Foot sole monitoring method and system for diabetic patient based on Internet of Things

Abstract

The invention relates to the technical field of health management of diabetic foot high-risk patients, in particular to a foot sole monitoring method and system for diabetic patients based on the Internet of Things, and the system comprises a foot sole data acquisition module, a data analysis module, a dynamic correction module and a user interaction terminal. Pelma temperature, pressure and gait data are collected in real time through a sensor, abnormity is analyzed and risks are evaluated through a deep learning model, a personalized correction instruction is generated, the height of the insole and the gait angle are dynamically adjusted, and precise correction is achieved. The foot sole health state can be monitored in real time, the complication risk can be predicted, a personalized correction scheme is provided, the gait of a patient is effectively improved, the diabetic foot attack risk is reduced, and the life quality of the patient is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of medical health and Internet of Things technology, and particularly relates to a plantar monitoring method and system for diabetic patients based on Internet of Things. BACKGROUND

[0002] With the increase in the number of diabetic patients and the improvement in health management needs, plantar monitoring technology plays an important role in the early screening and risk warning of diabetic foot high-risk patients. However, the existing monitoring system often fails to comprehensively collect data or accurately analyze data when dealing with individual differences of different patients, resulting in insufficient intervention measures and increased health risks of patients. For example, the changes in plantar skin temperature, pressure distribution and gait characteristics of diabetic foot high-risk patients will be different due to individual living habits and disease development. If the monitoring system fails to dynamically adjust the monitoring parameters and provide personalized feedback in this case, it will be difficult to achieve comprehensive and accurate risk management. SUMMARY

[0003] The application relates to the technical field of health management of diabetic foot high-risk patients, and particularly relates to a plantar monitoring method and system for diabetic patients based on Internet of Things.

[0004] To solve the above technical problems, the application is implemented by the following technical scheme: a plantar monitoring method and system for diabetic patients based on Internet of Things, which comprises a plantar data acquisition module, a data analysis module, a dynamic correction module and a user interaction terminal.

[0005] The plantar data acquisition module is distributed in different areas in the shoes of patients and is used for real-time acquisition of plantar skin temperature, plantar pressure distribution and gait characteristic data. The plantar data acquisition module comprises a temperature sensor array, a pressure sensor matrix and an accelerometer. The temperature sensor array is arranged in key areas where the surface of an insole contacts the plantar, and is used for acquisition of skin temperature of each part of the plantar. The pressure sensor matrix is embedded in the inside of the insole, and is used for measurement of pressure distribution of each area of the plantar. The accelerometer is installed at the heel part, and is used for recording gait characteristic data, including step frequency, step length and gait stability. Each sensor transmits data to a central processing unit through a low-power Bluetooth communication mode.

[0006] The data analysis module is configured to receive data from the plantar data acquisition module and analyze the data based on a preset algorithm to identify abnormal change trends in the plantar region, predict potential risks, and generate correction instructions. The data analysis module includes an abnormality detection submodule, a risk assessment submodule, and a correction strategy generation submodule. The abnormality detection submodule identifies abnormal change patterns by performing real-time comparative analysis on the collected plantar skin temperature, pressure distribution, and gait characteristic data. The risk assessment submodule assesses the patient's plantar health status and determines the risk level by combining historical data and current data. The correction strategy generation submodule generates corresponding correction instructions based on the risk assessment results to guide the operation of the dynamic correction module.

[0007] The dynamic correction module includes an adjustable insole and a gait correction device. The dynamic correction module is communicatively connected to the data analysis module, receives correction instructions, and performs corresponding correction actions. The adjustable insole is composed of multiple layers of air bags. Each layer of air bag is independently controlled by a micro air pump to adjust the inflation amount, thereby adjusting the insole height and support intensity. The gait correction device includes a vibration feedback unit and a posture adjustment unit mounted on the sole. The vibration feedback unit reminds the patient to adjust the gait through slight vibration. The posture adjustment unit helps the patient correct the gait deviation by driving the sole angle through the built-in motor.

[0008] The user interaction terminal interacts with the data analysis module. The patient can view real-time monitoring data, risk warning information, and personalized suggestions through the user interaction terminal. The user interaction terminal includes a touch screen display, a voice prompt module, and a manual setting function. The touch screen display is used to display the plantar health status and risk assessment results. The voice prompt module issues a voice alarm when an abnormality is detected. The manual setting function allows the patient to adjust the monitoring parameters and correction intensity according to their own needs.

