A stretchable wristband high-density electromyographic system for thumb micro-gesture recognition

Abstract

The present invention discloses a stretchable wristband high-density electromyography system for thumb micro-gesture recognition, including a high-density electromyography array and a micro wireless acquisition module; the high-density electromyography array includes three layers of flexible materials, the middle layer of flexible material is printed with flexible electrodes and interconnects, the uppermost and lowermost layers of flexible materials can be removed when in use, and the high-density electromyography array is formed with three parts: a soft skin sensing component, a flexible interconnection, and a rigid pluggable contact pad, wherein the soft skin sensing component is used to closely conformally adhere to the wrist skin, and a flexible electrode is arranged on it, which is electrically connected to the rigid pluggable contact pad through a flexible interconnection, and the rigid pluggable contact pad is a plug-and-play FPC interface for connecting to the micro wireless acquisition module. The system can be directly fitted with the wrist, capture high-fidelity surface electromyography of dynamic and fine gestures centered on the thumb, and apply it to human-computer interaction control.

Description

Technical Field

[0001] The invention belongs to the technical field of gesture recognition, and relates to a stretchable wristband high-density electromyographic system for thumb micro-gesture recognition and a preparation method thereof. Background Art

[0002] Traditional human-computer interaction technologies are often limited to input methods such as keyboards, mice or touch screens, which cannot meet user needs in some cases. Although the emergence of visual sensors has made up for this defect to a certain extent, it is still affected by visual impairments, especially in low light or dark environments. The recognition effect based on visual sensors will be seriously affected. This dilemma is even more prominent for user groups who have difficulty in gesture recognition, such as patients with finger defects.

[0003] Surface electromyography (sEMG)-based hand or wrist gesture recognition is a viable solution, but there are still two major challenges that greatly limit their practicality for thumb-dominated dynamic fine gesture recognition, which is often required in intuitive human-computer interaction. First, the location of electrodes is crucial to improve sEMG efficacy. Traditional electromyographic pattern recognition research has focused on forearm placement, mainly driven by prosthetic control applications, while anatomical understanding shows that the wrist joint is a hotspot for key muscles that control thumb movement. Therefore, wrist-worn electrodes can improve the fidelity of collected signals and meet consumer preferences for wrist-worn devices. Current work on forearm-placed sEMG does not provide a path for accurate recognition of thumb-dominated dynamic fine gestures. Second, subtle and dynamic thumb-dominated gesture recognition requires reliable ultra-conformable neuromuscular interfaces and large-area and high-density sEMG arrays. Current soft electrode arrays either lack a hybrid design of a central soft sensing part and a rigid connection part, resulting in unstable connection with an external rigid acquisition module, or are difficult to manufacture in large areas or on a large scale. Therefore, an ideal sEMG device for fine gesture control should be able to discern subtle dynamic movements of the thumb, ensure soft and intimate contact, provide comprehensive area coverage, and provide high-density sensing. In addition, it should utilize an optimal location, especially on the wrist, for detailed muscle activation studies.

[0004] Based on this, the present invention proposes a wireless, wrist-worn stretchable high-density electromyographic array system, which aims to collect, identify and control the electromyographic signals of thumb micro-movements. The main purpose of this technology is to build an immersive human-computer interface with thumb-controlled gesture recognition as the core. This function is achieved by capturing the electromyographic signals of the user's wrist area and analyzing and interpreting them efficiently and accurately.

[0005] Purpose of the Invention

[0006] The purpose of the present invention is to address the deficiencies in the prior art and provide a stretchable wristband high-density electromyography system for thumb micro-gesture recognition and a preparation method thereof. The system can provide strong skin fit and plug-and-play capability, can directly fit the wrist, capture high-fidelity surface electromyography of dynamic fine gestures centered on the thumb, and apply it to human-computer interaction control.

