the hand, the system proposed in this patent uses a high-density electrode array (with an electrode spacing of about 1 cm, the number of electrodes required to achieve full coverage of the forearm is about 300, and the coverage area is 300 cm). 2 ; Arranged in 16 rows (circling the forearm) and 20 columns (covering from the wrist to the elbow); Using a small number of electrodes to cover a part of the muscle group can also achieve partial finger rehabilitation function, so it also falls within the scope of protection of this patent) to improve spatial resolution and achieve fine motor rehabilitation of the hand. The finger extensors and flexors are extremely important in the fine movements of the hand. The anatomical structure of the finger extensors and flexors is extremely complex. They are composed of multiple smaller muscle groups, and the movement of each finger is driven by different muscle groups of the finger muscles. We use high-density electrode arrays to cover all muscle groups that control hand movements, and accurately stimulate different muscle groups to produce different fine motor movements of the hand, including extension and flexion of a single finger and wrist. At the same time, previous studies have shown that the hotspot position of the forearm electromyographic signal generated by the patient's autonomous hand movement is very close to the position of the electrode required to activate the electrical stimulation to produce this movement. Therefore, the flexible high-density electrical stimulation array placed in this system covers the forearm area.

[0062] The core difficulty of using electrical stimulation to achieve hand function and movement rehabilitation is how to quickly search for electrode combinations that induce different hand fine movements for different patients. Due to individual differences in the anatomical structure of the hand and forearm, the movements produced by the electrical stimulation site will also be different. In addition, different individuals and different parts of the body are also sensitive to the pain caused by electric current. For this reason, this system uses a forearm sleeve flexible robot that integrates a high-density electrical stimulation array and a glove robot with hand joint movement detection function. It adopts a multi-round automatic search strategy to find as many different hand fine movement patterns as possible. The specific steps are as follows ( Figure 1 ):

[0063] a) In the first round of search, after a single electrode is selected and connected to the reference electrode, the current gradually increases from the initial value of 15mA. When the following two conditions are met, the test of this electrode is completed: 1) the bending angle of a joint reaches the required value for rehabilitation training; or 2) the user feels pain. The current at this time is recorded as the maximum current allowed for the electrode. When a certain electrode stimulates a movement, an unsupervised learning classification method (such as principal component analysis, K-Means algorithm, etc.) is used to classify this action. If this action cannot be classified into an existing similar category, the action is identified as a new category. The first round of search requires traversing the test conditions of all single electrodes and ground electrodes. In addition, if the user feels pain when a certain electrode is connected to the initial 15mA, this electrode is eliminated.

[0064] b) In the second round of search, after connecting the current between the two electrodes, search for the hand movement that can be generated. The current increases from the initial value of 15mA. When the following two conditions are met, the test of this electrode combination is completed: 1) the bending angle of a certain joint reaches the required value for rehabilitation training; or 2) the current amplitude reaches the maximum current allowed by a certain electrode recorded in step (a). In order to increase the search efficiency, the search strategy intends to use a high-density electrode array to collect and quantify the muscle electrical signals to obtain a two-dimensional hot spot distribution map generated by the finger muscles during the movement of a single finger as prior knowledge ( Figure 2 ), first search for the two electrode combinations near the hot spot area, and then search for the electrode combinations after the hot spot area, and search for as many different hand movements as possible in the shortest time. Whenever a two-electrode combination stimulates a movement, similar to the case of a single electrode, the unsupervised learning classification method is used to classify this movement. If this movement cannot be classified into an existing similar category, a new movement category is established for the movement. The second round of search test ends when the following two conditions are met: 1) The different fine hand movements stimulated have met the needs of rehabilitation training; 2) All two-electrode combinations have been traversed.

[0065] c) Third round of search (optional): If all two-electrode combinations in the second round fail to meet the requirements for fine motor rehabilitation training, a three-electrode combination will be added. The search method is the same as for the two-electrode case, except for the addition of one electrode.

[0066] d) In the fourth search, the results from the second and third searches are combined in pairs to generate more fine motor movements of the hand. Furthermore, a delay can be added between combinations to observe the resulting hand movements. For example, in the applicant's research, we found that electrical stimulation of the middle extensor digitorum muscles can induce extension of both the ring and pinky fingers, while stimulation of the medial flexor digitorum muscles near the elbow can induce flexion of the ring finger. Combining these two methods can produce fine motor movements involving extension of only the pinky finger.

[0067] Traditional muscle electrical stimulation typically uses a square wave current with an amplitude of 20-40mA, a pulse width of 300-800μs, and a frequency of 20-50Hz. Recent research indicates that using currents with varying frequencies and pulse widths can alleviate pain and muscle fatigue. Therefore, this system further adjusts the current parameters used for electrical stimulation to achieve a more diverse impact on rehabilitation training. For example, this can be achieved by appropriately increasing or decreasing the wavelength; replacing square waves with sine waves and other waveforms; and using a high-frequency, narrow-pulse current of 10kHz.

