Novel plantar pressure and shear force signal decoupling measurement method based on deep learning

Through deep learning methods, the multi-dimensional perceived sensor structure is designed, which solves the problem of complex coupling between pressure and shear force under complex stress states of the sole, and realizes high-precision plantar pressure and shear force measurement to meet the intelligent wearable device monitoring requirements.

CN119014856AActive Publication Date: 2024-11-26SOUTH CHINA UNIV OF TECH

Patent Information

Application Number
CN202411313751.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-20
Publication Date
2024-11-26
Estimated Expiration
2044-09-20

AI Technical Summary

Technical Problem

When the prior art measures the complex stress state of the sole of the foot, the pressure and shear force are complexly coupled, resulting in insufficient measurement accuracy and applicability, which cannot meet the needs of smart wearable devices for foot stress monitoring.

Method used

Deep learning method is used to design a multi-dimensional sensor structure, and the precise distribution of pressure and shear force sensors in key parts of the sole of the foot is combined with low-power Bluetooth data transmission, adaptive Kalman filtering, and intelligent feature extraction algorithms to perform high precision Data acquisition and signal analysis enable measurement of shear force in 360° direction.

Benefits of technology

It significantly improves the measurement accuracy and diversity of sole pressure and shear force, provides reliable data support for smart wearable devices, and improves the system's intelligence level and decision-making support capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119014856A_ABST
    Figure CN119014856A_ABST
Patent Text Reader

Abstract

The invention discloses a novel plantar pressure and shear force signal decoupling measurement method based on deep learning, and relates to the technical field of plantar stress and gait measurement, and the method comprises the following steps: S1, making a sensor structure; step S2, integrating multi-dimensional perception; and S3, pressure and shear force signal analysis is carried out. According to the novel plantar pressure and shear force sensor based on deep learning and the signal decoupling method thereof, high-precision measurement of plantar pressure and shear force is realized, diversity and accuracy of data acquisition are remarkably improved, and reliable data support is provided for intelligent wearable equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of plantar force and gait measurement, and in particular to a novel deep learning method for decoupling plantar pressure and shear force signals. Background Art

[0002] Gait is one of the key patterns in daily human activities, which contains rich physiological information about human movement. The force on the plantar surface is a treasure trove of information that needs to be explored. However, gait is a complex behavioral characteristic that involves central commands, body balance and coordination control, coordinated movements of joints and muscle groups of the lower limbs, and postures of the upper limbs and trunk. In addition, the bone convex structure and various soft tissues of the plantar surface work together to make it often in a complex stress state of coupled pressure and shear force. How to effectively decouple pressure and shear force in this complex stress state and measure the uncertainty of the shear force direction has become a major challenge.

[0003] The force state of the sole of the foot is complex and changes dynamically with time and space. The coupling of pressure and shear force makes accurate measurement extremely difficult. The existing technology has the following problems:

[0004] 1) A preliminary consensus on plantar pressure sensors was reached, but complex force conditions were not addressed.

[0005] 2) There have been preliminary studies on the simultaneous measurement of foot pressure and shear force, but there are deficiencies in the measurement, accuracy and applicability of the shear force direction.

[0006] 3) Plantar force measurement is combined with artificial intelligence, but there are still obvious deficiencies in the accurate acquisition of raw plantar force data and the feature extraction under the cross-effects of gait and multiple diseases.

[0007] 4) Differences in the performance of the arch, bone convexity and plantar tissue lead to complex plantar stress conditions

[0008] In summary, due to the variability of human activities, the complexity of the plantar bone structure, and the combined effects of various soft tissue structures, the plantar is often in a complex stress state of coupled pressure and shear force. At present, the ability to decouple and measure complex plantar forces is insufficient, resulting in the inability of existing force sensors to meet the needs of smart wearable devices for accurate monitoring of plantar forces. Summary of the invention

[0009] The purpose of the present invention is to provide a new deep learning method for decoupling plantar pressure and shear force signals to solve the problems raised in the above-mentioned background technology, to perform data analysis and prediction in real time and efficiently, to provide users with accurate and reliable results, to improve the intelligence level of the system, and to provide a solid foundation for decision support and optimization.

[0010] To achieve the above object, the present invention provides a novel deep learning method for decoupling plantar pressure and shear force signals, comprising the following steps:

[0011] Step S1, making a sensor structure;

[0012] Step S2, integrating multi-dimensional perception;

[0013] Step S3: analyzing the pressure and shear force signals.

[0014] Preferably, the specific steps of step S1 are as follows:

[0015] Step S11, designing a multi-dimensional pressure and shear force sensor structure;

[0016] Step S12: assemble and manufacture the sensor according to the structural design.

[0017] Preferably, in step S2, data collection is performed by a data collection unit, and the collected data includes plantar pressure and shear force data during the user's daily activities. After the collection is completed, the sensor structure transmits real-time data to the data processing unit via low-power Bluetooth.