[0009] Preferably, after receiving the data from the plantar data acquisition module, the data analysis module extracts key features based on a preset algorithm and performs classification processing on the data to identify abnormal change patterns in the plantar region. The abnormality detection submodule uses an abnormality detection model based on deep learning. This model can accurately identify features such as increased plantar skin temperature, uneven pressure distribution, and gait abnormalities by training a large amount of plantar data from high-risk diabetic foot patients.

[0010] The risk assessment submodule combines individualized data of the patient, including age, disease duration, and lifestyle habits, to establish a comprehensive risk assessment model. The comprehensive risk assessment model uses a support vector machine (SVM) algorithm to combine plantar data features with individualized data to predict the risk probability of the patient developing plantar complications in the near future. The correction strategy generation submodule generates personalized correction instructions based on the risk assessment results, including insole height adjustment range, vibration feedback frequency, and posture adjustment angle.

[0011] Preferably, the training process for the anomaly detection model is as follows: Foot data of high-risk diabetic foot patients is obtained from a historical database, including foot skin temperature, pressure distribution, and gait characteristics; the historical data is cleaned to remove noise and outliers to ensure data quality; the cleaned data is stored according to time series to form a historical dataset for analysis and model training. The time series includes divisions by hour, day, and week.

[0012] The data receiving module of the central processing unit acquires individualized patient data from medical institutions or professional health management platforms, including blood glucose levels, body mass index, and past medical history; high-frequency noise in the data is removed using wavelet transform to obtain a smooth dataset; data validation rules are used to check the rationality of the data to ensure that the data meets the requirements of subsequent analysis and model input; at the same time, the data is uniformly converted into a standardized format to facilitate integration with historical datasets.

[0013] Features related to plantar health are extracted from historical datasets and individualized datasets. These features are then normalized and mapped to the [0, 1] interval to obtain plantar data features and individualized data features, respectively. This ensures that different features have the same dimension, avoiding the impact of large differences in feature values ​​on model training performance.

[0014] Convolutional Neural Networks (CNNs) were selected as the basic architecture for the anomaly detection model. The number of convolutional layers, the number of filters per layer, and the activation function were determined. This included setting up a neural network with four convolutional layers, with the number of filters per layer being [32, 64, 128, 256], and the Leaky ReLU activation function being used. Experiments were conducted to verify that the selected model structure could achieve good performance on the training set without overfitting.

[0015] The processed plantar data features and corresponding risk labels are used as training samples and input into the constructed model for training. The Adam optimization algorithm is used to adjust the model parameters so that the model learns the mapping relationship between plantar data and abnormal changes. The risk labels are obtained through clinical diagnosis results or expert annotation.

[0016] During training, the dataset is divided into training, validation, and test sets with a ratio of 70%:15%:15%. The model's performance metrics—accuracy, recall, and F1 score—are evaluated on the validation set to monitor for overfitting or underfitting. If the performance metrics on the validation set no longer improve or begin to deteriorate, training is stopped, resulting in an anomaly detection model. This prevents the model from overlearning noise in the training data and losing its generalization ability.

[0017] Individualized data features are input into the anomaly detection model; based on the learned patterns and rules, the anomaly detection model calculates and outputs prediction results, including the probability values ​​of abnormal changes in various areas of the patient's sole in the near future.

[0018] Preferably, the dynamic correction module adjusts the insole height and corrects gait according to the correction instructions as follows: The dynamic correction module first receives correction instructions from the data analysis module via wireless communication; the correction instructions include the set target insole height, vibration feedback frequency, and posture adjustment angle, and the microprocessor built into the dynamic correction module parses the correction instructions and converts them into operation signals; based on the operation signals, the dynamic correction module sends control signals to the adjustable insole and the gait correction device, wherein the adjustable insole adjusts the airbag inflation volume through a micro air pump to change the insole height and support strength; the gait correction device works in concert with the vibration feedback unit and the posture adjustment unit to adjust the vibration frequency and sole angle respectively, thereby achieving real-time correction of the patient's gait.

[0019] Each air bladder in the adjustable insole is independently controlled, with a micro-pump regulating the gas pressure within the bladder to precisely adjust the insole height and support. The posture adjustment unit uses a built-in motor to fine-tune the sole angle within ±5° to adapt to the gait needs of different patients. During adjustment, the dynamic correction module continuously monitors changes in foot data to ensure the correction effect is directed towards the target. Simultaneously, it works in conjunction with the user interface terminal to dynamically optimize correction parameters based on patient feedback, avoiding over- or under-correction.

