Photoelectric dual-mode collaborative stimulation wearable rehabilitation auxiliary equipment

Abstract

The invention discloses a wearable rehabilitation assisting device with photoelectric dual-mode collaborative stimulation. The wearable rehabilitation assisting device comprises a signal acquisition unit, a stimulation execution unit, a main control unit and a circuit board, the signal acquisition unit is used for acquiring sensing signals of the wearing area and comprises a myoelectricity sensor and a strain sensor of a hydrogel substrate. The stimulation execution unit comprises a hydrogel electrode, a micro LED array layer and a hydrogel light guide layer. The circuit board is integrated with a signal conditioning circuit, a myoelectricity stimulation generation circuit and an optical genetic stimulation control circuit; a sensing signal is transmitted to the main control unit through the signal conditioning circuit; a deep learning algorithm based on a rehabilitation action evaluation model is built in the main control unit, and a stimulation instruction is generated according to the sensing signal. According to a stimulation instruction, the electromyographic stimulation generation circuit controls the hydrogel electrode to stimulate muscle groups, and the optical genetic stimulation control circuit controls the micro LED array layer to stimulate neurons. The electromyographic stimulation generation circuit and the optical genetic stimulation control circuit work cooperatively through a time data synchronization algorithm and a space mapping algorithm.

Description

Technical Field

[0001] The present invention relates to the field of medical rehabilitation equipment, and in particular to a wearable rehabilitation auxiliary device with multimodal synergistic stimulation. Background Art

[0002] In current neurorehabilitation treatments, single-modality devices based on electrical stimulation or light stimulation still have technical defects. Traditional electrical stimulation technology, represented by functional electrical stimulation (FES), directly triggers muscle contraction by applying electrical pulses. Although it can improve muscle activity in the short term, its mechanism of action only stays at the muscle level and cannot synchronously activate or reshape damaged nerve conduction pathways. When such devices are used for a long time, patients often show "passive muscle contraction without neural control reconstruction." For example, in the upper limb rehabilitation of stroke patients, although electrical stimulation can temporarily lift the arm, it cannot restore the brain's autonomous control of fine hand movements, causing the muscles to be overly dependent on external stimulation. On the other hand, the emerging optogenetic stimulation technology uses light of specific wavelengths to precisely regulate neuronal activity, which can theoretically promote the regeneration of neural pathways, but its driving efficiency for muscles is insufficient when acting alone. Especially in patients with severe nerve damage, there is often a disconnect between neuronal and muscle responses.

[0003] Furthermore, the sensors used in most traditional wearable rehabilitation assistive devices are prone to partial delamination in areas of frequent joint movement due to the misalignment between the bonding material and the human skin. This bonding failure not only attenuates the stimulation signal but also creates motion artifacts that distort the EMG signal acquisition, triggering inappropriate electrical stimulation.

[0004] Therefore, how to achieve multimodal synergistic stimulation of muscle nerves while ensuring high biocompatibility has become a key path to breaking through the current bottleneck of neuromuscular rehabilitation. Summary of the Invention

[0005] The main technical problem to be solved by the present invention is to provide a wearable rehabilitation assistive device with multimodal synergistic stimulation, which avoids the defect of traditional devices in disconnecting nerve and muscle regulation, so as to improve the recovery effect of patients' motor function.

[0006] In order to solve the above technical problems, the present invention provides a wearable rehabilitation assistive device with photoelectric dual-modal synergistic stimulation, which is characterized by comprising a signal acquisition unit, a stimulation execution unit, a main control unit and a circuit board;

[0007] The signal acquisition unit is used to collect sensor signals of the wearable area, including a myoelectric sensor based on a hydrogel substrate and a strain sensor based on an organogel substrate;

[0008] The stimulation execution unit includes a hydrogel electrode, a micro-LED array layer and a hydrogel light-guiding layer; the hydrogel light-guiding layer is arranged on the side of the micro-LED array layer facing the skin; the hydrogel electrode is arranged on the side of the hydrogel light-guiding layer facing the skin;

[0009] The circuit board integrates a signal conditioning circuit, an electromyographic stimulation generating circuit, and an optogenetic stimulation control circuit; an input end of the signal conditioning circuit is connected to the signal acquisition unit to pre-process the sensor signal;

[0010] The main control unit includes at least a single-chip microcomputer arranged on the circuit board; the input end of the single-chip microcomputer is connected to the output end of the signal conditioning circuit, and the output end is respectively connected to the electromyographic stimulation generating circuit and the optogenetic stimulation control circuit; the main control unit is equipped with a deep learning algorithm based on a rehabilitation movement evaluation model, which is used to generate stimulation instructions according to the sensor signal;

[0011] The output end of the electromyographic stimulation generating circuit is connected to the hydrogel electrode to control the hydrogel electrode to stimulate the muscle groups in the wearable area according to the stimulation instruction; the output end of the optogenetic stimulation control circuit is connected to the micro LED array layer to control the micro LED array layer to stimulate the neurons in the wearable area according to the stimulation instruction.