[0018] Preferably, the specific steps of the analysis process in step S3 are as follows:

[0019] Step S31: performing intelligent feature extraction processing of multi-dimensional force signals: using an intelligent feature extraction algorithm to identify key multi-dimensional mechanical features from the processed signals, the multi-dimensional mechanical features including pressure magnitude, shear force magnitude and shear force direction;

[0020] Step S32: Perform high-precision data cleaning and standardization processing: Use refined data cleaning and standardization processing methods to ensure the integrity and consistency of all data.

[0021] Therefore, the present invention adopts the above-mentioned new deep learning plantar pressure and shear force signal decoupling measurement method to achieve high-precision measurement of plantar pressure and shear force, significantly improves the diversity and accuracy of data acquisition, and provides reliable data support for smart wearable devices.

[0022] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of the structure of a sensor for a novel deep learning method for measuring plantar pressure and shear force signal decoupling;

[0024] Figure 2A schematic diagram of the measurement principle of a sensor structure of a novel deep learning method for decoupling plantar pressure and shear force signals of the present invention;

[0025] Figure 3 This is a flow chart of a novel deep learning method for decoupling plantar pressure and shear force signals of the present invention. DETAILED DESCRIPTION

[0026] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0027] Unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0028] See also Figure 1-3 The present invention provides a novel deep learning method for decoupling plantar pressure and shear force signals, which can measure the shear force in 360° direction, including the following steps:

[0029] Step S1, making a sensor structure that can sense pressure and shear force at the same time to improve the diversity and accuracy of data collection. Embed multiple high-precision pressure and shear force sensor structures in the insole, and accurately distribute the sensors at key force-bearing parts of the sole (such as the forefoot, midfoot, and hindfoot) based on the research results of human motion biomechanics.

[0030] Step S11: Design a multi-dimensional pressure and shear force sensor structure. Figure 1 shown.

[0031] Step S12: assemble and manufacture the sensor according to the structural design. The detailed steps are as follows:

[0032] The pressure sensing unit in the sensor structure is assembled and manufactured with reference to the biomechanics of human motion. Figure 1As shown, the sensor structure includes a semicircular contact block arranged below the contact surface, and the bottom of the semicircular contact block is connected to an elastic support column and one end of multiple pressure sensing units, and the other end of the multiple pressure sensing units and the elastic support column are connected to the base. The elastic support column has excellent sensitivity and durability, and can provide stable support and rebound effects during repeated loading and unloading. The pressure sensing unit is made of polymer flexible material, has excellent sensitivity and durability, and can provide stable and accurate data acquisition under various foot force conditions. The changes and differences in the sensing signals can effectively reflect different compression and shear stress states. The flexible pressure sensing unit significantly reduces its interference with foot movement and effectively improves the comfort of wearing. The pressure sensing unit is arranged on the circumferential side ribs, and the accuracy of sensing the direction of shear force is significantly enhanced with the increase in the number of units.

[0033] Use foot 3D scanning technology to obtain accurate foot topography data. Based on the foot topography data, use computer-aided design (CAD) software to design the geometry and size of the pressure sensing unit and shear force sensing unit. Consider mechanical properties and ergonomics during design to ensure that the sensor unit fits tightly with the curved surface of the sole. Design the layout of the sensor unit to ensure that the main force-bearing areas of the sole, such as the forefoot, heel, and arch, are covered by sensor units.

[0034] Step S2, integrated multi-dimensional perception. Data collection is performed through a data collection unit, and the collected data includes plantar pressure and shear force data in the user's daily activities. After the collection is completed, the sensor structure transmits real-time data to a data processing unit via low-power Bluetooth.

[0035] In the integration process of multi-dimensional sensing, the layout and performance of the sensor structure are optimized to achieve accurate measurement of plantar pressure and shear force. The detailed steps are as follows:

[0036] First, the sensor is positioned: when the sensor structure is embedded in the insole, the placement of the sensor unit is scientifically selected based on the mechanical distribution characteristics of the sole. Through biomechanical analysis of the sole, the forefoot, heel and arch are identified as the key areas where pressure and shear force are concentrated. The pressure sensing unit is placed in the forefoot and heel, which are subject to greater vertical pressure when walking and standing; while the shear force sensing unit is preferentially placed in the arch and toe base to capture the shear force generated by the sole under different motion states. This positioning scheme can ensure that the sensor structure covers all major force areas of the sole, fully capture dynamic force conditions, and improve the accuracy of data collection.

[0037] Secondly, ensure close contact: To ensure close contact between the sensor structure and the sole of the foot, ensure that the sensor can be in close contact with the sole surface in various foot movements, thereby ensuring the accuracy of data collection. The fit between the sensor and the sole of the foot is improved, effectively preventing the sensor position from shifting during dynamic movement, and ensuring continuous high-quality data collection.