[0007] The technical solution adopted by the present invention is as follows:

[0008] A stretchable wristband high-density electromyography system for thumb micro-gesture recognition comprises a high-density electromyography array and a micro wireless acquisition module; the high-density electromyography array is used to be attached to the wrist to obtain electromyography signals and transmit them to the micro wireless acquisition module, comprising three layers of flexible materials, a middle layer of flexible material is printed with flexible electrodes and interconnection lines, the uppermost and lowermost layers of flexible materials can be removed when in use, and three parts, a soft skin sensing component, a flexible interconnection and a rigid pluggable contact pad, are formed on the high-density electromyography array, wherein the soft skin sensing component is used to closely and conformally adhere to the wrist skin, a flexible electrode is arranged on the soft skin sensing component, and is electrically connected to the rigid pluggable contact pad through the flexible interconnection, and the rigid pluggable contact pad is a plug-and-play FPC interface for connecting to the micro wireless acquisition module.

[0009] In the above technical solution, further, the three-layer flexible material includes: a surface layer using PET material, a middle layer using PU material, and a bottom layer using PET material; the middle layer is formed by screen printing flexible Ag / AgCl silver paste to form the flexible electrode and interconnection line, and between the middle layer and the surface layer, except for the electrode sensing functional unit, that is, the flexible electrode, the part is covered with skin double-sided tape for adhering to the skin, and the electrode sensing functional unit, that is, the surface of the flexible electrode is covered with conductive paste or conductive hydrogel.

[0010] Furthermore, the skin double-sided adhesive is an acrylic double-sided skin adhesive.

[0011] Furthermore, when in use, first tear off the surface PET material to expose the skin double-sided adhesive tape and conductive paste or conductive hydrogel, stick the high-density electromyography array on the wrist with this side facing down, connect the rigid pluggable contact pad to the micro wireless acquisition module, and remove the bottom PET material.

[0012] Furthermore, the flexible electrode includes 2 common ground electrodes and 32 working electrodes, which evenly cover the wrist area for recording 16-channel differential surface electromyography signals.

[0013] Furthermore, the flexible electrode adopts a protruding structure, and in particular, can adopt a protruding point array.

[0014] Furthermore, the surface of the interconnection line on the middle layer at the rigid easy-plug contact pad is covered with carbon paste, and the thickness of the flexible material thereon is thicker than that of the flexible interconnection part.

[0015] Furthermore, the micro wireless acquisition module includes an electromyographic signal module, a signal processing module, and a wireless module, which are respectively used to complete the acquisition of electromyographic signals transmitted by the high-density electromyographic array, and the processing and wireless transmission of the acquired signals.

[0016] The wristband high-density electromyographic system disclosed in the present invention has low modulus and high scalability in the soft skin sensing part, which enables it to flexibly adapt to skin deformation, thereby achieving accurate surface electromyography recording. The electromyographic array is designed to fit the wrist, which not only improves the wearing comfort, but also enhances the stability, thereby ensuring the accuracy and reliability of signal acquisition, and can effectively capture the electromyographic signals in the wrist area, so as to accurately identify the micro-gestures of the thumb. Compared with traditional electromyographic sensors, the array has excellent signal sensitivity and can adapt to and perfectly fit the skin surface of complex curved surfaces. By using the wristband high-density electrode system, it is not only possible to recognize and respond to the user's gestures, but also to break through visual obstacles through real-time monitoring and interpretation of electromyographic signals, bringing new possibilities for gesture recognition and human-computer interaction, especially to achieve reliable gesture recognition and human-computer interaction under conditions of dim external environment or insufficient light. This design provides a more convenient and accurate interactive experience, especially in the face of visual obstacles and finger defects, etc., it has significant advantages and value. At the same time, the system uses wireless communication technology, allowing users to move freely and interact with the device, improving the convenience and flexibility of use, and increasing the comfort and fluency of user experience. In addition, the myoelectric system can not only help patients with finger amputations solve operational difficulties, but can also be used in virtual reality, augmented reality and other fields, expanding new possibilities for human-computer interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is an overall schematic diagram of an example of a high-density myoelectric system of the present invention.

[0018] Figure 2 A top view of an example of a high-density electromyography array of the present invention.

