A physiotherapy control method, a wearable physiotherapy device and a computer program product

Abstract

The embodiment of the application is suitable for the technical field of physiotherapy equipment, and provides a physiotherapy control method, a wearable physiotherapy device and a computer program product. The method is applied to a signal processing component of a wearable physiotherapy device. The signal processing component is connected with a textile body. The textile body is provided with a sensing node and a physiotherapy execution component. The method comprises the following steps: acquiring a human feature signal collected by the sensing node; determining a fatigue state grade according to the human feature signal; and sending a physiotherapy control signal corresponding to the fatigue state grade to the physiotherapy execution component, so as to instruct the physiotherapy execution component to execute a physiotherapy strategy corresponding to the physiotherapy control signal. The embodiment of the application can realize a full-automatic closed loop from the sensing of a wearable physiotherapy device from a human feature signal, to the sensing of a fatigue state grade obtained by analyzing the human feature signal, and to the execution of a corresponding physiotherapy strategy according to the fatigue state grade.

Description

Technical Field

[0001] This application belongs to the field of physiotherapy equipment technology, and in particular relates to a physiotherapy control method, wearable physiotherapy equipment, and computer program product. Background Technology

[0002] Smart wearable devices are devices that can be worn on the body and integrate computing power, sensors and wireless communication functions. Health monitoring is one of the common intelligent functions of smart wearable devices.

[0003] Physiotherapy equipment uses physical factors (such as electricity, light, sound, heat, magnetism, force, etc.) to act on the human body in order to relieve pain, promote blood circulation, accelerate tissue repair, and restore function. Physiotherapy equipment is usually a standalone device that requires manual operation and has fixed functions.

[0004] The health monitoring function of smart wearable devices is completely separate from the physical intervention function of physiotherapy devices, which are usually integrated into smart wearable devices. Physiotherapy devices cannot dynamically, accurately, and adaptively adjust according to the user's real-time physical condition (such as the fatigue level of specific muscles), resulting in a poor user experience and limited effectiveness.

[0005] Therefore, there is an urgent need for a wearable device that includes health monitoring and physical therapy functions. Summary of the Invention

[0006] In view of this, embodiments of this application provide a physiotherapy control method, a wearable physiotherapy device, and a computer program product, which enable the wearable physiotherapy device to achieve a fully automatic closed loop from sensing human body characteristic signals, to obtaining fatigue level perception based on human body characteristic signal analysis, and then to executing corresponding physiotherapy strategies based on fatigue level.

[0007] A first aspect of this application provides a physiotherapy control method, applied to a signal processing component of a wearable physiotherapy device, the signal processing component being connected to a textile body: the textile body is provided with sensing nodes and a physiotherapy execution component; the method includes: Acquire human feature signals collected by sensor nodes; The fatigue level is determined based on human body characteristic signals; Send a physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component to instruct the physiotherapy execution component to execute the physiotherapy strategy corresponding to the physiotherapy control signal.

[0008] In some implementations of the first aspect, the sensing node is located in the therapeutic area of ​​the textile body, and the sensing node is formed by interlacing magnetic sensing fibers and conductive yarns; the human body characteristic signal is a strain signal sequence; the fatigue state level is determined based on the human body characteristic signal, including: The strain signal sequence is input into the trained artificial intelligence model to instruct the model to determine the fatigue state level of the physiotherapy area based on the strain signal sequence, and to determine the physiotherapy control signal corresponding to the fatigue state level. The fatigue state level includes the non-fatigue state and one or more fatigue levels corresponding to the fatigue state.

[0009] In some implementations of the first aspect, the strain signal sequence is input into a trained artificial intelligence model, including: The strain signal sequence is preprocessed to obtain the preprocessed signal; The preprocessed signal is sliced ​​according to a certain time window to obtain the sample to be processed corresponding to each time window. The sample to be processed is input into the trained artificial intelligence model; Preprocessing includes filtering and / or normalizing the amplitude of each strain signal in the strain signal sequence.

[0010] In some implementations of the first aspect, after sending the physiotherapy control signal corresponding to the fatigue state level to the physiotherapy execution component, the method further includes: Acquire physiotherapy feedback signals; physiotherapy feedback signals include re-acquired human characteristic signals; Determine the expected physical therapy effects corresponding to the physical therapy strategy; If the physiotherapy feedback signal does not match the expected physiotherapy effect, adjust the physiotherapy control signal.

[0011] In some implementations of the first aspect, the method also includes: Record personalized training samples; personalized training samples include corresponding physical therapy feedback signals and adjusted physical therapy control signals; The AI ​​model is trained using personalized training samples to instruct it to generate personalized sub-models.

[0012] In some implementations of the first aspect, the wearable physiotherapy device is further provided with at least one physiological signal sensor; the physiotherapy feedback signal also includes physiological signals collected by the physiological signal sensor; the physiological signals include at least one of skin temperature signal and heart rate signal.

[0013] In some implementations of the first aspect, the physiotherapy execution component includes a mechanical drive; the physiotherapy control signal includes a first control command; and the mechanical drive is used to periodically contract in response to the first control command to apply mechanical pressure.

[0014] In some implementations of the first aspect, the physiotherapy execution component includes a thermal energy unit; the physiotherapy control signal includes a second control command; and the thermal energy unit is used to release thermal energy in response to the second control command.

