Good group disease prediction shoe based on gait monitoring

Abstract

The invention relates to the field of intelligent shoes, in particular to an elderly group disease prediction shoe based on gait monitoring, which comprises a shoe body, an insole, a data acquisition module, a data transmission module and a main control terminal, the data acquisition module comprises a GPS module and a pressure sensor group, the pressure sensor group comprises a plurality of plantar pressure sensors, and the plurality of plantar pressure sensors are connected with the main control terminal. The data transmission module comprises a first power supply module, a slave and a first NB-IOT module; the master control terminal comprises a master control chip, a serial port screen, a second NB-IOT module and a second power supply module; the data acquisition module is used for acquiring pressure data and step number of each subarea of the sole; the data transmission module is used for packaging the distance data, the step number data and the pressure data in a JSON format and transmitting the data to a cloud MQTT server; the main control terminal is used for interacting with the second NB-IOT module through serial port transmission, subscribing cloud information and performing mode selection on a serial port screen interface; after corresponding operation, disease judgment is carried out. According to the invention, a multi-disease early warning function is realized.

Description

Technical Field

[0001] The present invention relates to the field of intelligent shoes, and particularly to a disease prediction shoe for the elderly based on gait monitoring. Background Art

[0002] In the existing intelligent shoes, the sensor module is fixed to the shoe upper through elastic straps, and the wearing method is cumbersome and affects the comfort of daily walking; relying on multi-sensor collaborative calibration and complex algorithms (such as 3D kinematic modeling, periodic drift correction) in the laboratory environment, the calculation amount is large and the real-time performance is insufficient, making it difficult to adapt to the low-cost health monitoring requirements of the home scenario. In addition, the function of detecting the plantar pressure distribution is not integrated, and it is impossible to capture key abnormal features such as narrowing of the step distance and concentration of pressure at the front end of the foot, and there is a lack of an Internet of Things module to support remote data transmission and real-time warning, so the application scenario is limited. Summary of the Invention

[0003] In order to solve the problems in the prior art that key abnormal features such as narrowing of the step distance and concentration of pressure at the front end of the foot cannot be captured and the single sensor dimension is insufficient, the present invention provides a disease prediction shoe for the elderly based on gait monitoring, including: a shoe body, an insole, a data acquisition module, a data transmission module and a main control terminal. The data acquisition module includes a GPS module and a pressure sensor group. The pressure sensor group includes several plantar pressure sensors. The data transmission module includes a first power module, a slave machine, and a first NB-IOT module. The main control terminal includes a main control chip, a serial port screen, a second NB-IOT module, and a second power module;

[0004] The data acquisition module is used to obtain the pressure data and the number of steps of each partition of the sole;

[0005] The data transmission module is used to pack the distance, the number of steps and the pressure data in JSON format and transmit them to the cloud MQTT server;

[0006] The main control terminal is used to interact with the second NB-IOT module through serial port transmission and subscribe to cloud information, and select a mode on the serial port screen interface; the mode includes a learning mode and an application mode. The learning mode is used to calculate the step length and step frequency according to the distance, the number of steps and the pressure information, combined with the timer count value, and obtain the pressure ratio of the left and right feet using the pressure data; perform regression fitting on several groups of step length and step frequency data to obtain the characteristic index of the fitting function; the application mode is used to substitute the step length and step frequency into the fitting function for matching, and judge whether the pressure ratio of the left and right feet is within the normal range, so as to make a disease judgment and update the relevant data and disease judgment results in real time.

[0007] A computer-readable storage medium stores a computer program, which, when executed by a processor, realizes the prediction function of the disease prediction shoe for the elderly based on gait monitoring.

[0008] The beneficial effects brought by the technical solution provided by the present invention are as follows: The plantar pressure sensor of the present invention is embedded in the insole, without external straps, and there is no burden during daily walking. It supports long-term monitoring in the home environment and has wearing comfort and applicability. By integrating multiple parameters such as plantar pressure distribution, stride length, and walking frequency, it can simultaneously warn of Parkinson's disease (narrow stride length and fast walking frequency) and cerebral infarction (unbalanced pressure sensing ratio), breaking through the single function of only targeting fall risk assessment and realizing the function of multi-disease warning. The hardware cost of the present invention is less than $50 (the cost of the IMU solution is more than $200), and there is no need for laboratory calibration. Users can independently complete the configuration of the learning mode, with low cost and easy deployment. Therefore, the present invention is significantly superior to the prior art in terms of wearing comfort, disease warning, and cost, providing a low-cost and highly reliable health management tool for the elderly group. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present invention will be further described below in conjunction with the drawings and embodiments. In the drawings:

[0010] Figure 1 is the structural block diagram of the disease prediction shoe for the elderly group based on gait monitoring in the embodiment of the present invention;

[0011] Figure 2 is the operation flow chart of the data transmission module in the embodiment of the present invention;

[0012] Figure 3 is the operation flow chart of the main control terminal in the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] In order to have a clearer understanding of the technical features, objectives, and effects of the present invention, the specific embodiments of the present invention will be described in detail below with reference to the drawings.