[0019] Figure 3 The exploded diagram of an example of a high-density electromyographic system of the present invention.

[0020] Figure 4 The figure is a schematic diagram of the structure of a micro wireless acquisition module example of the present invention. DETAILED DESCRIPTION

[0021] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0022] like Figure 1 As shown in FIG. 1 , a specific embodiment of the present invention is shown. The stretchable wristband high-density electromyography system (LS-EMG patch for short) for thumb micro-gesture recognition is composed of a high-density electromyography array 20 and a micro wireless acquisition module 30 connected thereto.

[0023] The high-density electromyographic array 20 is made of three layers of flexible materials, and the surface layer 21 is made of PET film to protect the high-density electromyographic array 20. When using the high-density electromyographic array 20, the user can tear off the surface layer 21. The middle layer 22 is made of PU film, and Ag / AgCl is screen-printed on its surface. It is used to receive electromyographic signals and is covered with double-sided skin glue to ensure that the high-density electromyographic array is attached to the wrist for a long time. It also includes a bottom layer 25, which is made of PET film as a temporary support and protective layer; after the user attaches the middle layer 22 to the wrist 10, tear off the bottom layer 25, and then connect the micro wireless acquisition module 30 to the FPC interface of the high-density electromyographic array 24. At this time, the entire high-density electromyographic system is worn. The user generates electromyographic signals by performing thumb micro-gestures, and these signals are collected by the high-density electromyographic array 20 and transmitted to the micro wireless acquisition module. The micro wireless acquisition module processes the electromyographic signals and transmits them wirelessly, transferring the micro-gestures to the robot or human-computer interaction control terminal, thus realizing the conversion of the user's thumb gestures into gestures on the robot or human-computer interaction.

[0024] like Figure 2 , 3 As shown, it is a top view of the high-density electromyography system and a schematic diagram of each layer structure. By designing it into a fan shape, it conforms to the structure of the wrist, thereby improving the portability of operation and wearing comfort. According to a specific embodiment of the present invention, the flexible electrode has two uppermost common ground electrodes and 32 working electrodes, which are intended to record 16-channel differential sEMG signals in a bipolar form. Each electrode has an in-plane size of 5.5×5.5mm 2 The total coverage area of ​​the LS-EMG patch is approximately 182×53mm 2 , enough to cover the entire circumference of the wrist so as to collect sEMG signals from all directions at the same time. The electrode design adopts a convex structure. During the test, the sensing electrode is higher than the surrounding packaging layer, which is easier to contact with the human skin, effectively improving the test stability and quality of the electromyographic signal. The multi-channel convex surface electromyographic electrode is lightweight, wearable, and reusable. It can comfortably and non-invasively measure electromyographic signals with sufficient spatial resolution and sufficient accuracy for a long time without the need for medical conductive gel. Figure 3The multilayer structure of the LS-EMG patch is schematically illustrated, which includes: (i) an ultrathin soft polyurethane (PU) substrate 22 (20 μm, ~6 MPa), (ii) stretchable Ag / AgCl traces 23 (10 μm) used as electrodes and interconnects, (iii) a tough conductive hydrogel film (55 μm, ~60 kPa) used as an ultrasoft conductive bioadhesive interface with the skin (not shown in the figure), and (iv) a patterned acrylic-based double-sided skin adhesive (30 μm) to ensure isolated encapsulation and firm skin adhesion (not shown in the figure); in addition, there is a bottom layer 25 with a temporary support and protective PET layer, and a surface protective layer PET film 21, the surface layer is used to protect the acrylic double-sided skin adhesive of the middle layer, ensure the adhesion of the electromyographic array and protect the middle layer, and extend the service life of the electromyographic array. The bottom layer 25 is used to protect the high-density electromyographic array and long-term storage, ensuring the stability and durability of the entire electromyographic system. More importantly, this PET support can reduce wrinkles and greatly facilitate manual operation of large-area ultrathin soft PU films. This approach allows the printing of stretchable Ag / AgCl traces and durable carbon pads directly onto ultrathin, soft PU substrates. In addition, a composite film of acrylic-based double-sided skin adhesive (30 μm) and a release PET protective layer, or top layer (25 μm), was carefully bonded on top, so that the soft skin sensing part was wrapped between two release layers of relatively hard PET film to enhance stability and operability. Crucially, this strategy allows for large-scale production of LS-EMG patches to meet a wide range of usage needs.