[0015] A third aspect of this application provides a wearable physiotherapy device, including: a signal processing component and a textile body. The signal processing component is provided with a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the wearable physiotherapy device implements the physiotherapy control method of the first aspect described above.

[0016] A fourth aspect of this application provides a computer program product, including a computer program that, when run, causes the physiotherapy control method of the first aspect described above to be executed.

[0017] The fifth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the physiotherapy control method of the first aspect described above.

[0018] The wearable physiotherapy device in this embodiment includes an interconnected signal processing component and a textile body. The textile body is equipped with a sensing node and a physiotherapy execution component. The signal processing component acquires human characteristic signals collected by the sensing node; determines the fatigue level based on the human characteristic signals; and sends a physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component to instruct the physiotherapy execution component to execute the physiotherapy strategy corresponding to the physiotherapy control signal. This enables the wearable physiotherapy device to achieve a fully automatic closed loop from sensing human characteristic signals to obtaining fatigue level sensing based on human characteristic signal analysis, and then executing the corresponding physiotherapy strategy (physical intervention) based on the fatigue level, thus solving the functional separation problem of fatigue detection and physical intervention in the prior art. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of a physiotherapy control method provided in an embodiment of this application; Figure 2 This is a partial schematic diagram of a textile body provided in an embodiment of this application; Figure 3 This is a cross-sectional view of a physiotherapy area provided in an embodiment of this application; Figure 4 This is a schematic diagram of strain signal sequence processing provided in an embodiment of this application; Figure 5This is a schematic diagram of a signal conditioning circuit provided in an embodiment of this application; Figure 6 This is a schematic diagram of the physiotherapy feedback mechanism provided in the embodiments of this application; Figure 7 This is a schematic diagram of a physiotherapy control device provided in an embodiment of this application; Figure 8 This is an example diagram of an application of the wearable physiotherapy device provided in the embodiments of this application. Detailed Implementation

[0021] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0022] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0023] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0024] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0025] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0027] One of the inventive concepts of this application is to prepare functional yarns with different functions such as sensing, conductivity, and driving; to integrate the different functional yarns into a wearable textile body through textile processes such as weaving and embroidery; and then to electrically connect the textile body to a signal processing component and embed a physiotherapy control algorithm in the signal processing component to obtain a wearable physiotherapy device. The technical solution of this application will be described below through specific embodiments.

[0028] Reference Figure 1 The diagram shows a schematic of a physiotherapy control method provided in an embodiment of this application. The embodiment of this application applies to a signal processing component of a wearable physiotherapy device. The signal processing component is connected to a textile body. The textile body is provided with a sensing node and a physiotherapy execution component.

[0029] The specific steps included in this application embodiment are as follows: Step 101: Acquire human feature signals collected by the sensor nodes; The embodiments of this application do not limit the human body parts to which the textile body is applied. For example, the textile body can be a sleeve, which is suitable for the arm; the textile body can be an upper garment, which is suitable for the torso and upper limbs; the textile body can be trousers, which is suitable for the lower limbs.

[0030] Fatigue detection can first acquire one or more dimensions of human characteristic signals, analyze these signals, and determine the fatigue detection result based on the analysis results. This application does not limit the human characteristic signals; they can include human-related physiological signals (e.g., heart rate, blood pressure, nerve electrical activity, muscle electrical activity) and / or human-related mechanical signals (e.g., skin deformation, muscle deformation, vascular pulsation). Correspondingly, appropriate sensing nodes can be set for different physiological / mechanical signals. For example, if the human characteristic signals include heart rate and muscle deformation, sensing nodes for sensing heart rate and sensing nodes for highly sensitive sensing of muscle deformation can be set in the textile body.

[0031] In practice, based on one or more factors such as cost, the part of the user wearing the device, and user needs, appropriate types of sensor nodes can be set in the textile body.

[0032] The signal processing component can receive and process human characteristic signals collected by various sensing nodes.

[0033] Step 102: Determine the fatigue level based on human body characteristic signals; Since human body feature signals are fatigue-related, when fatigue detection is performed on the same user and the same body part, the human body feature signals change with the degree of fatigue in that area. Therefore, signal processing components can be used to process the human body feature signals to analyze the user's current fatigue level and / or predict the user's future fatigue level. Specifically, according to actual needs, fatigue levels can be divided into multiple fatigue state levels, and different levels can be processed differently in subsequent steps to achieve refined processing of fatigue state levels.

[0034] Step 103: Send a physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component to instruct the physiotherapy execution component to execute the physiotherapy strategy corresponding to the physiotherapy control signal.

[0035] The signal processing component can generate a corresponding physical therapy control signal based on the fatigue level. The signal processing component sends the physical therapy control signal to the physical therapy execution component, which responds to the signal and executes a physical therapy strategy on the user at its location. This strategy includes physical intervention for a certain duration targeting the fatigued areas of the user.

[0036] The embodiments of this application do not limit the physiotherapy methods corresponding to the physiotherapy strategies, and may include one or more physiotherapy methods.

