Intelligent insole for gait analysis
By combining acceleration sensors and electrical impedance scanning modules in gait analysis insoles, using electrical impedance imaging technology and machine learning algorithms, the problem of gait analysis results in the existing technology is solved, and the step of higher accuracy and reliability is achieved Status information collection.
Patent Information
- Application Number
- CN202411278923.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-12
- Publication Date
- 2024-12-06
AI Technical Summary
The existing gait analysis insole sensors have a single type, making it difficult to fully collect gait information, resulting in one-sided analysis results and the flexible circuit is easily damaged.
The acceleration sensor and electrical impedance scanning module are used to obtain pressure change images and dynamic acceleration changes through electrical impedance imaging technology, and use machine learning algorithms to perform data analysis to provide more comprehensive gait information.
Improves the accuracy and reliability of gait analysis, avoids damage to flexible circuits, and provides more accurate and comprehensive general gait information.
Smart Images

Figure CN119073701A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent insoles, and in particular to an intelligent insole for gait analysis. Background Art
[0002] Most of the existing solutions transmit the plantar pressure information by distributing the pressure sensor on the surface of the insole, and usually lay the flexible PCB on the insole and cover the polyvinyl chloride layer as protection. However, the flexible PCB is easily damaged by the movement of the circuit, resulting in the error of gait analysis. In addition, the force sensor can only transmit the plantar pressure information, and the data obtained by this solution is relatively single, and it is difficult to fully parse out the complete gait information, resulting in a relatively one-sided analysis result. For this reason, the present invention adopts the impedance scanning image detection technology, which can obtain the pressure change image by the impedance imaging method, and at the same time combines the acceleration sensor to collect the dynamic acceleration change of the foot during the gait process. Through the combination of these multi-dimensional data, the intelligent insole can provide a more comprehensive and accurate gait analysis, avoiding the defects of the flexible circuit being easily damaged and the one-sided data in the traditional solution, and greatly improving the accuracy and reliability of gait analysis. Summary of the invention
[0003] Based on this, the present invention provides a smart insole for gait analysis to solve the problem that existing gait analysis insole sensors are of a single type and difficult to fully collect gait information.
[0004] The present invention provides a smart insole for gait analysis, comprising:
[0005] Insole body, sensor element, main control module, Bluetooth communication module and host computer;
[0006] The sensor element, the main control module and the Bluetooth communication module are all arranged inside the insole body. The main control module receives the data information collected by the sensor element, and the main control module sends the data information to the host computer through the Bluetooth communication module. The host computer receives the data information and processes the data information, and uses a machine learning classification algorithm to analyze the data information after the data processing to obtain the current gait information.
[0007] The sensor element includes an acceleration acquisition module and an electrical impedance scanning module;
[0008] The acceleration acquisition module is used to acquire acceleration changes caused by gait changes;
[0009] The electrical impedance scanning module is used to collect electrical impedance changes of the insole caused by changes in walking posture and gait.
[0010] The acceleration acquisition module includes an MPU-6050 chip U13, a pull-up resistor R21, a pull-up resistor R19, a current limiting resistor R20, and a current limiting resistor R20, and the main control module includes a main control chip U1;
[0011] Pin No. 13 and Pin No. 8 of the MPU-6050 chip U13 are connected to a 3.3V power supply, Pin No. 18 of the MPU-6050 chip U13 is grounded, a 0.1uF capacitor is connected between Pin No. 1 and Pin No. 8 of the MPU-6050 chip U13, Pin No. 23 of the MPU-6050 chip U13 is connected to a 3.3V power supply through the pull-up resistor R21, Pin No. 23 of the MPU-6050 chip U13 is also electrically connected to the current limiting resistor R20, Pin No. 24 of the MPU-6050 chip U13 is connected to a 3.3V power supply through the pull-up resistor R19, Pin No. 24 of the MPU-6050 chip U13 is also electrically connected to the current limiting resistor R18, Pin No. 23 and Pin No. 24 of the MPU-6050 chip U13 are electrically connected to Pin No. 58 and Pin No. 59 of the main control chip U1 respectively.