[0014] Embodiment 1

[0015] The embodiment of the present invention provides a disease prediction shoe for the elderly group based on gait monitoring. The structural block diagram is as Figure 1 shown, including: a shoe body, an insole, a data acquisition module, a data transmission module, and a main control terminal. The insole is placed inside the shoe body, and the data acquisition module is arranged inside the insole. The data acquisition module sends the collected data to the data transmission module through a serial port, and then transmits it to the MQTT server. The main control terminal obtains the data from the MQTT server and performs corresponding processing. The insole can be suitable for a variety of shoe types, facilitating user use.

[0016] For the data acquisition module, including the GPS module and the pressure sensor group, both are embedded in the insole and used to obtain the pressure data of each partition of the sole and the number of steps; the pressure sensor group includes several plantar pressure sensors, and the plantar pressure sensors are used to obtain the plantar pressure data, and the GPS module is used to obtain the change in walking distance. During operation, the pressure data, the number of steps, and the change in distance will be centrally transmitted to the main control terminal through the data transmission module. The main control terminal will calculate the cadence and stride according to the change in distance, pressure, and step information, and infer the walking posture based on the pressure data of each partition received to know the force on the user's instep.

[0017] The plantar pressure sensors adopt an array design. There are 18 sensing units in a single sensor, and each sensing point is independently designed to achieve minimum interference. In the present invention, through the cooperation of the 18-point array pressure sensor and the GPS, the plantar pressure distribution, stride, and cadence are synchronously captured, overcoming the limitations of only relying on acceleration data and the problem of insufficient dimensions of a single sensor in the prior art. Gait information has the characteristics of a large amount of information and a fast change rate. The NB-IOT module is responsible for data transmission, which can quickly connect devices, and at the same time ensure the fast and accurate sending and receiving of data, and read the gait information as completely as possible. At the same time, compared with the Bluetooth module, the transmission distance of the NB-IOT module can reach 15KM, fully meeting the daily activity needs of the elderly. The GPS module is used to obtain the change in the user's position, and the stride is calculated by combining the change in the data of the plantar pressure sensor to obtain the number of steps the user walks. Combining with the interval time between two times when the main control receives the change in position, the cadence of the user can be calculated. In the gait analysis algorithm, the present invention uses the upper computer to collect the walking data of healthy people and performs function fitting with MATLAB, and finds that the change rule approximately conforms to the function y = e^ -kx + b, where x is the cadence and y is the stride. When the main control is in the learning mode, this model is used to perform regression fitting on the cadence and stride data to inversely calculate k and b, and write them into the flash area, thereby saving the gait characteristic values of this user, that is, the cadence and stride.

[0018] For the data transmission module, including the first power supply module, the slave device, and the first NB-IOT module, the operation process is as follows Figure 2 , the slave device transmits the GPS data collected through the serial port and the plantar pressure data collected by the plantar pressure sensors, calculates the pressure sensing ratio and the number of steps according to the plantar pressure data, calculates the moving distance according to the GPS data, packs the distance, number of steps, and pressure data in JSON format, and transmits the data packet to the first NB-IOT module through the serial port transmission, and the first NB-IOT module sends it to the cloud MQTT server. The NB-IoT module supports ultra-long-distance data transmission (15km). Combining with data compression algorithms (such as differential coding), the power consumption is reduced to 1 / 3 of the traditional Bluetooth solution, which shows that the Internet of Things of the present invention has low power consumption. The first power supply module is used to supply power to the slave device.

[0019] For the main control terminal (host), including the main control chip, serial port screen, second NB-IOT module, second power supply module; with personalized gait modeling, the operation process is as follows Figure 3 , the STM32F407ZGT6 main control chip interacts with the second NB-IOT module through serial port transmission and subscribes to cloud information, and selects the mode on the serial port screen interface, which includes learning mode and application mode. The "learning mode" is used to construct user baseline data, and the nonlinear fitting function y=e^ -kx +b Dynamically adapt to individual differences to avoid the high false alarm rate of the traditional threshold method. The second power module is used to power the main control terminal.

[0020] In learning mode, the main control terminal collects 20 sets of step frequency, stride length, and pressure data to construct a data set, and uses linear fitting to extract the user's gait feature value. The main control chip solves the data packet obtained from the cloud server to obtain the distance, number of steps and pressure information, calculates the stride and step frequency in combination with the timer count value, and uses the pressure data to obtain the pressure ratio of the left and right feet. After receiving a certain number of data packets, several sets of stride and step frequency data are regressed and fitted to obtain the characteristic index of the fitting function. Finally, the characteristic index and several sets of stride, step frequency, and left and right foot pressure ratios are written into the on-chip flash area. After the learning mode ends, select the application mode.