[0020] Preferably, the process by which the adjustable insole automatically adjusts to its optimal state based on real-time foot data is as follows: A temperature sensor array and a pressure sensor matrix installed inside the insole monitor the temperature and pressure distribution data of the sole skin in real time. The temperature sensor array accurately measures the real-time temperature of various parts of the sole, while the pressure sensor matrix measures the pressure distribution in different areas of the sole. The sensors transmit the collected temperature and pressure data to the central processing unit in the form of electrical signals, ensuring timely data acquisition and accurate transmission, providing basic data support for the adjustment of the adjustable insole. Based on the anomaly detection model and the personalized correction goals set by the user's interactive terminal, the central processing unit calculates the target insole height and support strength for each area at the current moment. After receiving the real-time temperature and pressure data, the central processing unit compares it with the target values, calculating the height deviation and pressure deviation. The height deviation is the difference between the actual height and the target height, and the pressure deviation is the difference between the actual pressure and the target pressure.

[0021] The adjustable insole's built-in control algorithm is based on fuzzy control theory. It calculates the airbag inflation adjustment amount based on height and pressure deviations. The output control signal is calculated based on the fuzzy membership function of the deviation, which reflects the magnitude and trend of the current deviation. By adjusting the membership function parameters, the control of the adjustable insole can be made more precise and stable. The inflation adjustment signal calculated by the control algorithm is transmitted to a micro-air pump. The micro-air pump controls the gas pressure inside the airbag according to the received signal, changing the insole height and support strength. After adjusting the height, the adjustable insole continues to monitor temperature and pressure data in real time, repeating the above process of deviation calculation, control algorithm operation, and air pump adjustment, forming a closed-loop control system.

[0022] Preferably, the dynamic correction module achieves more efficient gait correction as follows: Feature extraction: Extract gait correction-related features from historical correction data and current plantar data; including: using plantar pressure change rate, gait stability index, and pressure difference between adjacent areas as input features, and using the adjustment angle of the gait correction device as output features.

[0023] Model Training and Optimization: A self-learning model for the gait correction device is constructed using a Deep Q-Network (DQN) algorithm from reinforcement learning. Extracted feature data is input into the model, which learns the optimal correction strategy under different conditions to achieve efficient gait correction movements, maximizing long-term corrective effects and patient comfort. During training, model parameters are continuously optimized to accurately predict the adjustment angles the gait correction device should take under various working conditions, achieving a more efficient gait correction goal. This includes using the reward function in the DQN algorithm to evaluate the effect of each correction operation. If gait stability improves after correction and patient feedback is positive, a positive reward is given; otherwise, a negative reward is given, guiding the model to learn the optimal correction strategy.

[0024] Condition assessment and strategy selection: Based on real-time monitored plantar data and environmental parameters, a trained self-learning model is used to assess the current working condition. The model selects the most suitable correction strategy from its learned strategy space based on input feature data: current plantar pressure distribution, gait stability index, and pressure conditions in other areas, determining the adjustment direction and amplitude of the gait correction device. Correction execution: The gait correction device, according to the correction strategy determined by the model, controls the sole angle through a drive mechanism to achieve real-time gait adjustment. During the correction process, gait changes are continuously monitored to ensure that the correction adjustment proceeds towards the target value. Simultaneously, it works in conjunction with an adjustable insole to dynamically balance the correction intensity in each area according to the overall correction needs of the system, avoiding over- or under-correction in certain areas and achieving efficient gait correction.

[0025] Effectiveness evaluation and feedback collection: Immediately after completing one gait correction session, collect plantar data feedback information. This includes actual changes in gait stability, patient subjective feedback scores, and changes in the energy consumption of the correction device. By comparing various indicators before and after correction, evaluate the effectiveness and efficiency of this correction procedure for gait correction. If gait stability is significantly improved after correction and patient feedback is positive, the correction is considered effective. Conversely, if significant gait fluctuations or poor patient feedback occur, further analysis of the causes is required.

[0026] Model parameter update and optimization: Based on the feedback evaluation results, the parameters of the self-learning model are updated using the learning mechanism in the reinforcement learning algorithm. If a correction operation achieves good results and high patient satisfaction, the model will strengthen the parameters related to the correction strategy that led to the result, making it more likely to be selected in similar situations in the future. Conversely, if the correction effect is poor, the model will adjust the parameters accordingly to avoid using similar unsatisfactory correction strategies again. By continuously repeating the above process of data collection, analysis, decision-making, feedback, and model update, the gait correction device can gradually improve its correction accuracy in long-term operation, continuously adapt to various changes in patients, achieve more efficient and accurate gait correction, and improve the stability of the entire system and the patient's user experience.