[0012] The electrical impedance scanning module includes an AD5933 chip U7 and an AD820 operational amplifier chip U12;
[0013] Pin 7 of the AD820 op amp chip U12 is connected to a 5V power supply, and pin 7 of the AD820 op amp chip U12 is also grounded through a 0.1uF capacitor. Pin 6 of the AD820 op amp chip U12 is connected to pin 2 of the pin header H11, and pin 1 of the pin header H11 is electrically connected to pin 1 of the pin header H10, and pin 1 of the pin header H10 is also connected to a voltage output terminal VOUT. Pin 2 of the AD820 op amp chip U12 is connected to pin 3 of the pin header H10 through a resistor R14, and pins 2 and 4 of the pin header H10 are connected to pin 6 of the AD5933 chip U7. Pin, pin 5 of the AD5933 chip U7 is connected to the voltage input terminal VIN, pin 4 of the AD5933 chip U7 is connected to the voltage input terminal VIN through a resistor R11, pin 15 of the AD5933 chip U7 is electrically connected to pin 59 of the main control chip U1 through a resistor R13, pin 15 of the AD5933 chip U7 is also connected to a 5V power supply through a pull-up resistor R6, pin 16 of the AD5933 chip U7 is electrically connected to pin 58 of the main control chip U1 through a resistor R12, and pin 16 of the AD5933 chip U7 is also connected to a 5V power supply through a pull-up resistor R5.
[0014] include,
[0015] The GPIO pins of the main control chip U1 are respectively connected to two ADG1608 multiplexer switches, and the PA0 pin and PA4 pin of the main control chip U1 are used to enable the strobe switch module.
[0016] Also includes,
[0017] When the PA0 pin of the main control chip U1 inputs a high level, the multiplexing switch U10 is enabled; when the PA4 pin of the main control chip U1 inputs a high level, the multiplexing switch U11 is enabled; when the PA0 pin and PA4 pin of the main control chip U1 input a low level, the multiplexing switch U10 and the multiplexing switch U11 are disabled.
[0018] It also includes a power management module, which supplies power to the acceleration acquisition module, the impedance scanning module and the main control module.
[0019] The host computer includes a Bluetooth communication module and a data processing module. The host computer receives the data information collected by the sensor element through the Bluetooth communication module. The data processing module displays the data information through acceleration data and impedance scanning images according to the built-in machine learning algorithm, and finally obtains the current gait information.
[0020] The insole body includes an upper surface suede cloth T1, a filling layer, an intelligent gait analysis mechanism T5, and a bottom support layer carbon fiber reinforced polymer T6. A supporting structure thermoplastic polyurethane lattice T2 is arranged in the filling layer, and a hydrogel T4 is filled inside the filling layer. A sensor element placement grid T3 is also arranged in the filling layer. The upper surface suede cloth T1, the filling layer, the intelligent gait analysis mechanism T5, and the bottom support layer carbon fiber reinforced polymer T6 are arranged in sequence from top to bottom.
[0021] Electrodes for real-time monitoring of plantar pressure distribution are evenly arranged inside the filling layer. The key areas of the insole include the forefoot, the arch and the heel, and the key areas are subjected to different proportions of plantar pressure.
[0022] The host computer reconstructs the acquired data information by imaging, and uses an image reconstruction algorithm to convert the data collected by the electrodes into an impedance image, and the impedance image displays the impedance distribution of different areas of the sole of the foot.
[0023] Beneficial effects: The present invention adds electrical impedance scanning image detection and acceleration detection on the basis of collecting pressure information, which can obtain the patient's gait information more comprehensively for more accurate analysis; the smart insole is designed as a regular tetradecahedron lattice structure, which can more safely implant microcircuits and multiple sensors while ensuring lightness and comfort, reducing the risk of circuit damage.
[0024] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The accompanying drawings are used to better understand the present invention and do not constitute a limitation of the present invention.
[0026] Figure 1 is a flow chart provided according to the present invention;
[0027] Figure 2 is a circuit diagram of an acceleration acquisition module provided according to the present invention;
[0028] Figure 3 is a circuit diagram of an electrical impedance scanning module provided according to the present invention;
[0029] Figure 4 is a circuit diagram of a main control module provided according to the present invention;
[0030] Figure 5 is a circuit diagram of two ADG1608 multiplexing switches provided according to the present invention;
[0031] Figure 6 Schematic diagram of a smart insole with a tetradecahedron lattice structure provided by the present invention;
[0032] Figure 7 It is a schematic diagram of the internal structure of the insole body provided according to the present invention. DETAILED DESCRIPTION
[0033] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.