[0025] like Figure 4 FIG. 1 is a schematic diagram of the structure of a micro wireless acquisition module 30 in an embodiment of the present invention.

[0026] In order to achieve an electrical interface with external electronic equipment for collecting data, traditional soft electromyographic signal patch designs usually rely on manually assembled flexible cables (such as anisotropic conductive films) or micro-wires, which can be very time-consuming and easy to debond. In contrast, the LS-EMG patch of the present invention can provide more powerful skin fit and plug-and-play capabilities. According to a specific embodiment of the present invention, by selectively bonding thin (25μm) and thick (3mm) PET films to the surfaces of the flexible interconnection part and the rigid pluggable contact pad part, the LS-EMG patch is divided into three functional parts when worn on the wrist, and has different mechanical properties. Including soft skin sensing parts, flexible interconnections and rigid pluggable contact pads. The soft skin sensing part is mainly ultra-thin (e.g., an overall thickness of 50μm and 75μm at the electrode), with inherent stretchability and self-adhesion, ensuring tight and conformal attachment to the flexible skin even under significant compression. The soft skin sensing part is preferably shaped into a fan shape to better conform to the three-dimensional configuration of the wrist. The rigid easy-plug contact pads combined with the flexible interconnects provide a rugged and durable plug-and-play electrical interface to the standard flexible printed circuit (FPC) connector, bridging the soft sensing components and the rigid acquisition module. The scratch-resistant conductive carbon paste is used to enhance the mechanical reliability of the contact pads during repeated plugging and unplugging.

[0027] When used specifically, the entire operation process is similar to using a commercially available skin bandage, involving four simple steps: (i) tearing off the PET protective film to expose the skin adhesive layer, (ii) gently pressing to place the soft sensing part on the subject's wrist, (iii) removing the PET support film, and (iv) connecting to the wireless data acquisition module for immediate use; after the user connects the micro wireless acquisition module to the high-density electromyographic array, the acquisition module can be started by long pressing the switch on the acquisition module. The power module 35 inside the acquisition module is equipped with a small lithium battery, which can meet the needs of long-term wear and use, and is equipped with a power display light to facilitate users to understand the working status of the system. When the user makes a thumb gesture, the generated electromyographic signal will be collected by the electromyographic signal collection module 31 and then transmitted to the signal processing module 32 for processing. Subsequently, the processed electromyographic signal is transmitted to the first wireless communication module 33, which converts the physiological electrical signal into a digital signal for gesture recognition and human-computer interaction control. This technology can not only solve the problem of gesture recognition of visual sensors in the case of visual impairment or insufficient optical fiber, but also help patients with finger defects to manipulate the manipulator, thereby improving the quality of life of patients. If the electromyographic signals are not received or processed in time, they will be stored in the storage module 34 for later analysis or processing, thereby ensuring the stability and reliability of the system.

[0028] The thumb's ability to flex, extend, abduct, and adduct enables it to independently perform a variety of dynamic gestures, such as sliding in all directions, such as thumb slide left (TSL), thumb slide right (TSR), thumb slide up (TSU), thumb double-tap down (TDCD), thumb slide clockwise (TCW), etc. These are not just arbitrary movements, but closely simulate the movements that people may take when browsing digital menus or interacting with guided user interfaces. Especially in immersive environments, such dynamic gestures are expected to achieve more subtle and refined control. After collecting electromyographic signals, artificial intelligence can be used to assist in the recognition of thumb-dominated micro-gestures. In the field of gesture recognition based on electromyographic signals, various classifiers have been explored. These algorithms cover traditional machine learning algorithms, such as support vector machines (SVM) and hidden Markov models (HMM), as well as a series of deep learning algorithms, including convolutional neural networks (CNN) and recurrent neural networks (RNN). In the deep learning algorithm, according to a specific embodiment of the present invention, a lightweight neural network based on 1D-CNN is constructed, which includes three blocks, each of which consists of four parallel convolutional neural networks. These networks have convolutional layers with kernel sizes of 1, 11, 21, 31, and 41, and the counts of grouped convolutions are set to 96, 32, 32, 32, and 32. By utilizing multi-scale networks, the corresponding sEMG signal features can be more comprehensively discerned from various thumb motion gestures.