[0027] Preferably, the dynamic correction module also includes the following steps in achieving more efficient gait correction: Historical correction data accumulation: During system operation, the dynamic correction module continuously records the correction time of each correction operation, gait data before correction, gait data after correction, corresponding plantar pressure distribution data, ambient temperature and humidity data, and correction parameters set by the patient: target gait stability and correction schedule, thus obtaining historical correction data.

[0028] Current plantar data monitoring: The central processing unit acquires real-time plantar data, including the correction needs predicted for each area based on plantar pressure distribution, temperature changes, and patient habits, as well as the correction settings that patients adjust in real-time through the user interface terminal, including temporary adjustments to insole height or gait correction intensity. Simultaneously, it closely monitors the operating status of other correction devices and the correction intensity in other areas, and their impact on current correction needs. By combining the above information, it determines the actual changes in correction needs for each correction device's area at the current moment, thus obtaining the current plantar data.

[0029] A method for monitoring the plantar flexures of diabetic patients based on the Internet of Things (IoT) includes the following steps: Step 1: Foot data acquisition: Real-time data on foot skin temperature, pressure distribution, and gait characteristics are collected in different areas inside the patient's shoe, and the collected data is transmitted to the central processing unit via wireless communication.

[0030] Step Two: Foot Data Analysis and Risk Assessment: Collect historical foot data and obtain individualized patient data, then extract foot health-related features, and use a deep learning model for training to obtain an anomaly detection model; collect patient correction habit data, extract features to establish a correction behavior pattern model, clarify the quantitative relationship between behavior patterns and correction needs, and obtain correction needs prediction results based on patient behavior.

[0031] Step 3: Based on the fusion strategy, the patient behavior prediction results are integrated with the historical plantar data and the individualized data prediction model output to obtain the final correction demand prediction model, predict future correction needs, obtain the integrated prediction results, and generate correction instructions based on the integrated prediction results.

[0032] Step 4: Based on the correction command, the adjustable insole and gait correction device automatically adjust the correction parameters according to the target insole height and gait correction angle, so as to realize the dynamic adjustment of the correction force and correction angle in each area and ensure the balance of correction effect.

[0033] Step 5: The patient interacts with the central processing unit through the user interface terminal to view the real-time foot health status and set personalized correction settings. Attached Figure Description

[0034] Figure 1 The diagram shows the overall structure of the system of the present invention, illustrating the connection relationship and data flow direction between the plantar data acquisition module, the data analysis module, the dynamic correction module, and the user interaction terminal.

[0035] Figure 2 This is a schematic diagram of the layout of the foot data acquisition module, showing the distribution of the temperature sensor array, pressure sensor matrix, and accelerometer in the insole and their communication connection with the central processing unit.

[0036] Figure 3 This is a workflow diagram for the data analysis module, describing the data processing steps and logical relationships of the anomaly detection submodule, risk assessment submodule, and correction strategy generation submodule.

[0037] Figure 4 This is a structural diagram of the dynamic correction module, showing the main components of the adjustable insole and gait correction device and how they work together.

[0038] Figure 5 This is a schematic diagram of the user interface of the interactive terminal, showing the operation area and information display content of the touch screen display, voice prompt module, and manual setting function.

[0039] Figure 6The flowchart illustrates the process of executing correction instructions, from the data analysis module generating correction instructions to the dynamic correction module adjusting the insole height and gait correction angle.

[0040] The attached figures are labeled as follows: 1. Foot data acquisition module; 2. Data analysis module; 3. Dynamic correction module; 4. User interaction terminal; 5. Temperature sensor array; 6. Pressure sensor matrix; 7. Accelerometer; 8. Anomaly detection submodule; 9. Risk assessment submodule; 10. Correction strategy generation submodule; 11. Adjustable insole; 12. Gait correction device; 13. Touch screen display; 14. Voice prompt module; 15. Manual setting function. Detailed Implementation

[0041] This invention provides a method and system for monitoring the soles of diabetic patients based on the Internet of Things (IoT), the specific implementation of which is as follows. (Combined with...) Figures 1 to 6 As shown, the system of the present invention includes a foot data acquisition module 1, a data analysis module 2, a dynamic correction module 3, and a user interaction terminal 4. The modules communicate with each other via wireless communication to achieve data transmission and collaborative work.