[0037] In this embodiment, the wearable physiotherapy device is equipped with an interconnected signal processing component and a textile body. The textile body is equipped with a sensing node and a physiotherapy execution component. The signal processing component acquires human characteristic signals collected by the sensing node; determines the fatigue level based on the human characteristic signals; and sends a physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component to instruct the physiotherapy execution component to execute the physiotherapy strategy corresponding to the physiotherapy control signal. This enables the wearable physiotherapy device to achieve a fully automatic closed loop from sensing human characteristic signals to obtaining fatigue level perception based on human characteristic signal analysis, and then executing the corresponding physiotherapy strategy (physical intervention) based on the fatigue level, thus solving the functional separation problem of fatigue detection and physical intervention in the prior art.

[0038] In some implementations of this application, multiple therapeutic areas can be provided for the parts of the user wearing the textile body, and the sensing nodes are located in the therapeutic areas of the textile body. Correspondingly, the therapeutic execution components in each therapeutic area correspond to the sensing nodes in the same therapeutic area.

[0039] As an example, the textile body can be obtained by using weaving equipment based on appropriate weaving techniques (e.g., computerized horizontal weaving) to weave three types of yarn—sensing yarn, drive yarn, and comfort yarn—in different areas as needed. In the therapeutic area, the sensing and drive yarns are woven synchronously and electrically interconnected, forming a sensing-drive unit; in the non-therapeutic area, comfort yarn is mainly used to ensure comfort. All electrical connections are naturally formed during weaving, ultimately resulting in a seamless, soft, and fully integrated smart garment—the textile body.

[0040] As another example, the textile body can be obtained by weaving magnetic sensing fibers (e.g., magnetic nanocomposite fibers) and conductive yarns (e.g., silver-plated conductive yarns) as warp and weft yarns at a 1:1 ratio into a mesh-like fabric base structure. Each interlacing point forms a sensing node. The resistance of the magnetic sensing fibers changes sensitively with fabric strain (caused by muscle activity), while the conductive yarns are responsible for signal transmission. The magnetic sensing fibers and conductive yarns converge at the edge of the therapeutic area and connect to a signal processing component, enabling array-style signal reading of the sensing network composed of the individual sensing nodes.

[0041] Reference Figure 2 This illustration shows a partial schematic diagram of a textile body provided in an embodiment of this application. The therapeutic area of ​​the textile body includes interwoven magnetic sensing fibers. Figure 2 (solid thread) and conductive yarn ( Figure 2 (Solid and dashed lines in the middle), the sensing nodes are formed by the interweaving of magnetic sensing fibers and conductive yarns; the human body characteristic signal is a strain signal sequence, which is composed of the strain signals generated by each sensing node.

[0042] Reference Figure 3 This image shows a cross-sectional view of a physiotherapy area provided in an embodiment of this application. As an example, at the sensor node location of the physiotherapy area (e.g., ... Figure 2 (As shown at point A), the textile body, from its outer surface to its inner surface, includes a protective layer, magnetic sensing fibers, conductive yarns, an insulating layer, and a skin-friendly layer (the physiotherapy execution component is not shown). In non-physiotherapy areas, the textile body, from its outer surface to its inner surface, consists of a protective layer and a skin-friendly layer, respectively.

[0043] In another example, the conductive yarn in the treatment area can be located on top of the magnetic sensing fiber.

[0044] Before step 101, the training and deployment of the artificial intelligence model are required, including: (1) Training phase: using a pre-collected and expert-labeled "strain signal-fatigue state level" paired dataset as training data, the pre-set artificial intelligence model is trained in a supervised manner. The cross-entropy loss function and model optimizer are used for iterative optimization until the model reaches a stable accuracy on the validation set. (2) Deployment phase: the artificial intelligence model that has completed iterative training is converted into a lightweight format (such as TensorFlow Lite) and deployed on the processor of the signal processing component (e.g., deployed on the embedded processor contained in the signal processing component) to realize real-time and offline strain signal analysis and fatigue state level identification.

[0045] In a practical implementation, the processor of the signal processing component can be a digital signal processor or a microcontroller.

[0046] Through the training and deployment of the above-mentioned artificial intelligence model, the signal processing component can process the strain sequence signal. Step 102 includes: inputting the strain signal sequence into the trained artificial intelligence model to instruct the artificial intelligence model to determine the fatigue state level of the physiotherapy area based on the strain signal sequence, and to determine the physiotherapy control signal corresponding to the fatigue state level; wherein, the fatigue state level includes a non-fatigue state, and one or more fatigue levels (e.g., mild fatigue, moderate fatigue, severe fatigue) corresponding to the fatigue state.

[0047] Different strain signals can correspond to different physiotherapy areas. The strain signal sequence is input into a trained artificial intelligence (AI) model, which can determine the fatigue level corresponding to the strain signal in the same physiotherapy area. For example, the strain signal sequence determines physiotherapy area 1 as non-fatigue, physiotherapy area 2 as mild fatigue, and physiotherapy area 3 as severe fatigue. After determining the fatigue level of each physiotherapy area, the AI ​​model can further determine the corresponding physiotherapy control signal for each fatigue level. Continuing the example, for physiotherapy areas 2 and 3 corresponding to fatigue, it further determines physiotherapy control signal 1 for physiotherapy area 2 and physiotherapy control signal 2 for physiotherapy area 3. In this way, when a physiotherapy area is determined to correspond to a fatigue state, it can be determined that the user's physiotherapy area is in a fatigue state. Based on the fatigue level of that physiotherapy area, the corresponding physiotherapy control signal is determined, and step 103 enables the wearable physiotherapy device to automatically execute different physiotherapy strategies according to the fatigue level.