[0012] In a preferred embodiment: the main control unit also includes a host computer; the circuit board also integrates a wireless transmission circuit connected to the single-chip microcomputer; the host computer is wirelessly connected to the single-chip microcomputer through the wireless transmission network; the deep learning algorithm is built into the host computer.

[0013] In a preferred embodiment: the deep learning algorithm adopts a convolutional neural network algorithm, which at least includes a convolution layer, an activation function, a pooling layer and a fully connected layer.

[0014] In a preferred embodiment: the device further includes a vibration feedback unit bound to the VR scene mode; the vibration feedback unit includes a plurality of multi-point distributed vibrators.

[0015] In a preferred embodiment: the main control unit has built-in real-time data synchronization algorithm and spatial mapping algorithm; the real-time data synchronization algorithm is used to control the timing matching between the electromyography stimulation generation circuit and the optogenetic control circuit; the spatial mapping algorithm is used to control the active area of the hydrogel electrode and the micro LED array layer.

[0016] In a preferred embodiment, the hydrogel substrate includes a network made of polyacrylamide, sodium alginate, tannic acid, and partially reduced graphene oxide.

[0017] In a preferred embodiment, the organic gel substrate includes a network made of polyacrylamide, sodium alginate, and tannic acid, and a conductive layer composited with silver nanowires and a conductive polymer poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonic acid).

[0018] In a preferred embodiment: the signal acquisition unit and the stimulation execution unit are respectively embedded in a flexible self-adhesive patch; the flexible self-adhesive patch is made of a composite material of medical silica gel and breathable fabric.

[0019] In a preferred embodiment, the electric pulse signal generated by the electromyography stimulation generation circuit has a frequency range of 10-100 Hz and a pulse width of 100-500 μs.

[0020] In a preferred embodiment, the wavelength range of the micro LED array layer is 450-650nm, and the dynamic adjustable range of light intensity is 0.1-10mW / mm 2 .

[0021] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0022] The rehabilitation assistive device provided by the present invention innovatively integrates electrophysiological regulation and optogenetic intervention through a photoelectric dual-modal synergistic stimulation mechanism, constructing a dual-path optimization system for neuromuscular function reconstruction. The two stimulation modalities achieve complementary synergy in response timing and spatial distribution with the help of a high-precision spatiotemporal synchronization algorithm, thereby significantly improving the efficiency of neural signal conduction and the coordination of muscle contraction, greatly accelerating the patient's motor function recovery. On this basis, the rehabilitation assistive device adopts dual-modal perception of electromyographic sensors and strain sensors to achieve multi-dimensional capture of muscle electrical signals and tiny deformations, improving the perception accuracy of the patient's movement intentions; both the sensing and execution units use an improved low-modulus gel base to achieve flexible conformal contact with human skin with high deformation coordination and fit, improving wearing comfort while reducing signal capture errors caused by motion artifacts; integrating a vibration feedback unit linked to the VR scene mode to strengthen real-time guidance and sensory feedback for patients during immersive training, improving the systematicness and interactivity of rehabilitation training, and improving patient training compliance. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of wearing the rehabilitation assistive device according to an embodiment of the present invention;

[0024] Figure 2 Schematic diagram of a layer cut of the stimulation execution unit according to an embodiment of the present invention;

[0025] Figure 3 This is a flowchart of the working process of the rehabilitation assistive device described in an embodiment of the present invention;

[0026] Figure 4 This is a workflow diagram of the convolutional neural network algorithm in the host computer described in an embodiment of the present invention.