[0034] like Figure 1 As shown, the present invention provides a smart insole for gait analysis, comprising:
[0035] Insole body, sensor element, main control module, Bluetooth communication module and host computer;
[0036] The sensor element, the main control module and the Bluetooth communication module are all arranged inside the insole body. The main control module receives the data information collected by the sensor element, and sends the data information to the host computer through the Bluetooth communication module. The host computer receives the data information and processes the data information, and uses the machine learning classification algorithm to analyze the data information after the data processing to obtain the current gait information.
[0037] The sensing element includes an acceleration acquisition module and an electrical impedance scanning module;
[0038] The acceleration acquisition module is used to acquire foot acceleration changes caused by gait changes;
[0039] The electrical impedance scanning module is used to collect the electrical impedance changes of the insole caused by the changes in the sole pressure.
[0040] like Figure 2 As shown, the acceleration acquisition module includes an MPU-6050 chip U13, a pull-up resistor R21, a pull-up resistor R19, a current limiting resistor R20, and a current limiting resistor R20, and the main control module includes a main control chip U1;
[0041] Pin 13 and pin 8 of the MPU-6050 chip U13 are connected to a 3.3V power supply, pin 18 of the MPU-6050 chip U13 is grounded, a 0.1uF capacitor is connected between pin 1 and pin 8 of the MPU-6050 chip U13, pin 23 of the MPU-6050 chip U13 is connected to a 3.3V power supply through a pull-up resistor R21, pin 23 of the MPU-6050 chip U13 is also electrically connected to a current limiting resistor R20, pin 24 of the MPU-6050 chip U13 is connected to a 3.3V power supply through a pull-up resistor R19, pin 24 of the MPU-6050 chip U13 is also electrically connected to a current limiting resistor R18, and pins 23 and 24 of the MPU-6050 chip U13 are electrically connected to pins 58 and 59 of the main control chip U1 respectively.
[0042] Furthermore, a 3.3V power supply is connected to pin No. 13 of the MPU-6050 chip U13 to provide a working voltage for the MPU-6050 chip U13, and pin No. 18 of the MPU-6050 chip U13 is grounded to form a power supply loop.
[0043] Pins 23 and 24 of the MPU-6050 chip U13 are electrically connected to the main control chip U1 through the I2C interface to achieve data communication between the MPU-6050 chip U13 and the main control chip U1.
[0044] The acceleration acquisition module of the present invention adopts the MPU-6050 chip U13, which integrates sensors of a three-axis accelerometer and a three-axis gyroscope, and is used to measure acceleration and angular velocity. The circuit communicates with the MPU-6050 chip U13 through the main control chip U1 to obtain acceleration and angular velocity data; Pin 13 and Pin 8 of the MPU-6050 chip U13 are connected to the 3.3V power supply, and Pin 18 of the MPU-6050 chip U13 is grounded to ensure the stability of the circuit; A 0.1uF capacitor is connected between Pin 1 and Pin 8 of the MPU-6050 chip U13 to filter out power supply noise and ensure the stable operation of the acceleration sensor; Pin 23 and Pin 24 of the MPU-6050 chip U13 communicate with the main control chip U1 through the I2C interface, Pin 23 and Pin 24 of the MPU-6050 chip U13 are connected to 3.3V through a 10kΩ pull-up resistor respectively, and the signal line is protected by a 330Ω current limiting resistor, the CLKIN pin can be connected to an external crystal oscillator, and the CPOUT pin is used for external clock output. The INT pin is connected to the PA7 pin of the main control chip U1 to receive the interrupt signal from the MPU-6050 chip U13. When the power is turned on, the MPU-6050 chip U13 starts to work and communicates with the main control chip U1 through the I2C bus to transmit acceleration and angular velocity data. The main control chip U1 uses an STM32 microcontroller, such as Figure 4 shown.