[0042] In practical applications, the plantar data acquisition module 1 is placed in different areas inside the patient's shoe to collect real-time data on plantar skin temperature, plantar pressure distribution, and gait characteristics. For example... Figure 2 As shown, the foot data acquisition module 1 includes a temperature sensor array 5, a pressure sensor matrix 6, and an accelerometer 7. The temperature sensor array 5 is positioned on key areas of the insole surface that contact the foot, such as the heel, arch, and forefoot, to collect skin temperature data from various parts of the foot. The pressure sensor matrix 6 is embedded inside the insole, covering the entire insole surface, to measure the pressure distribution in different areas of the foot. The accelerometer 7 is mounted in the heel area to record gait characteristic data, including cadence, stride length, and gait stability. The temperature sensor array 5, pressure sensor matrix 6, and accelerometer 7 all transmit data to the central processing unit via Bluetooth Low Energy communication to ensure real-time data transmission and accuracy.

[0043] Data analysis module 2 receives data from plantar data acquisition module 1, analyzes the data based on a preset algorithm, identifies abnormal plantar changes, predicts potential risks, and generates corrective instructions. For example... Figure 3As shown, the data analysis module 2 includes an anomaly detection submodule 8, a risk assessment submodule 9, and a correction strategy generation submodule 10. The anomaly detection submodule 8 identifies abnormal change patterns by performing real-time comparative analysis of collected plantar skin temperature, pressure distribution, and gait characteristic data. The risk assessment submodule 9 combines historical and current data to assess the patient's plantar health status and determine the risk level. The correction strategy generation submodule 10 generates corresponding correction instructions based on the risk assessment results to guide the work of the dynamic correction module 3. The anomaly detection submodule 8 employs a deep learning-based anomaly detection model. This model, trained on a large amount of plantar data from high-risk diabetic foot patients, can accurately identify features such as elevated plantar skin temperature, uneven pressure distribution, and gait abnormalities. The risk assessment submodule 9 combines the patient's individualized data, including age, disease duration, and lifestyle habits, to establish a comprehensive risk assessment model. This comprehensive risk assessment model uses a support vector machine algorithm to combine plantar data features with individualized data to predict the probability of the patient developing plantar complications in the near future. The correction strategy generation submodule 10 generates personalized correction instructions based on the risk assessment results, including the insole height adjustment range, vibration feedback frequency, and posture adjustment angle.

[0044] The dynamic correction module 3 includes an adjustable insole 11 and a gait correction device 12. The dynamic correction module 3 is communicatively connected to the data analysis module 2, receiving correction commands and executing corresponding correction actions. For example... Figure 4As shown, the adjustable insole 11 consists of a multi-layered airbag structure. Each airbag layer's inflation volume is independently controlled by a micro-pump, thereby adjusting the insole height and support strength. The gait correction device 12 includes a vibration feedback unit and a posture adjustment unit installed on the sole. The vibration feedback unit reminds the patient to adjust their gait through slight vibrations, while the posture adjustment unit uses a built-in motor to fine-tune the sole angle, helping the patient correct gait deviations. The dynamic correction module 3 adjusts the insole height and corrects the gait according to the correction instructions as follows: The dynamic correction module 3 first receives correction instructions from the data analysis module 2 via wireless communication. The correction instructions include the set target insole height, vibration feedback frequency, and posture adjustment angle. The microprocessor built into the dynamic correction module 3 parses the correction instructions and converts them into operation signals. Based on the operation signals, the dynamic correction module 3 sends control signals to the adjustable insole 11 and the gait correction device 12. The adjustable insole 11 adjusts the airbag inflation volume through a micro-pump, changing the insole height and support strength. The gait correction device 12 works in concert with a vibration feedback unit and a posture adjustment unit to adjust the vibration frequency and sole angle, thereby achieving real-time gait correction for the patient. Each air bladder in the adjustable insole 11 is independently controlled, with a micro-pump regulating the gas pressure within the bladder to precisely adjust the insole height and support. The posture adjustment unit uses a built-in motor to fine-tune the sole angle, with an adjustment range of ±5° to adapt to the gait needs of different patients. During adjustment, the dynamic correction module 3 continuously monitors changes in foot data to ensure the correction effect moves in the target direction. It also works in concert with the user interaction terminal 4 to dynamically optimize correction parameters based on patient feedback, avoiding over- or under-correction.

[0045] The user interaction terminal 4 interacts with the data analysis module 2, allowing patients to view real-time monitoring data, risk warning information, and personalized suggestions through the user interaction terminal 4. Figure 5 As shown, the user interaction terminal 4 includes a touchscreen display 13, a voice prompt module 14, and a manual setting function 15. The touchscreen display 13 shows the foot health status and risk assessment results. The voice prompt module 14 issues a voice alarm when an abnormality is detected. The manual setting function 15 allows the patient to adjust monitoring parameters and correction intensity according to their needs. The user interaction terminal 4 exchanges data with the data analysis module 2 wirelessly. The patient can view the changing trends of foot health status and understand the risk assessment results through the touchscreen display 13, and take timely measures based on the alarm information from the voice prompt module 14. The manual setting function 15 allows the patient to adjust correction parameters according to personal habits, such as adjusting the insole height or vibration feedback frequency, to meet individual needs.