[0027] Marked in the figure: 1-electromyographic sensor, 2-strain sensor, 3-flexible magnetic interface, 4-stimulation execution unit, 41-hydrogel electrode, 42-micro LED array layer, 43-hydrogel light guide layer, 44-flexible transparent substrate, 5-flexible printed circuit board, 6-single-chip microcomputer. DETAILED DESCRIPTION

[0028] The technical solutions 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; it is obvious that the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0029] In the description of the present invention, it should be noted that the terms "upper," "lower," "inner," "outer," "top / bottom," and the like, indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0030] In the description of the present invention, it should be noted that, unless otherwise clearly stipulated and limited, the terms "installed", "provided with", "set / connected", "connected", etc. should be understood in a broad sense. For example, "connection" can be a wall-mounted connection, a detachable connection, or an integral connection. It can be a mechanical connection or an electrical connection. It can be a direct connection or an indirect connection through an intermediate medium. It can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0031] like Figures 1 to 4 As shown, an embodiment of the present invention provides a wearable rehabilitation assistive device with photoelectric dual-modal synergistic stimulation, including a signal acquisition unit, a stimulation execution unit 4, a main control unit and a circuit board.

[0032] The signal acquisition unit includes at least an electromyographic sensor 1. The electromyographic sensor 1 is used to collect and conduct electromyographic signals from parts of the human body, and includes a hydrogel substrate composited with a conductive reinforcing material and a first electrode layer for conducting electromyographic signals. In this embodiment, the main body of the hydrogel substrate is a low-modulus network made of polyacrylamide, sodium alginate, and tannic acid (referred to as a PAM-SA-TA network), and is composited with a partially reduced graphene oxide film (referred to as PRGO). It achieves self-adhesion to the skin through a physical-covalent double cross-linking structure. Specifically, the components of the hydrogel substrate include: 10-15wt% polyacrylamide (PAM), 2-5wt% sodium alginate (SA), 0.5-1.5wt% tannic acid (TA), 0.1-0.3wt% partially reduced graphene oxide (PRGO), 5-8wt% sodium chloride (NaCl), and the rest is a glycerol-water binary solvent with a volume ratio of 3:7. The output end of the first electrode layer is constructed with a flexible magnetic interface 3 so as to be independently connected to the circuit board through a wire.

[0033] In this embodiment, the signal acquisition unit also includes a strain sensor 2. The strain sensor 2 is used to capture the slight movement and deformation of the human body part and transmit it through electrical signals, and includes an organic gel substrate and a second electrode layer. In this embodiment, the organic gel substrate body adopts the PAM-SA-TA network and is coated with a conductive layer composed of silver nanowires (AgNWs) and a conductive polymer poly (3,4-ethylenedioxythiophene) - poly (styrene sulfonic acid) (PED0T-PSS). The organic gel substrate is constructed with a microcrack network. When the human body part produces a slight movement, the organic gel attached to the skin surface is stretched, and the microcracks rapidly expand, causing the conductive path to break and the resistance to increase sharply, thereby generating an electrical signal. The second electrode layer also transmits the generated electrical signal to the circuit board through the flexible magnetic interface 3.

[0034] like Figure 2 As shown, the stimulation execution unit 4 integrates the hydrogel electrode 41, the micro LED array layer 42 and the hydrogel light-guiding layer 43. Specifically, the hydrogel light-guiding layer 43 is arranged on the side of the micro LED array layer 42 facing the skin. It adopts polydimethylsiloxane (PDMS for short) with high light transmittance as the supporting structure, and is composited with titanium dioxide scattering particles. The transmittance is not less than 90%, and the scattering angle is controlled within the range of 30° to 60°. The hydrogel electrode 41 is arranged on the side of the hydrogel light-guiding layer 43 facing the skin. Additionally, the side of the micro LED array layer 42 facing away from the hydrogel light-guiding layer 43 is encapsulated with a flexible transparent substrate 44. Similarly, as the execution hardware, the input end of the stimulation execution unit 4 is connected to the circuit board through a magnetic interface.

[0035] In this embodiment, the myoelectric sensor 1, strain sensor 2, and stimulation actuator 4 are independently embedded in a flexible, self-adhesive patch. This patch is made of a composite material of medical-grade silicone and breathable fabric and is secured to various body parts at multiple points via adjustable straps. The patch boasts an adhesion strength of no less than 7 kPa, an elastic modulus of 0.1-2 MPa, and a skin-fitting tolerance of no more than 0.5 mm, providing ideal conformability.