[0045] Acceleration characteristics are physical quantities that describe the rate of change of an object's velocity. They refer to the amount of velocity change per unit time. Accelerometers and gyroscopes are usually used to detect acceleration information. The smart insole of the present invention analyzes the acceleration change information through an accelerometer, thereby analyzing the patient's gait data.
[0046] like Figure 3 As shown, the impedance scanning module includes an AD5933 chip U7 and an AD820 operational amplifier chip U12;
[0047] Pin 7 of the AD820 op amp chip U12 is connected to a 5V power supply, and pin 7 of the AD820 op amp chip U12 is also grounded through a 0.1uF capacitor. Pin 6 of the AD820 op amp chip U12 is connected to pin 2 of the pin header H11, and pin 1 of the pin header H11 is electrically connected to pin 1 of the pin header H10. Pin 1 of the pin header H10 is also connected to the voltage output terminal VOUT. Pin 2 of the AD820 op amp chip U12 is connected to pin 3 of the pin header H10 through resistor R14, and pins 2 and 4 of the pin header H10 are connected to pin 6 of the AD5933 chip U7. , Pin 5 of the AD5933 chip U7 is connected to the voltage input terminal VIN, Pin 4 of the AD5933 chip U7 is connected to the voltage input terminal VIN through a resistor R11, Pin 15 of the AD5933 chip U7 is electrically connected to Pin 59 of the main control chip U1 through a resistor R13, Pin 15 of the AD5933 chip U7 is also connected to a 5V power supply through a pull-up resistor R6, Pin 16 of the AD5933 chip U7 is electrically connected to Pin 58 of the main control chip U1 through a resistor R12, and Pin 16 of the AD5933 chip U7 is also connected to a 5V power supply through a pull-up resistor R5.
[0048] Furthermore, pin 7 of the AD820 op amp chip U12 is connected to a 5V power supply to provide a working voltage for the AD820 op amp chip U12. Pin 7 of the AD820 op amp chip U12 is also grounded through a 0.1uF capacitor to form a power supply loop to ensure the stability of the circuit.
[0049] As an integrated impedance converter, the AD5933 chip U7 is responsible for frequency scanning and impedance measurement. It can generate an AC excitation signal and calculate the impedance by measuring the returned voltage and current. Pins 15 and 16 of the AD5933 chip U7 communicate with the main control chip U1 through the I2C interface to transmit the impedance data to the main control chip U1.
[0050] include,
[0051] The GPIO pins of the main control chip U1 are connected to two ADG1608 multiplexer switches respectively, and the PA0 pin and PA4 pin of the main control chip U1 are used to enable the strobe switch module, such as Figure 5 shown.
[0052] Also includes,
[0053] When the PA0 pin of the main control chip U1 inputs a high level, the multiplexing switch U10 is enabled; when the PA4 pin of the main control chip U1 inputs a high level, the multiplexing switch U11 is enabled; when the PA0 pin and PA4 pin of the main control chip U1 input a low level, the multiplexing switch U10 and the multiplexing switch U11 are disabled.
[0054] The present invention utilizes the AD5933 chip U7 and two ADG1608 multiplexing switches to realize the function of multi-channel impedance scanning. The AD5933 chip U7 is an integrated impedance converter, responsible for frequency scanning and impedance measurement, capable of generating an AC excitation signal, and calculating impedance by measuring the returned voltage and current. Since the AD5933 chip U7 chip is not sufficient to provide sufficient current drive, an AD820 operational amplifier chip is used to realize the load capacity.