[0048] In some implementations of this application, inputting the strain signal sequence into a trained artificial intelligence model includes: preprocessing the strain signal sequence to obtain a preprocessed signal; slicing the preprocessed signal according to a certain time window to obtain samples to be processed corresponding to each time window; and inputting the samples to be processed into the trained artificial intelligence model. The preprocessing includes filtering and / or normalizing the amplitude of each strain signal in the strain signal sequence.

[0049] To improve the stability and accuracy of fatigue state level determination using artificial intelligence models to process strain signal sequences, preprocessing can be performed on the strain signal sequences before sending them to the AI ​​model to obtain preprocessed signals. Preprocessing can include at least one of filtering and amplitude normalization. Since different strain signals in the strain signal sequence are sensed by different sensing nodes, amplitude normalization normalizes the amplitudes of the strain signals corresponding to different sensing nodes to a certain range.

[0050] Since the strain signal sequence can be a real-time acquired data stream, the artificial intelligence model needs to continuously sense the strain signal at the sensing node and predict the fatigue state level from the strain signal sequence. Therefore, the preprocessed signal needs to be sliced ​​according to a certain time window to obtain the sample to be processed. The artificial intelligence model uses one or more samples to be processed to make predictions and obtain the fatigue state level.

[0051] In this embodiment, the time window can be a fixed window or a sliding window, for example, the size of the time window is 2 seconds. This embodiment can divide the preprocessed signal into non-overlapping samples to be processed using a fixed window method, or it can divide the preprocessed signal into partially overlapping samples to be processed using a sliding window method.

[0052] The following is combined with Figure 4 The process of determining the fatigue state level of each physiotherapy area based on the strain signal sequence is further explained. (Refer to...) Figure 4 This illustration shows a schematic diagram of strain signal sequence processing provided in an embodiment of this application. In this embodiment, an artificial intelligence model is used to determine the fatigue state level, which is divided into three stages, including: (1) Data Input and Preprocessing. This includes filtering, normalization, and segmentation. The original strain signal sequence sent by the textile body is low-pass filtered to remove high-frequency noise and motion artifacts. The amplitude of the strain signal is normalized to the [0,1] interval according to different sensing nodes to eliminate individual user differences and sensing node sensitivity differences, resulting in a preprocessed signal. The continuous preprocessed signal is segmented into segments according to a fixed time window (e.g., 2 seconds). Each segment of the strain signal is used as one input data for the artificial intelligence model, i.e., the sample to be processed.

[0053] (2) Feature extraction: The artificial intelligence model is equipped with a one-dimensional convolutional neural network to automatically learn and extract abstract features in the signal that are highly correlated with muscle fatigue, replacing the complex process of manually designing features (such as mean, variance, and spectral energy) in traditional methods. The one-dimensional convolutional neural network consists of several one-dimensional convolutional units stacked in sequence. A one-dimensional convolutional unit includes a convolutional layer and a pooling layer.

[0054] Convolutional layers are used to slide a one-dimensional convolutional kernel across the time dimension of a strain signal, automatically extracting temporal features from local to global perspectives (such as oscillation patterns at specific frequencies and trends in signal amplitude variation). As an example, the first convolutional kernel can be set with a small width to capture details, with subsequent layers gradually increasing the width.

[0055] Pooling layers are used to reduce data dimensionality and computational cost by employing specific values ​​(such as max pooling values) while preserving the most salient features, thereby enhancing the model's robustness to small time shifts.

[0056] (3) State classification and output: The multidimensional feature map output by the convolutional neural network is converted into a one-dimensional vector and input into several (e.g., 1 or 2) fully connected layers. The fully connected layers perform nonlinear combination and mapping of the extracted features and input them into the classifier. For each physiotherapy area, the classifier outputs the probability distribution of each fatigue state level and takes the one with the highest probability as the final output.

[0057] By following the above three steps, the fatigue level corresponding to each physiotherapy area can be obtained.

[0058] In some implementations of this application, the signal processing component includes a signal conditioning circuit connected to the sensing node. The signal conditioning circuit can be integrated into a low-power design, packaged on a thin, flexible printed circuit board, for gating, amplifying, filtering, digitizing, and initially packaging multi-channel sensing signals. This flexible circuit board is electrically connected to the conductive yarn bus of the textile body via conductive buttons, anisotropic conductive adhesive, or direct soldering. Because the printed circuit board is flexible, it can adapt to the deformation scenarios of wearable physiotherapy devices.

[0059] Reference Figure 5 The diagram shows a signal conditioning circuit according to an embodiment of this application. The signal conditioning circuit may include: an input protection unit, a multiplexing unit, an adjustable gain amplifier, an anti-aliasing filter, a sample and hold unit, an analog-to-digital converter, a conditioning controller, a clock unit, and a reference voltage source.

[0060] The input protection unit consists of a series of transient voltage suppression diodes and RC networks. Each sensor signal channel is equipped with an independent set, and each sensor signal channel is used to transmit the human body characteristic signal output by one sensor node. The input protection unit is used to absorb transient overvoltages such as electrostatic pulses from the human body or environment, protecting the subsequent precision analog circuits from damage, and also serves as a current limiter.