[0036] In this embodiment, the circuit board adopts a flexible printed circuit board 5, hereinafter referred to as FPC. The FPC integrates a signal conditioning circuit, an electromyographic stimulation generation circuit, an optogenetic stimulation control circuit and a wireless transmission circuit. The main control unit includes a single chip microcomputer 6 (hereinafter referred to as MCU) arranged on the FPC. Figure 3 As shown, the input end of the signal conditioning circuit is connected to the electromyographic sensor 1 and the strain sensor 2 via wires, so that the received electrical signal is converted into a digital signal after amplification and filtering, and then transmitted to the single-chip microcomputer 6 through the output end. The single-chip microcomputer 6 is connected to the wireless transmission circuit to wirelessly transmit the above-mentioned digital signal to the host computer which also belongs to the main control unit. The host computer uses a convolutional neural network (CNN) algorithm to extract features from the received digital signal to identify the patient's movement intention. Then, based on the rehabilitation movement assessment model, the patient's movement intention is compared with the standard movement library. Based on the comparison results, stimulation parameters are generated and wirelessly transmitted to the single-chip microcomputer 6 on the FPC. The output end of the single-chip microcomputer 6 is respectively connected to the input end of the electromyographic stimulation generation circuit and the optogenetic stimulation control circuit, so that stimulation instructions are sent to the electromyographic stimulation generation circuit and the optogenetic control circuit according to the stimulation parameters. According to the stimulation instructions, the electromyographic stimulation generation circuit generates a low-frequency electrical pulse signal to control the hydrogel electrode 41 to stimulate the muscle groups in the wearable area. The optogenetic stimulation control circuit includes an LED driving module, a wavelength control module and a light intensity feedback module, which controls the micro LED array layer 42 to generate light of specific wavelength and intensity according to the stimulation instruction, and stimulates the neurons in the wearable area by activating the photosensitive protein. In this way, the electromyography stimulation generation circuit and the optogenetic stimulation control circuit form a photoelectric dual-modal synergistic stimulation, realizing dual-path optimization and regulation of neural signal conduction and muscle contraction efficiency. In terms of working parameters, the intensity of electromyography stimulation supports multi-level adjustment, the frequency range of the electric pulse signal is 10-100Hz, and the pulse width is 100-500μs. The specific stimulation parameters are dynamically adjusted by the host computer. The wavelength range of the micro LED array layer is 450-650nm, and the dynamic adjustable range of light intensity is 0.1-10mW / mm 2 .

[0037] Furthermore, to achieve synchronous linkage of the optoelectronic dual-modal stimulation modules, the single-chip microcomputer 6 has a built-in real-time data synchronization algorithm, which simultaneously controls the timing matching of the electromyographic stimulation generation circuit and the optogenetic control circuit with microsecond accuracy, and strictly matches the timing. The single-chip microcomputer 6 also has a built-in spatial mapping algorithm to coordinate the positioning of the active areas of the hydrogel electrode 41 and the micro-LED array to avoid conflict between the two in the spatial domain. In summary, the device forms a complete "perception-decision-execution" closed-loop link by collecting sensor signals in real time, intelligent decision-making feedback, and dynamically adjusting the electrical stimulation amplitude, light stimulation intensity, and wavelength. This provides efficient and accurate movement correction and guidance for patients, thereby accelerating the reconstruction of neuromuscular function.

[0038] With the rapid development of cutting-edge technologies, the computing performance of the single-chip microcomputer 6 continues to improve. As a simple alternative to this embodiment, in other embodiments, the single-chip microcomputer 6 can be equipped with a built-in deep learning algorithm based on a rehabilitation movement assessment model, independently completing patient movement recognition, database comparison, and feedback of stimulation parameters without the need for an additional host computer connection. For those skilled in the art, this alternative solution has a mature industry technical foundation.

[0039] like Figure 4 As shown, the host computer's process for extracting and identifying sensor signals will now be described. Prior to this, the signal conditioning circuit has already converted the myoelectric and strain signals into standardized digital signals through amplification, filtering, and analog-to-digital conversion. As a mature existing technology, the specific process of signal pre-conditioning will not be detailed in this article.

[0040] Then we enter the feature extraction stage: the sliding window technique is used to segment the time domain data in the sensor signal into multiple subsequences and project them into a two-dimensional space for subsequent classification model processing. The sliding window size is set to 32 and the step size is set to 16 to maintain a 50% overlap rate. This method can capture the time-varying nature of the signal while retaining sufficient information for effective feature extraction. The specific feature extraction process is as follows: (1) The sensor signal of each patient action is segmented to form a window list; (2) The data in the window list is spliced into the input signal list inputs and the corresponding label value is assigned to each window; (3) The input signal list inputs and the label list outputs are converted into numpy arrays; (4) The data dimension is adjusted; (5) Finally, the function returns the processed signal data and the corresponding label data.