[0055] Two ADG1608 multiplexing switches are connected to the GPIO pins of the main control chip U1 respectively, and the PA0 pin and PA4 pin of the main control chip U1 are used to enable the strobe switch module. Furthermore, when the PA0 pin of the main control chip U1 inputs a high level, the multiplexing switch U10 is enabled; when the PA4 pin of the main control chip U1 inputs a high level, the multiplexing switch U11 is enabled; when the PA0 pin and PA4 pin of the main control chip U1 input a low level, the multiplexing switch U10 and the multiplexing switch U11 are disabled. After enabling the strobe multiplexing switch U10, the specific channel of the multiplexing switch U10 is selected by controlling the high and low level combination of the PA1, PA2, and PA3 pins of the main control chip U1. For example, when PA1, PA2, and PA3 are all low levels, the S1 channel is enabled; when PA1 is high and PA2 and PA3 are low levels, the S2 channel is enabled; when PA1 and PA2 are both low levels and PA3 is high, the S3 channel is enabled, and so on. Similarly, the specific channel of the multiplexer switch is selected by controlling the high and low level combination of the PA5, PA6, and PA7 pins. For example, when PA5, PA6, and PA7 are all low levels, the S1 channel is enabled; when PA5 is high and PA6 and PA7 are low levels, the S2 channel is enabled; when PA5 and PA6 are both low levels and PA7 is high, the S3 channel is enabled, and so on. In this way, the main control chip U1 can accurately control and select different impedance channels, realize the multi-channel impedance scanning function, and thus realize the selection and measurement of different impedance channels.
[0056] In the specific application, the GPIO and I2C interfaces of the main control chip U1 are first initialized, and then the working mode of the AD5933 chip U7 is configured. Before each measurement, the channel selection of the ADG1608 multiplexer is switched through the main control chip U1, and the measurement data of the AD5933 chip U7 is obtained, finally realizing the real-time scanning of multi-channel impedance measurement.
[0057] It also includes a power management module, which supplies power to the acceleration acquisition module, the impedance scanning module and the main control module.
[0058] The host computer includes a Bluetooth communication module and a data processing module. The host computer receives the data information collected by the sensor element through the Bluetooth communication module. The data processing module displays the data information in the form of acceleration data and impedance scanning images according to the built-in machine learning algorithm, and finally obtains the current gait information.
[0059] The host computer receives the impedance scan data through the Bluetooth communication module and converts the received impedance data into a matrix format that can be used for imaging processing. The imaging algorithm uses an image reconstruction algorithm based on electrical impedance imaging (EIT) and uses the physical position and distribution of the electrodes on the insole to generate a finite element model (FEM) as a mathematical representation of the imaging area. The system scans each pair of electrodes to obtain complete voltage data for the plantar area. The impedance value between each electrode pair is calculated using the known current and voltage and input into the imaging model. The impedance distribution map generated by the reconstruction algorithm reflects the changes in the impedance of different areas of the plantar.
[0060] The acceleration data display is used to show the three-axis acceleration, and the electrical impedance scanning image display is used to display the impedance image after analysis and conversion; the current gait information includes gait cycle, stride, pace, support phase, swing phase duration, sole load distribution, gait symmetry, posture stability, dynamic balance and energy consumption and other gait information.
[0061] Specifically, the host computer operation interface adopts the Qt development environment and is built using the C++ programming language. Through the rich controls and custom class libraries of the Qt framework, developers can quickly create modern, customizable user interfaces and support multiple operating systems such as Windows, Linux, and macOS. The system collects electrical impedance scanning data and inertial sensor data through smart insoles, and uses a support vector machine (SVM) classifier for gait analysis. The extracted features include maximum, minimum, and average values. The data is divided into 80% training sets and 20% test sets, and 5-fold cross-validation is performed, with an accuracy of more than 90%. The host computer interface displays impedance, pressure, and acceleration data, and analyzes information such as gait cycle, stride, and speed to achieve accurate analysis and display of gait information.
[0062] like Figure 7 As shown, the insole body includes an upper surface suede cloth T1, a filling layer, an intelligent gait analysis mechanism T5 and a bottom support layer of carbon fiber reinforced polymer T6. A supporting structure thermoplastic polyurethane lattice T2 is arranged in the filling layer, and the hydrogel T4 is filled inside the filling layer. A sensor element placement grid T3 is also arranged in the filling layer. The upper surface suede cloth T1, the filling layer, the intelligent gait analysis mechanism T5 and the bottom support layer of carbon fiber reinforced polymer T6 are arranged in sequence from top to bottom.
[0063] Electrodes for real-time monitoring of plantar pressure distribution are evenly arranged inside the filling layer. The key areas of the insole include the forefoot, the arch and the heel, and the key areas are subjected to different proportions of plantar pressure.