[0068] This system can also realize rehabilitation training that combines the patient's subjective intention with the flexible electrical stimulation robot (natural human-machine interaction). Unlike voluntary contraction, the muscle contraction caused by electrical stimulation is easier to activate several muscle groups at the same time because of the conduction of current along the muscle fibers, which can easily cause multiple fingers or fingers and wrists to extend and flex at the same time. Therefore, the electrical stimulation method may not be able to meet all the required fine hand movements, and the patient's subjective intention input should be implemented within the range of hand movements that can be obtained by electrical stimulation. The specific steps to combine the two are as follows ( Figure 3 ):

[0069] a) According to the above electrode search method, search for all categories of fine hand movements that can be produced by high-density electrodes.

[0070] b) According to the hand movement category generated by high-density electrical stimulation, the patient independently performs corresponding flexion and extension movements, and the muscle electrical signals generated are detected (Note: If the patient's movement disorder is more serious, the mirror movement of both hands is used to detect the electromyography of the healthy side), and then deep learning and machine learning algorithms [5] are used to identify different movements.

[0071] c) Using the patient's subjective movement intention as input, the system identifies the corresponding rehabilitation movement that generates myoelectricity. Upon receiving the recognition signal, the functional electrical stimulation system activates the corresponding electrode combination. The high-density electrical stimulation array stimulates the designated muscle group, providing the patient with appropriate assistance. The current intensity is gradually increased to increase the range of motion until the required rehabilitation training standard is achieved, completing the patient's intended movement and enabling real-time human-computer interaction. The same movement can be trained repeatedly to meet clinical rehabilitation training requirements.

[0072] (2) Implementation of functional electrical stimulation of hand tactile sensation:

[0073] Since the patient's sensory neural circuit is still intact, the system stimulates the three main sensory nerves in the patient's hand (median nerve, ulnar nerve and radial nerve) by adding functional electrical stimulation externally, artificially inducing the patient's corresponding tactile perception in the hand, thereby helping the patient gradually recover the tactile function of the hand.

[0074] The sensory nerves of the upper arm start from the brachial plexus and extend to the ends of the fingers. On the way, they run through the wrist, forearm and upper arm. Many parts are very close to the body surface. Figure 4 Since the forearm is mainly used for myoelectric intention recognition and motor function electrical stimulation, in order to avoid mutual interference between multiple electrical signals, this system uses the upper arm flexible robot for hand function tactile perception rehabilitation training.

[0075] The median nerve mainly controls the tactile perception of the thumb, index finger, middle finger and palm, the ulnar nerve mainly controls the tactile perception of the ring finger, little finger and palm, and the radial nerve mainly controls the tactile perception of the dorsal side of the hand. Electrical stimulation tactile perception rehabilitation training needs to cover the entire tactile perception area of the hand as much as possible; at the same time, anatomical studies have shown that the median nerve and radial nerve are closest to the body surface below the biceps in the upper arm, while the ulnar nerve is closest to the body surface near the elbow joint and below the biceps. This system places two sets of high-density electrical stimulation arrays (such as Figure 4 In order to cover the tactile perception area of the entire hand as much as possible, tactile rehabilitation adopts an automatic search strategy similar to fine motor rehabilitation. The specific steps are as follows ( Figure 5 ):

[0076] a) Divide the hand into 11 areas, such as Figure 5 As shown, if the patient can feel the touch in a certain area, it is believed that the area has received effective rehabilitation training.

[0077] b) After connecting the current between the single electrode and the reference electrode, ask the user how they feel. The current gradually increases from the initial value of 2mA. When the following three conditions are met, the test of this electrode is completed: 1) The user feels pain; 2) The current value is too large (usually more than 6mA), which will stimulate the user's hand movement in addition to tactile perception. Record the hand area where tactile perception can be obtained and the corresponding single electrode. If the tactile area generated is the same, select the electrode that the patient feels more comfortable. If the user feels pain when an electrode is connected to the initial value of 2mA current, this electrode will be eliminated in subsequent searches.

[0078] c) After connecting the current between the two electrodes, ask the user how they feel. The current is increased from the initial value of 2mA. When the following two conditions are met, the test of this electrode combination is completed: 1) the user feels pain; 2) in addition to tactile perception, the user's hand movement is stimulated. Similarly, the hand area where tactile perception can be obtained and the corresponding two electrodes are recorded. If the tactile area generated is the same as that of a single electrode or other two-electrode combination, the electrode that the patient feels more comfortable with is selected. The two-electrode search is completed when the following two conditions are met: 1) the induced tactile perception area has covered all fingers and most of the palm; 2) all two-electrode combinations have been traversed.

[0079] This system uses a new method to rehabilitate the patient's hand tactile perception. In previous studies, we used a square wave current with an amplitude of 2-3mA, a pulse width of 150-250μs, and a frequency of 150Hz. Figure 6 Shows the areas of the hand that correspond to tactile perception when certain electrode combinations are activated in previous research experiments.

[0080] In addition, the required electrical stimulation current intensity, frequency, pulse width, waveform, etc. will have an impact on the patient's perception. The specific rehabilitation training content mainly includes the following two aspects: Figure 6 ):

[0081] a) Use current with varying amplitudes as input, such as triangular or trapezoidal currents, to train the patient's ability to perceive changes in the size of hand tactile sensations.