[0061] The multiplexing unit employs an analog multiplexer chip (such as 16:1 or 32:1) with low on-resistance and high channel count. Multiple input channels of the multiplexing unit are connected to protected sensor signal channels, and the output is connected to subsequent amplification and filtering links. The address selection pin is controlled by the general-purpose input / output port of the conditioning controller. Under the control of the conditioning controller, the multiplexing unit sequentially switches multiple distributed sensor nodes to a single post-processing path in a preset order (such as line-by-line scanning), achieving time-division multiplexing measurement and greatly simplifying system complexity.

[0062] Instrumentation Amplifier. The instrumentation amplifier uses a high input impedance, low noise, and high common-mode rejection ratio (CMRR) preamplifier. The high input impedance ensures that excessive current is not applied to the high-resistance sensing fiber, thus preventing interference with measurements; the high CMRR effectively suppresses common-mode interference from the human body.

[0063] The adjustable gain amplifier is connected after the instrumentation amplifier. The gain factor of the adjustable gain amplifier (e.g., 1, 2, 4, 8…128 times) is dynamically configured by the conditioning controller via SPI or I2C bus. When the signal amplitude is detected to be too small, the gain is automatically increased; when the amplitude saturates, the gain is decreased, realizing adaptive range adjustment and ensuring that the dynamic range of the analog-to-digital converter is fully utilized.

[0064] The anti-aliasing filter employs a second- or third-order active low-pass filter whose cutoff frequency is strictly set according to the highest effective frequency of the signal (such as the muscle tremor frequency) and the sampling frequency of the subsequent analog-to-digital converter, in order to eliminate high-frequency noise and prevent sampling aliasing.

[0065] The sample-and-hold unit quickly captures and "freezes" the current analog voltage value before the analog-to-digital converter (ADC) starts operating, ensuring the stability of the signal input to the ADC throughout the conversion cycle. Its sampling timing is synchronized with the conditioning controller using the same clock, guaranteeing time alignment of multi-channel data.

[0066] The analog-to-digital converter (ADC) employs a 16-bit or 24-bit high-resolution, low-power successive approximation type. Its reference voltage is provided by a low-temperature-drift, high-precision bandgap reference voltage source, which is the cornerstone of ensuring the measurement accuracy of the entire system. The ADC's conversion start signal is also precisely controlled by a conditioning controller.

[0067] The conditioning controller is a low-power microcontroller used to schedule the various components of the entire conditioning signal assembly. The conditioning controller generates the channel selection address for the multiplexer, the gain control word for the PGA, the sampling pulse for the S / H, and the start conversion signal for the analog-to-digital converter, and reads the conversion result.

[0068] The clock unit provides a system-wide time reference for synchronization, either through an external temperature-compensated crystal oscillator or the high-precision clock within the conditioning controller, ensuring the accuracy of the sampling interval and timing. The conditioning controller, in conjunction with the time reference provided by the clock unit, determines the control logic and outputs corresponding control signals to the multiplexing unit, sample-and-hold unit, and digital-to-analog converter.

[0069] Taking the reading of an 8x8 sensor network (8*8 sensor nodes) as an example, the processing procedure of the conditioning signal component will be further explained. The processing procedure of the conditioning signal component is as follows: 1. Initialization: The wearable physiotherapy device is powered on, the conditioning controller initializes all I / O, configures the analog-to-digital converter parameters, sets the initial PGA gain, and sets the scan sequence.

[0070] 2. Channel selection: The conditioning controller sets the multiplexer address to the channel corresponding to the first sensor node (e.g., row 1, column 1).

[0071] 3. Signal stabilization: Wait a short time for the analog signal to stabilize after gating.

[0072] 4. Gain Adjustment: The conditioning controller reads the analog-to-digital converter data once and determines the amplitude. If it is not in the optimal range [ZYZL1.1], the PGA gain is adjusted and the signal is resampled until the signal amplitude is ideal.

[0073] 5. Synchronous Sampling and Conversion: The conditioning controller sends a synchronization pulse, which simultaneously triggers the S / H circuit to sample the current channel signal and start the analog-to-digital converter to begin conversion.

[0074] 6. Data Reading and Packaging: After the conversion is completed, the conditioning controller reads the digital result [ZYZL2.1] of the analog-to-digital converter through the SPI bus and packs the data with the current channel number (coordinate information).

[0075] 7. Looping and Transmission: The conditioning controller updates the multiplexer address to the next channel, repeating steps 2-6 until all preset nodes have been scanned. The packaged data frame (containing the coordinates and sensor data of 64 nodes) is sent to the back-end AI analysis engine (intelligent processing submodule) via a high-speed serial port (such as UART or USB).

[0076] 8. Low power management: During non-scanning periods, the conditioning controller can control the power supply of analog circuits such as PGA and analog-to-digital converter to shut down and enter sleep mode until it is woken up by the timer in the next sampling cycle.

[0077] The physiotherapy actuator is used to decouple the low-voltage physiotherapy control signal from the high-voltage drive power, ensuring the safety of the signal processing component, and enabling independent, addressable control of each actuator to achieve highly precise physiotherapy.

[0078] In some implementations of this application, the physiotherapy execution component includes a mechanical drive unit; the physiotherapy control signal includes a first control command; the mechanical drive unit is used to periodically contract in response to the first control command to apply mechanical pressure.