[0064] The filling layer is made of a conductive hydrogel with good elasticity and softness and a silicone elastomer to ensure comfortable protection when worn. The built-in impedance electrodes are distributed in the insole to monitor the resistance changes caused by pressure during walking. Under the action of external force, the hydrogel T4 deforms, and the electrodes detect the local conductivity changes caused by this. This change data, combined with the measurement results of the acceleration sensor, is processed through the electrical impedance imaging algorithm technology and the machine learning classification algorithm to finally generate gait information. The demand for gait analysis technology in the field of rehabilitation is growing, especially in rehabilitation treatment and rehabilitation training. Understanding and monitoring the patient's gait changes is crucial to evaluating the treatment effect and guiding the rehabilitation plan.
[0065] The smart insole has an electrode array embedded in the filling layer for electrical impedance tomography (EIT). The array consists of rectangular array electrodes. The electrodes are evenly distributed in the forefoot, arch and heel areas of the insole, covering the key stress-bearing areas of the sole. The arrangement of the electrode array ensures that the insole can monitor the pressure distribution of the sole of the foot in real time during walking, standing and other movements.
[0066] The smart insole combines conductive hydrogel with tomography technology. The physical properties of its hydrogel T4 can be customized as needed, and its performance can be further optimized by compounding with silicone. The changes in resistance are monitored by electrodes, and these data are used for tomography of electrical impedance tomography (EIT). Machine learning is applied to classify the pressure morphology obtained by electrical impedance scanning. Thanks to the mechanical properties of the conductive hydrogel-elastomer composite material, the flexibility of hydrogel T4 can buffer external forces.
[0067] The host computer reconstructs the acquired data information by imaging, and uses an image reconstruction algorithm to convert the data collected by the electrodes into an impedance image, and the impedance image displays the impedance distribution of different areas of the sole of the foot.
[0068] Electrodes are evenly arranged inside the filling layer of the insole to achieve real-time monitoring of plantar pressure distribution. The key areas of the insole include the forefoot, arch and heel, which bear different proportions of plantar pressure. Especially in gait analysis, these pressure changes can reflect the user's gait stability and health status. Rectangular array electrodes are arranged on the insole so that the electrodes can evenly cover the main pressure-bearing areas of the insole and ensure that the pressure distribution in each area is fully detected.
[0069] The voltage data acquired by the host computer is reconstructed by imaging, and an image reconstruction algorithm (such as a regularization method) is used to convert the electrode acquisition data into an impedance image to display the impedance distribution of different areas of the sole. These impedance distribution images are proportional to the pressure distribution of the sole. When the impedance value changes, it means that the pressure on the area is also changing.
[0070] The intelligent gait analysis mechanism T5 is provided with a Bluetooth communication module, a main control module, a power management module, and a sensor element integrated on a flexible PCB. The intelligent gait analysis mechanism T5 is also provided with a host computer.
[0071] like Figure 6 As shown, the smart insole structure of the present invention is composed of a cascade of regular tetradecahedral lattices. The tetradecahedral porous structure is composed of a plurality of tetradecahedral units tightly stacked, and these units are interconnected by shared faces, edges or vertices. The tetradecahedral porous structure has a high porosity and specific surface area, and its compressive resistance is relatively weak. When subjected to pressure, the pores in the structure will be compressed and deformed, thereby absorbing part of the energy and playing a role in buffering and shock absorption. Through this structural characteristic, good support, buffering and shock absorption effects and sensor integration can be achieved.