[0021] In application mode, the data packets are also received and solved from the cloud server, and the stride and frequency are substituted into the fitting function for matching, and the pressure ratio of the left and right feet is judged to be in the normal range, so as to judge the disease. The main control terminal will continue to collect data, and distinguish whether the current collected data is in the normal range or the disease range according to the gait feature value extracted originally, and update the relevant data and disease judgment results in real time, and write them into the flash area on the chip. If the pressure ratio of the left and right feet is in the abnormal range, the abnormal information will be displayed through the serial port screen. For example, patients with Parkinson's disease walk in a panic state with small steps and forward leaning at the beginning of the disease. Compared with normal walking, the stride is narrow and the frequency is fast, and the pressure on the sole of the foot is concentrated on the front of the foot. By reading these abnormal gait information, the main control terminal will infer that the user may have the disease and issue a timely warning. Through the collaborative design of hardware modules and software algorithms, natural gait feature collection, multi-dimensional data analysis and early warning of diseases can be realized. When data needs to be read, historical data can be read from the flash area on the chip and sent to the serial port screen for display. Based on the historical feature values ​​stored in Flash, the main control terminal compares gait parameters in real time and can achieve millisecond-level response (data processing cycle <50ms), which is better than the delay of laboratory-level IMU systems in existing technologies (>200ms), and realizes real-time anomaly detection function.

[0022] Example 2

[0023] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the prediction function of a disease prediction shoe for the elderly population based on gait monitoring.

[0024] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An elderly group disease prediction shoe based on gait monitoring, characterized in that, Including: A shoe body, an insole, a data acquisition module, a data transmission module, and a main control terminal. The data acquisition module includes a GPS module and a pressure sensor group. The pressure sensor group includes several plantar pressure sensors. The data transmission module includes a first power module, a slave device, and a first NB-IoT module. The main control terminal includes a main control chip, a serial port screen, a second NB-IoT module, and a second power module; The data acquisition module is used to obtain the pressure data and the number of steps of each partition of the sole; The data transmission module is used to package the distance, the number of steps, and the pressure data in JSON format and transmit them to the cloud MQTT server; The main control terminal is used to interact with the second NB-IoT module through serial port transmission and subscribe to cloud information, and select a mode on the serial port screen interface; The modes include a learning mode and an application mode. The learning mode is used to calculate the stride and cadence according to the distance, the number of steps, and the pressure information, combined with the timer count value, and obtain the pressure sensing ratio of the left and right feet using the pressure data; Perform regression fitting on several groups of stride and cadence data to obtain the characteristic index of the fitting function; The application mode is used to substitute the stride and cadence into the fitting function for matching, and determine whether the pressure sensing ratio of the left and right feet is within the normal range, so as to perform disease judgment and update the relevant data and disease judgment results in real time.

2. The disease prediction shoe for the elderly group based on gait monitoring according to claim 1, wherein The insole is internally provided with plantar pressure sensors. The plantar pressure sensors adopt an array design. There are 18 sensing units distributed in a single sensor, and each sensing point adopts an independent design, which is used to obtain the pressure data and the number of steps of each partition of the sole.

3. The disease prediction shoe for the elderly group based on gait monitoring according to claim 1, characterized in that, The fitting function is y = e^ -kx + b, where y represents the stride, x represents the cadence, and k and b represent the coefficients of the function. The coefficients k and b are inversely calculated by regression fitting using the cadence and stride data.

4. The disease prediction shoe for the elderly group based on gait monitoring according to claim 1, wherein The slave device acquires the data of the GPS module and the pressure sensor group through serial port transmission.

5. The disease prediction shoe for the elderly group based on gait monitoring according to claim 1, characterized in that, The main control chip is used to solve the data packet obtained from the cloud server to obtain the distance, the number of steps, and the pressure information.

6. The disease prediction shoe for the elderly group based on gait monitoring according to claim 1, characterized in that, The model of the main control chip is STM32F407ZGT6.

7. A disease prediction shoe for the elderly group based on gait monitoring according to claim 1, characterized in that The GPS module is used to obtain the user's position change amount, and the user's walking steps obtained by combining the data change of the plantar pressure sensors are used to calculate the stride. Then, combined with the interval time between two times when the main control terminal receives the position change amount, the user's cadence is calculated. According to the received pressure data of each partition, the force condition of the user's instep is known to infer the walking posture.

8. A disease prediction shoe for the elderly group based on gait monitoring according to claim 1, characterized in that If the pressure sensing ratio of the left and right feet is in the abnormal range, the abnormal information is displayed through the serial port screen.

9. A computer-readable storage medium, characterized in that, A computer program is stored, and when the program is executed by a processor, the prediction function of the disease prediction shoe for the elderly group based on gait monitoring is realized.

Images