[0079] The physiotherapy execution component includes a mechanical drive unit for converting electrical energy into mechanical force or deformation.

[0080] The therapeutic drive section is the core for simulating massage movements. Its material can be nickel-titanium based shape memory alloy (SMA) wire, with a diameter between 50-150 micrometers, woven or embroidered into the textile body in spiral, straight, or wavy patterns. SMA wire possesses a "shape memory effect." When a specific current is applied to the SMA wire, it heats up to above its austenitic phase transformation point due to Joule heating, altering the internal crystal structure of the alloy and causing rapid and significant macroscopic contraction (strain can reach 5-8%). After the current is cut off and it cools, it returns to its original length under the restoring force of the textile body.

[0081] The first control command can be a pulse width modulation signal. By controlling the on / off state of the SMA wire current through pulse width modulation, the contraction force, frequency and rhythm can be precisely controlled to simulate various massage modes such as pressing, kneading and vibration. This enables the mechanical drive unit to respond to the first control command and contract periodically to apply mechanical pressure and simulate massage actions.

[0082] In some implementations of the embodiments of this application, the physiotherapy execution component includes a thermal energy unit; the physiotherapy control signal includes a second control command; and the thermal energy unit is used to release thermal energy in response to the second control command.

[0083] The physiotherapy execution component includes a thermal unit that converts electrical energy into heat energy to achieve precise and safe localized heat application.

[0084] The thermal section can use flexible conductive yarns with high resistivity, such as carbon fiber yarns or carbon-plated / alloy-plated nylon yarns, woven in meandering, grid, or specific patterns into the therapeutic area of ​​the textile body.

[0085] When an electric current passes through the resistive yarn, electrical energy is converted into heat energy based on the Joule heating effect, causing the temperature of the resistive yarn and the surrounding area of ​​the textile body to rise.

[0086] The thermal unit can achieve precise temperature control (e.g., linear adjustment between 36°C and 45°C) by adjusting the current in the resistor wiring (including the presence or absence of current) in response to a second control command.

[0087] As an example, distributed miniature temperature sensors (such as negative temperature coefficient thermistors) can be integrated into the textile body to form a real-time temperature feedback closed loop, preventing overheating and ensuring safety.

[0088] In a specific implementation, the physiotherapy execution component also includes a flexible drive circuit. The input end of the flexible drive circuit is connected to the processor of the power supply and signal processing component, and the output end is connected to the mechanical drive unit and the thermal energy unit respectively. It is used to convert the low-voltage physiotherapy control signal into a controlled drive current for the mechanical drive unit and the thermal energy unit.

[0089] The flexible drive circuit consists of flexible printed circuits / conductive circuits attached to or woven onto the textile body, including: The control bus is used to receive low-voltage digital control signals (physiotherapy control signals) from the signal processing components.

[0090] A power bus is used to draw the high-power current required to drive the device from the power source of the physiotherapy wearable device, such as a flexible battery.

[0091] The drive unit, a miniature switching circuit (such as a MOSFET array) located next to the mechanical drive section and the thermal energy section, is used to connect or disconnect the circuit between the SMA filament or heating yarn and the power supply according to the physiotherapy control signal.

[0092] In some implementations of this application, after sending the physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component, the application further includes: acquiring a physiotherapy feedback signal; the physiotherapy feedback signal includes re-acquired human characteristic signals; determining the expected physiotherapy effect corresponding to the physiotherapy strategy; and adjusting the physiotherapy control signal if the physiotherapy feedback signal does not match the expected physiotherapy effect.

[0093] Following step 103, the signal processing component continuously acquires physiotherapy feedback signals. This feedback information may include human characteristic signals re-collected by sensor nodes in the physiotherapy area previously identified as being in a state of fatigue. The physiotherapy feedback signals are used to analyze whether the physiotherapy strategy executed by the physiotherapy execution component achieves the expected therapeutic effect. If the feedback signals do not match the expected effect, it indicates that the current physiotherapy strategy cannot achieve the desired result, and the corresponding physiotherapy control signals need to be adjusted. This provides a short-term closed-loop and control feedback mechanism based on physiotherapy feedback signals.

[0094] For example, after the physiotherapy execution component applies a corresponding physiotherapy strategy to a specific treatment area, the human characteristic signals of that area are used as physiotherapy feedback information. This feedback information is used to monitor changes in muscle tension, stiffness, or micro-vibrations in that area. If a decrease in the spectral components of muscle stiffness is detected, or if the human characteristic signals change in a relaxed state, the physiotherapy is deemed effective. If there is no change in the human characteristic signals or abnormal high-frequency vibrations appear (potentially indicating intolerance), the effect is deemed poor (not matching the expected physiotherapy effect). The artificial intelligence model can adjust the physiotherapy control signals in real time (e.g., adjusting the current magnitude or operating duration of the mechanical control unit in the next cycle) when the effect is determined to be poor.

[0095] In some implementations of the embodiments of this application, the embodiments of this application further include: recording personalized training samples; the personalized training samples include corresponding physiotherapy feedback signals and adjusted physiotherapy control signals; and using the personalized training samples to train the artificial intelligence model so as to instruct the artificial intelligence model to generate personalized sub-models.