[0072] The present invention conducts a series of gait recognition experiments for the verification of the performance of intelligent insoles. In the experiment, a total of 18 healthy volunteers were recruited, covering 6 different specifications of insoles. The volunteers walked at a normal speed on a 7-meter walkway for at least 20 times, and a total of 377 sets of valid data without packet loss were obtained. The impedance value of each electrode is read by the electrical impedance scanning module, and the maximum, minimum and average values are extracted. At the same time, the three-axis acceleration and angular velocity data are read by the inertial sensor module, and the maximum, minimum, average and standard deviation are extracted. After all the data are normalized, they are divided into training sets and test sets (80% for training and 20% for testing). The support vector machine (SVM) classifier is trained using the training set data, and its parameters are set to find the optimal decision boundary. The classification performance of the SVM model is then verified with the test set data, and the classification accuracy is calculated. By analyzing the results of the model output, the acceleration data is used to detect the moment when the foot touches the ground and calculate the gait cycle; the stride and speed are calculated by combining the impedance and acceleration data; the impedance and acceleration data are used to identify the moment when the foot touches the ground and leaves the ground, and the duration of the support phase and the swing phase is calculated; the pressure value of each electrode in the gait cycle is analyzed to calculate the load distribution; the gait parameters of the left and right feet are calculated, and the symmetry index is used to evaluate the gait symmetry; the body swing amplitude and frequency are calculated through the acceleration and angular velocity data to evaluate the posture stability; finally, the dynamic balance index and energy consumption are calculated through the gait analysis model. Finally, a classification accuracy of more than 90% is obtained. The intelligent insole proposed by the present invention can accurately identify gait parameters such as gait cycle, stride, speed, support phase, swing phase duration, sole load distribution, gait symmetry, posture stability, dynamic balance and energy consumption, and then evaluate the patient's gait rehabilitation.
[0073] The present invention can monitor the patient's walking posture, foot pressure distribution, stride length, cadence and other key data in real time and non-invasively in daily life through sensor elements. By continuously collecting and analyzing this data, the smart insole can help rehabilitation professionals more accurately assess the patient's gait problems, adjust the rehabilitation plan in time, and monitor the progress of rehabilitation.
[0074] Impedance is the relationship between current and voltage. It usually refers to the combination of resistance, inductance and capacitance of an object. Generally, the impedance of a capacitor is -j / ωC, the impedance of an inductor is jωL, and the impedance of a resistor is R. The impedance of an object can be obtained by measuring the sum of these values. Different material structures will cause changes in conductivity and dielectric constant, different temperatures will cause changes in resistance and thus affect the impedance of the object, and different surface conditions will also change the conductivity and resistivity of the object.
[0075] When the wearer walks, the pressure on the foot causes the resistance of each part of the insole to change. The built-in impedance detection electrodes of the insole are distributed in different positions. The GPIO pins of the main control chip U1 select different electrode channels by controlling the ADG1608 multiplexer to measure the impedance at different positions. The impedance data is processed to extract key feature values, such as the maximum, minimum, and average values of the impedance change. These features reflect the pressure changes at different positions of the foot. Using the support vector machine (SVM) classifier, the impedance and acceleration data are combined for gait recognition. By analyzing the results output by the machine learning model, gait parameters such as gait cycle, stride, pace, support phase duration, swing phase duration, load distribution, gait symmetry, and posture stability are calculated.
[0076] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.
[0077] The units may or may not be physically separated, and the components displayed as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0078] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0079] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.
[0080] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A smart insole for gait analysis, characterized in that: include: Insole body, sensor element, main control module, Bluetooth communication module and host computer; The sensor element, the main control module and the Bluetooth communication module are all arranged inside the insole body. The main control module receives the data information collected by the sensor element, and the main control module sends the data information to the host computer through the Bluetooth communication module. The host computer receives the data information and processes the data information, and uses a machine learning classification algorithm to analyze the data information after the data processing to obtain the current gait information.
2. The smart insole for gait analysis according to claim 1, characterized in that: The sensor element includes an acceleration acquisition module and an electrical impedance scanning module; The acceleration acquisition module is used to acquire foot acceleration changes caused by gait changes; The electrical impedance scanning module is used to collect the electrical impedance changes of the insole caused by the changes in the sole pressure.
3. The smart insole for gait analysis according to claim 2, characterized in that: The acceleration acquisition module includes an MPU-6050 chip U13, a pull-up resistor R21, a pull-up resistor R19, a current limiting resistor R20, and a current limiting resistor R20, and the main control module includes a main control chip U1; Pin No. 13 and Pin No. 8 of the MPU-6050 chip U13 are connected to a 3.3V power supply, Pin No. 18 of the MPU-6050 chip U13 is grounded, a 0.1uF capacitor is connected between Pin No. 1 and Pin No. 8 of the MPU-6050 chip U13, Pin No. 23 of the MPU-6050 chip U13 is connected to a 3.3V power supply through the pull-up resistor R21, Pin No. 23 of the MPU-6050 chip U13 is also electrically connected to the current limiting resistor R20, Pin No. 24 of the MPU-6050 chip U13 is connected to a 3.3V power supply through the pull-up resistor R19, Pin No. 24 of the MPU-6050 chip U13 is also electrically connected to the current limiting resistor R18, Pin No. 23 and Pin No. 24 of the MPU-6050 chip U13 are electrically connected to Pin No. 58 and Pin No. 59 of the main control chip U1 respectively.