[0041] In this embodiment, the classification model adopts the convolutional neural network (CNN) algorithm. Convolutional neural network is a deep learning model specially used to process data with grid structure. Its core idea is to extract features in the image through convolution layers and pooling layers, and map these features to different categories through fully connected layers. Its main components include convolution layers, activation functions, pooling layers, fully connected layers and loss functions. The model mainly consists of two convolutional neural network feature extraction blocks and a fully connected neural network classifier block. Among them, the first convolution layer CNN1 has 32 output filters, uses ReLU activation function, and the convolution kernel size is 3×5; the second convolution layer CNN2 has 64 output filters, uses ReLU activation function, and the convolution kernel size is 3×5.

[0042] Each CNN feature extraction block is followed by batch normalization, PReLU activation, spatial 2D Dropout, and maximum pooling layers. PReLU is a variant of ReLU. The difference is that PReLU introduces a learnable parameter that can adjust the slope of the activation function, so that the neural network can learn different activation function slopes for different parts during training, thereby enhancing the network's representational ability and adaptability. Spatial 2DDropout is also a regularization technique used to reduce overfitting problems. The principle of spatial 2D Dropout is to set the output of certain neurons to zero with a certain probability during training, that is, to randomly inactivate a part of neurons during forward propagation. This helps to reduce the co-adaptation between neurons and forces the network to learn more robust and generalized feature representations, thereby reducing the risk of overfitting; in spatial 2DDropout, not only the output of neurons will be randomly discarded, but also some features in the entire feature map will be discarded, which can better retain the spatial information between features and help improve the generalization ability of the network. This part of the work is Figure 4 The "2D channel random deactivation" embodied in .

[0043] The features are then flattened and classified using a fully connected neural network classifier. If the number of neurons is specified, an additional fully connected layer is added to the classifier block. The final output layer has num_classes neurons and uses a softmax activation function for classification to obtain the output recognition result. After the CNN completes the classification and recognition of the patient's movements, the output recognition result is compared with the standard library of the rehabilitation movement assessment model. Based on the comparison results, the stimulation adjustment strategy is determined to send the corresponding stimulation parameters to the microcontroller 6 to guide the stimulation execution unit 4 to synergistically stimulate the muscles and nerves in the wearable area.

[0044] To facilitate further understanding, the various components of the device are now reviewed from a functional perspective. The device includes a signal acquisition module, a decision-making module, and a stimulation module. The signal acquisition module includes the myoelectric sensor 1, the strain sensor 2, and the signal conditioning circuit integrated into the FPC, which is used to acquire and transmit bimodal sensor signals and convert them into digital signals. The decision-making module includes the single-chip microcomputer 6 configured in the FPC and the host computer in wireless communication with the single-chip microcomputer 6. After receiving the digital signal from the signal acquisition module, it processes and makes decisions based on the CNN algorithm and sends specific stimulation instructions to the photoelectric stimulation module. The stimulation module operates in a dual-modal manner through the myoelectric stimulation module and the optogenetic stimulation module. Specifically, the myoelectric stimulation module includes the myoelectric stimulation generation circuit and the hydrogel electrodes 41 connected thereto, which are used to respond to the stimulation instructions to provide electrical stimulation to the target muscle group. The optogenetic stimulation module includes the optogenetic stimulation control circuit and the micro-LED array layer 42 and hydrogel light-guiding layer 43 connected thereto, which are used to respond to the stimulation instructions to provide optogenetic stimulation to neurons in the target area.

[0045] Additionally, the device also includes a vibration feedback unit, which includes a plurality of vibrators distributed in a multi-point manner at key joint positions of the human limbs. In this embodiment, the vibrator adopts a micro linear motor. The triggering logic of the vibration feedback unit is bound to the VR scene mode of the device. When the user's action deviates from the preset, the micro linear motor provides the user with real-time correction prompts by triggering vibrations of different intensities and patterns. Thanks to the vibration feedback unit, the device has a stronger sense of immersion and interactivity, thereby improving the efficiency of rehabilitation training.