[0096] In this embodiment, after step 103, corresponding physiotherapy feedback signals and adjusted physiotherapy control signals can be recorded as personalized training samples. After completing the physiotherapy strategy, the artificial intelligence model is trained using the personalized training samples, thereby generating personalized sub-models in the artificial intelligence model. This makes the subsequent physiotherapy strategy generation more accurate and efficient, enabling the physiotherapy wearable device to provide more suitable personalized physiotherapy strategies for different users, thus providing a long-term model-optimized physiotherapy feedback mechanism.

[0097] As an example, training samples can also be output to an artificial intelligence model in real time, enabling the model to fine-tune the physiotherapy control signals online, thus adjusting the signals to better suit the current user.

[0098] In some implementations of the embodiments of this application, the wearable physiotherapy device is further provided with at least one physiological signal sensor; the physiotherapy feedback signal also includes physiological signals collected by the physiological signal sensor; the physiological signals include at least one of skin temperature signal and heart rate signal.

[0099] Among the three physiotherapy feedback mechanisms mentioned above, strain signals and physiological signals can also be integrated. Physiological signal sensors can be installed on the textile body to collect the current user's physiological signals.

[0100] For example, a physiological signal sensor can be a miniature temperature sensor used to sense the current skin temperature of the user and output a corresponding skin temperature signal, thereby verifying whether the thermal energy unit has reached and maintained the expected temperature; a physiological signal sensor can also be an electrocardiogram sensor (e.g., electrocardiogram textile electrode), which is used to output a heart rate signal. Through the heart rate signal, it is possible to assess whether the physical therapy has made the user generally more relaxed (HRV increased) or caused stress (heart rate increased).

[0101] By fusing strain signals (mechanical signals) and physiological signals, a multimodal signal fusion feedback mechanism is provided to adjust physiotherapy control signals and establish personalized models. For example, after the physiotherapy execution component executes a corresponding physiotherapy strategy on a certain physiotherapy area, the strain and physiological signals of that area are used as physiotherapy feedback information to conduct a more comprehensive evaluation of the user and the current physiotherapy strategy, thereby enabling a more accurate adjustment of physiotherapy control signals.

[0102] Reference Figure 6 The diagram illustrates a physiotherapy feedback mechanism provided in an embodiment of this application. In this embodiment, after a physiotherapy control signal is sent to the physiotherapy execution component, the signal processing component acquires the physiotherapy feedback signal and executes at least one of the following: a short-term closed-loop and control feedback mechanism, a long-term model optimization physiotherapy feedback mechanism, and a multimodal signal fusion feedback mechanism.

[0103] By performing the various steps of the embodiments of this application, wearable physiotherapy devices can be widely applied in multiple fields, including but not limited to the following: sports rehabilitation, real-time monitoring of athletes' muscle load, and automatic relaxation and recovery of fatigued muscle groups after exercise; daily health management, relieving chronic strain and stiffness in the neck, shoulders, waist, and back for office workers who sit for long periods; medical assistance, as a physical therapy aid for the long-term management of chronic pain (such as arthritis and myofascitis); rehabilitation nursing, helping stroke or postoperative patients perform timed and quantitative passive joint movements and muscle stimulation; and specific occupational fields, providing immediate fatigue relief for workers who need to maintain specific postures for long periods (such as drivers and craftsmen). It should be noted that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0104] Reference Figure 7 The diagram illustrates a physiotherapy control device according to an embodiment of this application, which may specifically include a human feature signal acquisition module 701, a human feature signal analysis module 702, and a physiotherapy control module 703, wherein: The human feature signal acquisition module 701 is used to acquire human feature signals collected by the sensor node.

[0105] The human body feature signal analysis module 702 is used to determine the fatigue level based on human body feature signals.

[0106] The physiotherapy control module 703 is used to send a physiotherapy control signal corresponding to the fatigue state level to the physiotherapy execution component, so as to instruct the physiotherapy execution component to execute the physiotherapy strategy corresponding to the physiotherapy control signal.

[0107] In some implementations of this application, the sensing node is located in the therapeutic area of ​​the textile body, and the sensing node is formed by interlacing magnetic sensing fibers and conductive yarns; the human body feature signal is a strain signal sequence; module 702 includes: The strain signal sequence input submodule is used to input the strain signal sequence into the trained artificial intelligence model, so as to instruct the artificial intelligence model to determine the fatigue state level of the physiotherapy area based on the strain signal sequence, and to determine the physiotherapy control signal corresponding to the fatigue state level. The fatigue state level includes the non-fatigue state and one or more fatigue levels corresponding to the fatigue state.

[0108] In some implementations of the embodiments of this application, the strain signal sequence input submodule includes: The signal preprocessing unit is used to preprocess the strain signal sequence to obtain the preprocessed signal; The signal slicing unit is used to slice the preprocessed signal according to a certain time window to obtain the sample to be processed corresponding to each time window. The signal input unit is used to input the sample to be processed into the trained artificial intelligence model; the preprocessing includes filtering and / or normalizing the amplitude of each strain signal in the strain signal sequence.

[0109] In some implementations of this application, after sending a physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component, the device further includes: The physiotherapy feedback signal acquisition module is used to acquire physiotherapy feedback signals; the physiotherapy feedback signals include re-acquired human characteristic signals; The expected physiotherapy effect determination module is used to determine the expected physiotherapy effect corresponding to the physiotherapy strategy; The physiotherapy control signal adjustment module is used to adjust the physiotherapy control signal when the physiotherapy feedback signal does not match the expected physiotherapy effect.