4. The smart insole for gait analysis according to claim 3, characterized in that: The electrical impedance scanning module includes an AD5933 chip U7 and an AD820 operational amplifier chip U12; Pin 7 of the AD820 op amp chip U12 is connected to a 5V power supply, and pin 7 of the AD820 op amp chip U12 is also grounded through a 0.1uF capacitor. Pin 6 of the AD820 op amp chip U12 is connected to pin 2 of the pin header H11, and pin 1 of the pin header H11 is electrically connected to pin 1 of the pin header H10, and pin 1 of the pin header H10 is also connected to a voltage output terminal VOUT. Pin 2 of the AD820 op amp chip U12 is connected to pin 3 of the pin header H10 through a resistor R14, and pins 2 and 4 of the pin header H10 are connected to pin 6 of the AD5933 chip U7. Pin, pin 5 of the AD5933 chip U7 is connected to the voltage input terminal VIN, pin 4 of the AD5933 chip U7 is connected to the voltage input terminal VIN through a resistor R11, pin 15 of the AD5933 chip U7 is electrically connected to pin 59 of the main control chip U1 through a resistor R13, pin 15 of the AD5933 chip U7 is also connected to a 5V power supply through a pull-up resistor R6, pin 16 of the AD5933 chip U7 is electrically connected to pin 58 of the main control chip U1 through a resistor R12, and pin 16 of the AD5933 chip U7 is also connected to a 5V power supply through a pull-up resistor R5.
5. The smart insole for gait analysis according to claim 3 or 4, characterized in that: include, The GPIO pins of the main control chip U1 are respectively connected to two ADG1608 multiplexer switches, and the PA0 pin and PA4 pin of the main control chip U1 are used to enable the strobe switch module.
6. The smart insole for gait analysis according to claim 5, characterized in that: Also includes, When the PA0 pin of the main control chip U1 inputs a high level, the multiplexing switch U10 is enabled; when the PA4 pin of the main control chip U1 inputs a high level, the multiplexing switch U11 is enabled; when the PA0 pin and PA4 pin of the main control chip U1 input a low level, the multiplexing switch U10 and the multiplexing switch U11 are disabled.
7. A smart insole for gait analysis according to any one of claims 2-4 and 6, characterized in that: It also includes a power management module, which supplies power to the acceleration acquisition module, the impedance scanning module and the main control module.
8. The smart insole for gait analysis according to claim 7, characterized in that: The host computer includes a Bluetooth communication module and a data processing module. The host computer receives the data information collected by the sensor element through the Bluetooth communication module. The data processing module displays the data information through acceleration data and impedance scanning images according to the built-in machine learning algorithm, and finally obtains the current gait information.
9. The smart insole for gait analysis according to claim 1, characterized in that: The insole body includes an upper surface suede cloth T1, a filling layer, an intelligent gait analysis mechanism T5, and a bottom support layer carbon fiber reinforced polymer T6. A supporting structure thermoplastic polyurethane lattice T2 is arranged in the filling layer, and a hydrogel T4 is filled inside the filling layer. A sensor element placement grid T3 is also arranged in the filling layer. The upper surface suede cloth T1, the filling layer, the intelligent gait analysis mechanism T5, and the bottom support layer carbon fiber reinforced polymer T6 are arranged in sequence from top to bottom.
10. The smart insole for gait analysis according to claim 9, characterized in that: Electrodes for real-time monitoring of plantar pressure distribution are evenly arranged inside the filling layer. The key areas of the insole include the forefoot, the arch and the heel, and the key areas are subjected to different proportions of plantar pressure.
11. The smart insole for gait analysis according to claim 10, characterized in that: The host computer reconstructs the acquired data information by imaging, and uses an image reconstruction algorithm to convert the data collected by the electrodes into an impedance image, and the impedance image displays the impedance distribution of different areas of the sole of the foot.