[0046] The specific steps of the correction instruction execution process are as follows: Figure 6As shown, the process from the data analysis module 2 generating correction instructions to the dynamic correction module 3 adjusting the insole height and gait correction angle is as follows: The data analysis module 2, based on real-time data provided by the foot data acquisition module 1, generates correction instructions through the anomaly detection submodule 8, risk assessment submodule 9, and correction strategy generation submodule 10. The correction instructions are transmitted wirelessly to the dynamic correction module 3. The built-in microprocessor in the dynamic correction module 3 parses the correction instructions and converts them into operation signals. These operation signals are sent to the adjustable insole 11 and the gait correction device 12, respectively. The adjustable insole 11 adjusts the airbag inflation volume using a micro-pump, changing the insole height and support strength. The gait correction device 12 works collaboratively with a vibration feedback unit and a posture adjustment unit to adjust the vibration frequency and sole angle, thereby achieving real-time gait correction for the patient. During the adjustment process, the dynamic correction module 3 continuously monitors changes in foot data to ensure the correction effect moves in the target direction. Simultaneously, it works in conjunction with the user interaction terminal 4 to dynamically optimize correction parameters based on patient feedback, avoiding over- or under-correction.

[0047] In actual operation, the system first collects real-time data on foot skin temperature, pressure distribution, and gait characteristics through the foot data acquisition module 1, and transmits the collected data to the data analysis module 2 via wireless communication. The data analysis module 2 processes the received data, extracts key features, classifies the data, and identifies abnormal foot change patterns. The risk assessment submodule 9, combined with the patient's individualized data, establishes a comprehensive risk assessment model to predict the probability of the patient developing foot complications in the near future. The correction strategy generation submodule 10 generates personalized correction instructions based on the risk assessment results and transmits them to the dynamic correction module 3 via wireless communication. The dynamic correction module 3 adjusts the insole height and corrects the gait according to the correction instructions to ensure the correction effect is directed towards the target. The user interaction terminal 4 interacts with the data analysis module 2. Patients can view real-time monitoring data, risk warning information, and personalized suggestions through the user interaction terminal 4, and adjust correction parameters according to the manual setting function 15 to meet their individual needs.

[0048] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principle of this invention will be further explained below in conjunction with a specific application scenario.

[0049] In practical applications, the system first collects real-time data on the patient's foot skin temperature, pressure distribution, and gait characteristics through the foot data acquisition module 1. For example... Figure 2As shown, temperature sensor array 5 is arranged in key areas of the insole surface that contact the sole of the foot, such as the heel, arch, and forefoot, to accurately measure skin temperature changes in various parts of the sole. Pressure sensor matrix 6 is embedded inside the insole, covering the entire insole surface, enabling comprehensive monitoring of pressure distribution in different areas of the sole. Accelerometer 7 is installed in the heel area to record gait characteristic data, including cadence, stride length, and gait stability. These sensors transmit data to the central processing unit via Bluetooth Low Energy communication, ensuring real-time data accuracy.

[0050] Data analysis module 2 receives data from plantar data acquisition module 1 and processes it to identify abnormal patterns. For example... Figure 3 As shown, the anomaly detection submodule 8 uses a deep learning model to perform real-time comparative analysis of collected plantar skin temperature, pressure distribution, and gait characteristic data to identify abnormal trends. This model, trained on a large amount of plantar data from high-risk diabetic foot patients, can accurately identify features such as elevated plantar skin temperature, uneven pressure distribution, and gait abnormalities. The risk assessment submodule 9 combines individualized patient data, including age, disease duration, and lifestyle habits, to establish a comprehensive risk assessment model. This model uses a support vector machine algorithm to combine plantar data features with individualized data to predict the probability of patients developing plantar complications in the near future. The correction strategy generation submodule 10 generates personalized correction instructions based on the risk assessment results, including the insole height adjustment range, vibration feedback frequency, and posture adjustment angle.