[0110] In some implementations of the embodiments of this application, the apparatus further includes: The personalized training sample recording module is used to record personalized training samples; the personalized training samples include corresponding physiotherapy feedback signals and adjusted physiotherapy control signals. The personalized sub-model training module is used to train the artificial intelligence model with personalized training samples, so as to instruct the artificial intelligence model to generate personalized sub-models.

[0111] In some implementations of the embodiments of this application, the wearable physiotherapy device is further provided with at least one physiological signal sensor; the physiotherapy feedback signal also includes physiological signals collected by the physiological signal sensor; the physiological signals include at least one of skin temperature signal and heart rate signal.

[0112] In some implementations of this application, the physiotherapy execution component includes a mechanical drive unit; the physiotherapy control signal includes a first control command; the mechanical drive unit is used to periodically contract in response to the first control command to apply mechanical pressure.

[0113] In some implementations of the embodiments of this application, the physiotherapy execution component includes a thermal energy unit; the physiotherapy control signal includes a second control command; and the thermal energy unit is used to release thermal energy in response to the second control command.

[0114] This application provides a physiotherapy control device, which can be used to implement the steps in the aforementioned method embodiments.

[0115] As the apparatus embodiments are basically similar to the method embodiments, they are described in a relatively simple manner. For relevant details, please refer to the description in the method embodiment section.

[0116] This application also provides a wearable physiotherapy device, including: a signal processing component and a textile body. The signal processing component is provided with a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the wearable physiotherapy device implements the physiotherapy control method as described in the first aspect above.

[0117] Reference Figure 8 The diagram illustrates an application example of a wearable physiotherapy device provided in an embodiment of this application, including a textile body 801 and an electrical mechanism 802, the electrical mechanism 802 including a signal processing component.

[0118] This application also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the physiotherapy control methods as described in the foregoing embodiments.

[0119] This application also discloses a computer program product, including a computer program, which, when run, causes the physiotherapy control methods as described in the foregoing embodiments to be executed.

[0120] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A physiotherapy control method, characterized in that, A signal processing component for wearable physiotherapy devices, the signal processing component being connected to a textile body; the textile body being provided with sensing nodes and a physiotherapy execution component; the method comprising: Acquire human feature signals collected by the sensing nodes; The fatigue level is determined based on the aforementioned human characteristic signals; Send a physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component to instruct the physiotherapy execution component to execute the physiotherapy strategy corresponding to the physiotherapy control signal.

2. The method according to claim 1, characterized in that, The sensing node is located in the therapeutic area of ​​the textile body, and the sensing node is formed by interlacing magnetic sensing fibers and conductive yarns; the human body characteristic signal is a strain signal sequence; determining the fatigue state level based on the human body characteristic signal includes: The strain signal sequence is input into a trained artificial intelligence model to instruct the model to determine the fatigue level of the physiotherapy area based on the strain signal sequence, and to determine the physiotherapy control signal corresponding to the fatigue level. The fatigue state level includes a non-fatigue state and one or more fatigue levels corresponding to the fatigue state.

3. The method according to claim 2, characterized in that, The step of inputting the strain signal sequence into the trained artificial intelligence model includes: The strain signal sequence is preprocessed to obtain a preprocessed signal; The preprocessed signal is sliced ​​according to a certain time window to obtain the sample to be processed corresponding to each time window. The sample to be processed is input into the trained artificial intelligence model; The preprocessing includes filtering and / or normalizing the amplitude of each strain signal in the strain signal sequence.

4. The method according to claim 2, characterized in that, After sending the physiotherapy control signal corresponding to the fatigue level to the physiotherapy execution component, the method further includes: Acquire physiotherapy feedback signals; the physiotherapy feedback signals include re-acquired human characteristic signals; Determine the expected physical therapy effect corresponding to the described physical therapy strategy; If the physiotherapy feedback signal does not match the expected physiotherapy effect, the physiotherapy control signal is adjusted.

5. The method according to claim 4, characterized in that, The method further includes: Record personalized training samples; the personalized training samples include corresponding physiotherapy feedback signals and adjusted physiotherapy control signals; The artificial intelligence model is trained using the personalized training samples to instruct the artificial intelligence model to generate personalized sub-models.

6. The method according to claim 4, characterized in that, The wearable physiotherapy device is also equipped with at least one physiological signal sensor; the physiotherapy feedback signal also includes physiological signals collected by the physiological signal sensor; the physiological signals include at least one of skin temperature signal and heart rate signal.

7. The method according to claim 1, characterized in that, The physiotherapy execution component includes a mechanical drive unit; the physiotherapy control signal includes a first control command; the mechanical drive unit is used to periodically contract in response to the first control command to apply mechanical pressure.

8. The method according to claim 1 or 7, characterized in that, The physiotherapy execution component includes a thermal energy unit; the physiotherapy control signal includes a second control command; the thermal energy unit is used to release thermal energy in response to the second control command.

9. A wearable physiotherapy device, characterized in that, include: The signal processing component includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it causes the wearable physiotherapy device to perform the method as described in any one of claims 1-8.

10. A computer program product, characterized in that, Includes a computer program, which, when run, causes the method as described in any one of claims 1-8 to be performed.

Images