[0046] In summary, the rehabilitation assistive device provided by the embodiment of the present invention innovatively integrates electrophysiological regulation and optogenetic intervention through a photoelectric dual-modal synergistic stimulation mechanism to construct a dual-path optimization system for neuromuscular function reconstruction. The electromyography stimulation module accurately drives muscle contraction based on adjustable low-frequency pulses, and the synchronously running optogenetic stimulation module targets and activates neurons through wavelength-specific light sources. The two stimulation modalities achieve complementary synergy in response timing and spatial distribution with the help of a high-precision spatiotemporal synchronization algorithm, thereby significantly improving the efficiency of neural signal conduction and the coordination of muscle contraction, greatly accelerating the patient's motor function recovery. On this basis, the rehabilitation assistive device also has the following optimizations: the dual-modal perception of the electromyographic sensor 1 and the strain sensor 2 realizes the multi-dimensional capture of muscle electrical signals and tiny deformations, and improves the perception accuracy of the patient's movement intentions; the sensing and execution units both adopt a modified low-modulus gel base, which achieves flexible conformal contact with human skin with higher deformation coordination and fit, while improving wearing comfort and reducing the signal capture error caused by motion artifacts; the vibration feedback unit linked to the VR scene mode is integrated to strengthen the real-time guidance and sensory feedback of patients during immersive training, improve the systematicness and interactivity of rehabilitation training, and improve the patient's training compliance.

[0047] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any technical equivalent transformation made using the contents of the present invention specification shall fall within the protection scope of the present invention.

Claims

1. A wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation, characterized by: It includes a signal acquisition unit, a stimulation execution unit, a main control unit and a circuit board; The signal acquisition unit is used to collect sensor signals of the wearable area, including a myoelectric sensor based on a hydrogel substrate and a strain sensor based on an organogel substrate; The stimulation execution unit includes a hydrogel electrode, a micro-LED array layer and a hydrogel light-guiding layer; the hydrogel light-guiding layer is arranged on the side of the micro-LED array layer facing the skin; the hydrogel electrode is arranged on the side of the hydrogel light-guiding layer facing the skin; The circuit board integrates a signal conditioning circuit, an electromyographic stimulation generating circuit, and an optogenetic stimulation control circuit; an input end of the signal conditioning circuit is connected to the signal acquisition unit to pre-process the sensor signal; The main control unit includes at least a single-chip microcomputer arranged on the circuit board; the input end of the single-chip microcomputer is connected to the output end of the signal conditioning circuit, and the output end is respectively connected to the electromyographic stimulation generating circuit and the optogenetic stimulation control circuit; the main control unit is equipped with a deep learning algorithm based on a rehabilitation movement evaluation model, which is used to generate stimulation instructions according to the sensor signal; The output end of the electromyographic stimulation generating circuit is connected to the hydrogel electrode to control the hydrogel electrode to stimulate the muscle groups in the wearable area according to the stimulation instruction; the output end of the optogenetic stimulation control circuit is connected to the micro LED array layer to control the micro LED array layer to stimulate the neurons in the wearable area according to the stimulation instruction.

2. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: The main control unit also includes a host computer; the circuit board also integrates a wireless transmission circuit connected to the single-chip microcomputer; the host computer is wirelessly connected to the single-chip microcomputer through the wireless transmission network; the deep learning algorithm is built into the host computer.

3. A wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to any one of claims 1 or 2, characterized in that: The deep learning algorithm adopts a convolutional neural network algorithm, which includes at least a convolution layer, an activation function, a pooling layer and a fully connected layer.

4. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: It also includes a vibration feedback unit bound to the VR scene mode; the vibration feedback unit includes a plurality of multi-point distributed vibrators.

5. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: The main control unit has built-in real-time data synchronization algorithm and spatial mapping algorithm; the real-time data synchronization algorithm is used to control the timing matching between the electromyography stimulation generation circuit and the optogenetic control circuit; the spatial mapping algorithm is used to control the active area of the hydrogel electrode and the micro LED array layer.

6. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: The hydrogel substrate comprises a network made of polyacrylamide, sodium alginate, tannic acid, and partially reduced graphene oxide.

7. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: The organic gel substrate comprises a network made of polyacrylamide, sodium alginate and tannic acid, and a conductive layer composited with silver nanowires and conductive polymer poly(3,4-ethylenedioxythiophene)-poly(styrene sulfonic acid).

8. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: The signal acquisition unit and the stimulation execution unit are respectively embedded in a flexible self-adhesive patch; the flexible self-adhesive patch is made of a composite material of medical silica gel and breathable fabric.

9. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: The electric pulse signal generated by the electromyography stimulation generation circuit has a frequency range of 10-100 Hz and a pulse width of 100-500 μs.

10. The wearable rehabilitation assistive device with optoelectronic dual-modal synergistic stimulation according to claim 1, characterized in that: The wavelength range of the micro LED array layer is 450-650nm, and the dynamic adjustable range of light intensity is 0.1-10mW / mm 2 .

Images