[0038] Finally, signal transmission and power management: The sensor structure adopts a low-power design to extend the service life of the device. Through low-power wireless communication, the collected pressure and shear force data are transmitted to the external device in real time for processing. This design not only reduces the consumption of the sensor structure battery and improves the overall efficiency of the sensor system, but also ensures the immediacy and stability of data transmission, and improves the practicality and reliability of the system.

[0039] Step S3: Analyze the pressure and shear force signals. The specific steps of the analysis process are as follows:

[0040] Step S31: Perform intelligent feature extraction processing of multi-dimensional force signals. First, perform denoising processing: Adaptive Kalman filtering method is used to perform high-precision denoising on the pressure and shear force signals collected in real time. This method can dynamically adjust the filter parameters according to the pressure and shear force signals collected in real time, so as to effectively remove high-frequency noise in the signal and retain the true mechanical characteristics. Unlike traditional static filters, adaptive Kalman filtering can intelligently identify and filter out different types of noise, such as environmental noise and sensor noise, to ensure the accuracy and stability of the output signal. Secondly, an intelligent feature extraction algorithm is used to identify key multi-dimensional mechanical features from the processed signal, and the multi-dimensional mechanical features include the temporal and spatial distribution characteristics of pressure, shear force, and shear force direction. The intelligent feature extraction algorithm can identify key parameters from the denoised multi-dimensional force signal, including the maximum pressure point, shear force direction, etc., which can reflect the actual situation of the sole force. The intelligent feature extraction algorithm extracts the most representative features by analyzing the multi-dimensional signals collected by the sensor, and uses these features for further analysis and application. It not only improves the speed and accuracy of feature extraction, but also shows good robustness in a variety of complex plantar force environments.

[0041] Step S32: Perform high-precision data cleaning and standardization processing: Use refined data cleaning and standardization processing methods to ensure the integrity and consistency of all data. The data cleaning step includes processing missing values ​​and outliers to prevent data noise from affecting the analysis results. The standardization step ensures that data from different sources and properties can be compared and analyzed under the same standard by unifying the dimensions and scales of the data. Through this process, high-quality data input is guaranteed, providing a reliable foundation for subsequent analysis and model training.

[0042] Therefore, the present invention adopts the above-mentioned new deep learning plantar pressure and shear force signal decoupling measurement method to achieve high-precision measurement of plantar pressure and shear force, significantly improves the diversity and accuracy of data acquisition, and provides reliable data support for smart wearable devices.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A new deep learning method for decoupling plantar pressure and shear force signals, characterized in that: The following steps are involved: Step S1, making a sensor structure; Step S2, integrating multi-dimensional perception; Step S3: analyzing the pressure and shear force signals.

2. According to claim 1, a novel deep learning method for decoupling plantar pressure and shear force signals is characterized in that: The specific steps of step S1 are as follows: Step S11, designing a multi-dimensional pressure and shear force sensor structure; Step S12: assemble and manufacture the sensor according to the structural design.

3. According to claim 2, a novel deep learning method for decoupling plantar pressure and shear force signals is characterized in that: In step S2, data collection is performed by a data collection unit, and the collected data includes plantar pressure and shear force data during the user's daily activities. After the collection is completed, the sensor structure transmits real-time data to the data processing unit via low-power Bluetooth.

4. According to claim 3, a novel deep learning method for decoupling plantar pressure and shear force signals is characterized in that: The specific steps of the analysis process in step S3 are as follows: Step S31: performing intelligent feature extraction processing of multi-dimensional force signals: using an intelligent feature extraction algorithm to identify key multi-dimensional mechanical features from the processed signals, the multi-dimensional mechanical features including pressure magnitude, shear force magnitude and shear force direction; Step S32: Perform high-precision data cleaning and standardization processing: Use refined data cleaning and standardization processing methods to ensure the integrity and consistency of all data.

Citation Information

Patent Citations

  • Sensor and method for detecting normal stress and shear force of sole precisely during walking

    CN110200633A

  • Foot sole stress detection system, device thereof, server, method thereof and storage medium

    CN110680328A

  • Sole soft tissue in-vivo multi-dimensional stress-strain detection equipment

    CN115500814A

  • Foot sole ground reaction force prediction method and system based on WOA-1DCNN-LSTM

    CN116369867A

  • Balance function evaluation method and device based on plantar area shear force distribution measurement

    CN116869479A

Cited By

  • Plantar shear force estimation method and device based on IMU and plantar pressure data

    CN121765280A

  • Method and device for estimating plantar shear forces based on imu and plantar pressure data

    CN121765280B

  • Grid type ultrathin pressure and shear force synchronous measurement sensor and signal decoupling method

    CN122192561A