[0082] b) Use different types of current as input, such as different frequencies and waveforms, to see if the trainee can feel other detailed tactile sensations besides the size of the touch, such as pain, movement, etc.

[0083] In this embodiment, a method for rapid electrode array search is proposed to address the extremely high number of electrode combinations in high-density electrode arrays. This method, based on each subject's unique muscle anatomy, quickly and accurately identifies electrode combinations corresponding to different fine motor movements and fine positional tactile sensations (Table 1). Furthermore, the time required for electrode array search is further shortened by leveraging prior knowledge provided by regional neural function quantification.

[0084] Table 1 Table of palm tactile evoked information (24-channel array electrical stimulation)

[0085]

[0086]

[0087] Figure 7 Disclosed is an electrical stimulation tactile perception system, which includes an electrical stimulation instrument, a multiplexer, a command transmitter, and a high-density electrode array. The high-density electrode array is installed on the upper arm and has 3 rows and 8 columns, with 24 channels. References:

[0088] [1]MB Popovic, DBPopovic, L.Schwirtlich, and T. “Functional Electrical Therapy (FET): Clinical Trial in Chronic Hemiplegic Subjects,” Neuromodulation:

[0089] Technology at the Neural Interface,vol.7,no.2,pp.133–140,2004,doi:

[0090] 10.1111 / j.1094-7159.2004.04017.x.

[0091] [2]A.Selfslagh et al.,“Non-invasive,Brain-controlled FunctionalElectrical Stimulation for Locomotion Rehabilitation in Individuals withParaplegia,”Sci Rep,vol.9,no.1,p.6782,

[0092] Dec.2019,doi:10.1038 / s41598-019-43041-9.

[0093] [3]S.Shokur et al.,“Assimilation of virtual legs and perception offloor texture by complete paraplegic patients receiving artificial tactilefeedback,”Sci Rep,vol.6,no.1,p.32293,Sep.

[0094] 2016,doi:10.1038 / srep32293.

[0095] [4]P.Bach-Y-Rita,C.C.Collins,F.A.Saunders,B.White,and L.Scadden,“Vision Substitution by Tactile Image Projection,”Nature,vol.221,no.5184,pp.963–964,Mar.

[0096] 1969,doi:10.1038 / 221963a0.

[0097] [5]X.Jiang et al.,“Open Access Dataset,Toolbox and BenchmarkProcessing Results of High-Density Surface Electromyogram Recordings,”IEEETransactions on Neural Systems and Rehabilitation Engineering,vol.29,pp.1035–1046,2021,doi:

[0098] 10.1109 / TNSRE.2021.3082551。

Claims

1. A method for rapidly searching for an optimal electrode array, characterized in that: The method uses an electrode array composed of functional electrical stimulation electrodes as a single electrical stimulation module and employs a multi-round automatic search strategy to search for electrical stimulation electrode combinations that produce different categories of fine hand movements. The specific steps are as follows: a) In the first round of search, a single electrode is selected and connected to the reference electrode. The current gradually increases from the initial value of 15mA. The test of this electrode is completed when one of the following two conditions is met: 1) The bending angle of a certain joint reaches the required value for training; 2) The user experiences pain; The current at this point is recorded as the maximum current allowed for that electrode. When a certain electrode stimulates a movement, the movement is classified using an unsupervised learning classification method. If the movement cannot be classified into an existing similar category, it is considered a new category. The first round of search requires traversing the test conditions of all single electrodes and the ground electrode. In addition, if the user feels pain when a certain electrode is connected to the initial 15mA, this electrode is eliminated. b) In the second round of search, a current is connected between two electrodes to search for hand movements that can be produced. The current is increased from the initial value of 15mA. When one of the following two conditions is met, the test for this electrode combination is completed: 1) The bending angle of a certain joint reaches the required value for training; 2) The current amplitude reaches the maximum current allowed by a certain electrode recorded in step a); Whenever a two-electrode combination stimulates a motor action, an unsupervised learning classification method is used to classify the action. If the action cannot be classified into an existing similar category, a new action category is created for the action. The second round of search testing ends when one of the following two conditions is met: 1) The different fine hand movements stimulated have met the training needs; 2) All possible combinations of two electrodes have been completed.

2. The method according to claim 1, characterized in that In step b), the two-dimensional hotspot distribution map generated by the finger muscles during single finger movement obtained by quantifying regional neural function is used as prior knowledge. First, two electrode combinations near the hotspot area are searched, and then the electrode combination after the hotspot area is searched.

3. The method according to claim 1, characterized in that If the second round of search has traversed all two-electrode combinations and cannot meet the hand fine motor training requirements, a third round of search will be performed based on the three-electrode combination.

4. The method according to claim 3, characterized in that The fourth round of search is also included, which combines the results obtained from the second and third rounds of search in pairs to produce more fine hand movements, or adds a certain delay between the two combinations to observe the resulting hand movement results.

Images