[0051] The dynamic correction module 3 executes corresponding correction actions based on the correction instructions. For example... Figure 4 As shown, the adjustable insole 11 consists of a multi-layered airbag structure. Each airbag layer's inflation volume is independently controlled by a micro-pump, thereby adjusting the insole height and support strength. The gait correction device 12 includes a vibration feedback unit and a posture adjustment unit. The vibration feedback unit reminds the patient to adjust their gait through slight vibrations, while the posture adjustment unit uses a built-in motor to fine-tune the sole angle, helping the patient correct gait deviations. The dynamic correction module 3 first receives correction instructions from the data analysis module 2 via wireless communication. These instructions include the set target insole height, vibration feedback frequency, and posture adjustment angle. The built-in microprocessor parses the correction instructions and converts them into operation signals. Based on the operation signals, the dynamic correction module 3 sends control signals to the adjustable insole 11 and the gait correction device 12. The adjustable insole 11 adjusts the airbag inflation volume using a micro-pump, changing the insole height and support strength. The gait correction device 12, through the coordinated operation of the vibration feedback unit and the posture adjustment unit, adjusts the vibration frequency and sole angle respectively, thereby achieving real-time correction of the patient's gait.

[0052] The user interaction terminal 4 interacts with the data analysis module 2, allowing patients to view real-time monitoring data, risk warning information, and personalized suggestions through the user interaction terminal 4. Figure 5 As shown, the touchscreen display 13 displays the foot health status and risk assessment results, the voice prompt module 14 issues a voice alarm when an abnormality is detected, and the manual setting function 15 allows the patient to adjust the monitoring parameters and correction intensity according to their own needs. The user interaction terminal 4 exchanges data with the data analysis module 2 via wireless communication. The patient can view the changing trend of the foot health status and understand the risk assessment results through the touchscreen display 13, and take timely measures based on the alarm information from the voice prompt module 14. The manual setting function 15 allows the patient to adjust the correction parameters according to personal habits, such as adjusting the insole height or vibration feedback frequency, to meet personalized needs.

[0053] The specific steps of the correction instruction execution process are as follows: Figure 6 As shown, the data analysis module 2, based on real-time data provided by the foot data acquisition module 1, generates correction instructions through the anomaly detection submodule 8, risk assessment submodule 9, and correction strategy generation submodule 10. These correction instructions are transmitted wirelessly to the dynamic correction module 3. The microprocessor built into the dynamic correction module 3 parses the correction instructions and converts them into operation signals. These operation signals are sent to the adjustable insole 11 and the gait correction device 12. The adjustable insole 11 adjusts the airbag inflation volume using a micro-pump, changing the insole height and support strength. The gait correction device 12, through the coordinated operation of a vibration feedback unit and a posture adjustment unit, adjusts the vibration frequency and sole angle respectively, thereby achieving real-time correction of the patient's gait. During the adjustment process, the dynamic correction module 3 continuously monitors changes in foot data to ensure the correction effect moves in the target direction. Simultaneously, it works in conjunction with the user interaction terminal 4 to dynamically optimize correction parameters based on patient feedback, avoiding over- or under-correction.

[0054] In a specific scenario, suppose a diabetic patient experiences abnormal plantar pressure distribution due to prolonged standing at work. The system detects a significant increase in pressure in the forefoot area via pressure sensor matrix 6 and a slight increase in skin temperature in this area via temperature sensor array 5. The anomaly detection submodule 8 in data analysis module 2 identifies this abnormal pattern and assesses a high probability of the patient developing plantar ulcers in the near future via risk assessment submodule 9. Correction strategy generation submodule 10 generates correction instructions, directing dynamic correction module 3 to adjust insole height and gait correction angle. Upon receiving the correction instructions, dynamic correction module 3 lowers the height of the forefoot area via adjustable insole 11 to reduce pressure in that area. Simultaneously, the vibration feedback unit of gait correction device 12 reminds the patient to adjust their gait, and the posture adjustment unit fine-tunes the sole angle to help the patient restore a normal gait. The patient views the correction effect through user interaction terminal 4 and manually adjusts the correction parameters as needed, ultimately achieving a balanced distribution of plantar pressure and stable gait recovery.

[0055] The above embodiments are only one of the preferred embodiments of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for monitoring the soles of the feet of diabetic patients based on the Internet of Things, characterized in that, Includes the following steps: The foot data acquisition module (1) collects real-time data on the skin temperature, pressure distribution, and gait characteristics of the patient's feet in different areas of the shoe. The collected data is transmitted to the data analysis module (2). The data analysis module (2) processes the received data, extracts foot health-related features, and identifies abnormal change patterns. Correction instructions are generated based on abnormal change patterns and individualized patient data, and the correction instructions are sent to the dynamic correction module (3). The dynamic correction module (3) adjusts the height of the adjustable insole (11) and the angle of the gait correction device (12) according to the correction command to achieve real-time correction of the foot condition; Users can view real-time monitoring data, risk warning information and personalized suggestions through the user interaction terminal (4), and adjust the correction parameters as needed.