[0029] The above disclosure is only part of the preferred embodiments of the present invention, but the present application is not limited thereto. Any modifications or equivalent substitutions made by relevant technical personnel within the core theoretical scope of the present invention should fall within the protection scope of the present invention.

Claims

1. A stretchable wristband high-density electromyographic system for thumb micro-gesture recognition, characterized in that: It includes a high-density electromyographic array and a micro wireless acquisition module; the high-density electromyographic array is used to be attached to the wrist to obtain electromyographic signals and transmit them to the micro wireless acquisition module, including three layers of flexible materials, the middle layer of flexible material is printed with flexible electrodes and interconnections, the uppermost and lowermost layers of flexible materials can be removed when in use, and the high-density electromyographic array is formed with three parts: a soft skin sensing component, a flexible interconnection and a rigid pluggable contact pad. The soft skin sensing component is used to closely and conformally adhere to the wrist skin, and a flexible electrode is arranged on it, which is electrically connected to the rigid pluggable contact pad through a flexible interconnection, and the rigid pluggable contact pad is a plug-and-play FPC interface for connecting to the micro wireless acquisition module.

2. The stretchable wristband high-density electromyographic system based on thumb micro-gesture according to claim 1 is characterized in that: The three-layer flexible material includes: a surface layer using PET material, a middle layer using PU material, and a bottom layer using PET material; the middle layer is formed by screen printing flexible Ag / AgCl silver paste to form the flexible electrode and interconnection line, and between the middle layer and the surface layer, except for the electrode sensing functional unit, i.e., the flexible electrode, the part is covered with skin double-sided tape for adhering to the skin, and the electrode sensing functional unit, i.e., the surface of the flexible electrode is covered with conductive paste or conductive hydrogel.

3. The stretchable wristband high-density electromyographic system for thumb micro-gesture recognition according to claim 2, characterized in that: The skin double-sided adhesive tape adopts acrylic double-sided skin adhesive.

4. The stretchable wristband high-density electromyographic system for thumb micro-gesture recognition according to claim 2, characterized in that: When in use, first tear off the surface PET material to expose the double-sided adhesive tape and conductive paste or conductive hydrogel, then stick the high-density electromyography array on the wrist with this side facing down, connect the rigid pluggable contact pad to the micro wireless acquisition module, and remove the bottom PET material.

5. The stretchable wristband high-density electromyographic system for thumb micro-gesture recognition according to claim 1, characterized in that: The flexible electrode includes 2 common ground electrodes and 32 working electrodes, which evenly cover the wrist area and are used to record 16-channel differential surface electromyography signals.

6. The stretchable wristband high-density electromyographic system for thumb micro-gesture recognition according to claim 1, characterized in that: The flexible electrode adopts a convex structure.

7. The stretchable wristband high-density electromyographic system for thumb micro-gesture recognition according to claim 1, characterized in that: The surface of the interconnection line on the middle layer at the rigid easy-plug contact pad is covered with carbon paste, and the thickness of the flexible material thereon is thicker than that of the flexible interconnection part.

8. The stretchable wristband high-density electromyographic system for thumb micro-gesture recognition according to claim 1, characterized in that: The micro wireless acquisition module comprises an electromyographic signal module, a signal processing module and a wireless module, which are respectively used for acquiring electromyographic signals transmitted by a high-density electromyographic array, and processing and wirelessly transmitting the acquired signals.

Images