2. The method for monitoring the soles of diabetic patients based on the Internet of Things according to claim 1, characterized in that: The foot data acquisition module (1) includes a temperature sensor array (5), a pressure sensor matrix (6), and an accelerometer (7). The temperature sensor array (5) is arranged on the surface of the insole in the key area that contacts the sole of the foot to collect the skin temperature of each part of the sole. The pressure sensor matrix (6) is embedded inside the insole to measure the pressure distribution in each area of ​​the sole. The accelerometer (7) is installed on the heel to record gait characteristic data.

3. The method for monitoring the plantar surface of diabetic patients based on the Internet of Things according to claim 1, characterized in that: The data analysis module (2) includes an anomaly detection submodule (8), a risk assessment submodule (9), and a correction strategy generation submodule (10). The anomaly detection submodule (8) identifies abnormal change patterns by performing real-time comparative analysis on the collected data of foot skin temperature, pressure distribution and gait characteristics. The risk assessment submodule (9) combines historical and current data to assess the patient’s foot health status and determine the risk level; The correction strategy generation submodule (10) generates corresponding correction instructions based on the risk assessment results.

4. The method for monitoring the soles of diabetic patients based on the Internet of Things according to claim 3, characterized in that: The anomaly detection submodule (8) adopts a deep learning-based anomaly detection model. This model is trained on a large number of foot data from high-risk diabetic foot patients and can accurately identify features such as increased foot skin temperature, uneven pressure distribution, and abnormal gait.

5. The method for monitoring the plantar surface of diabetic patients based on the Internet of Things according to claim 3, characterized in that: The risk assessment submodule (9) combines individual patient data, including age, disease course, and lifestyle habits, to establish a comprehensive risk assessment model; The comprehensive risk assessment model uses the support vector machine algorithm to combine plantar data features with individualized data to predict the probability of patients developing plantar complications in the near future.

6. The method for monitoring the plantar surface of diabetic patients based on the Internet of Things according to claim 1, characterized in that: The dynamic correction module (3) includes an adjustable insole (11) and a gait correction device (12). The adjustable insole (11) is composed of a multi-layer airbag structure. Each airbag is independently controlled by a micro air pump to adjust the insole height and support strength. The gait correction device (12) includes a vibration feedback unit and a posture adjustment unit. The vibration feedback unit reminds the patient to adjust their gait by slight vibration, and the posture adjustment unit drives the sole angle of the shoe to be finely adjusted by a built-in motor.

7. A method for monitoring the plantar surface of diabetic patients based on the Internet of Things according to claim 6, characterized in that: After receiving the correction command, the dynamic correction module (3) analyzes the correction command and adjusts the air volume of the adjustable insole (11) and the vibration frequency and sole angle of the gait correction device (12) respectively through the control signal. The dynamic correction module (3) continuously monitors changes in plantar data during the adjustment process and dynamically optimizes correction parameters based on patient feedback.

8. The method for monitoring the plantar surface of diabetic patients based on the Internet of Things according to claim 1, characterized in that: The user interaction terminal (4) includes a touch screen display (13), a voice prompt module (14), and a manual setting function (15). A touchscreen display (13) is used to display the health status and risk assessment results of the soles of the feet; The voice prompt module (14) is used to issue a voice alarm when an anomaly is detected; The manual setting function (15) allows patients to adjust monitoring parameters and correction intensity according to their own needs.

9. A method for monitoring the plantar surface of diabetic patients based on the Internet of Things according to claim 1, characterized in that: The training process of the anomaly detection model includes obtaining plantar data of high-risk diabetic foot patients from historical databases, including plantar skin temperature, pressure distribution and gait characteristics data; Clean historical data to remove noisy data and outliers; The cleaned data is stored in time series to form a historical dataset that can be used for analysis and model training.

10. A method for monitoring the plantar surface of diabetic patients based on the Internet of Things according to claim 1, characterized in that: During the execution of the correction command, the dynamic correction module (3) receives the correction command through wireless communication and converts the correction command into an operation signal; The operation signals are sent to the adjustable insole (11) and the gait correction device (12) respectively. The adjustable insole (11) adjusts the air volume of the airbag through a micro air pump to change the height and support of the insole. The gait correction device (12) works in concert with the vibration feedback unit and the posture adjustment unit to adjust the vibration frequency and the shoe sole angle respectively.

Images