Pressure array based geriatric steering steady state assessment method and system

Abstract

The application provides an elderly turning steady state evaluation method and system based on a pressure array, relates to the technical field of elderly ability evaluation, and comprises the following steps: acquiring the plantar pressure image of a tested person collected by a pressure array in response to the tested person standing on a platform type pressure array with direction marks and performing a turning task; identifying the standing posture orientation of the tested person relative to the pressure array based on the plantar pressure image to obtain the standing posture orientation identification result of the tested person; extracting the multi-dimensional plantar force feature of the tested person when performing the turning task based on the plantar pressure image; evaluating the turning ability of the tested person based on the standing posture orientation identification result and the multi-dimensional plantar force feature to obtain the turning ability evaluation result of the tested person; and generating the turning steady state evaluation result of the tested person based on the standing posture orientation identification result and the turning ability evaluation result. The application can solve the problem that the traditional plantar pressure analysis technology cannot accurately evaluate the standing posture orientation and turning ability of the elderly.

Description

Technical Field

[0001] This application relates to the field of elderly capacity assessment technology, and in particular to a method and system for assessing elderly people’s shift to steady state based on a pressure array. Background Technology

[0002] Plantar pressure analysis is a core technology in biomechanical testing, motor function assessment, and health risk screening. It is widely used in various scenarios, including clinical gait analysis, balance assessment, fall risk assessment, rehabilitation training effect feedback, and diabetic foot pressure monitoring. Plantar pressure analysis technology collects spatiotemporal distribution data of plantar force and extracts quantitative indicators such as peak plantar pressure, plantar contact area, stride length, stride width, gait cycle, and pressure center trajectory. This provides objective data support for motor function assessment and health status evaluation. The accuracy of the indicator calculations and their adaptability to different scenarios directly determine the reliability of the assessment results.

[0003] In related technologies, the parameter system and calculation logic of plantar pressure analysis methods are designed around two classic testing paradigms: straight-line walking and static standing, assuming that the test subject has a stable forward direction and periodic gait patterns. However, in the scenario of assessing the function of turning in place, the test subject does not have a fixed forward direction. During the movement, the foot will exhibit complex movement states such as planar rotation, small step adjustments, single-leg support transitions, and alternating weight transfer between both feet. The spatiotemporal distribution pattern of foot force is fundamentally different from that of straight-line walking. Traditional analysis methods cannot adapt to this type of non-periodic, dynamically changing foot movement pattern. They are unable to accurately extract core features such as the foot rotation amplitude, weight adjustment path, and support phase switching sequence during the turning process, nor can they complete the quantitative assessment of the test subject's standing orientation control ability and turning movement flexibility. Ultimately, this leads to a significant deviation between the assessment results of the elderly's standing orientation and turning ability and their actual functional level.

[0004] In summary, traditional plantar pressure analysis technology is limited by the paradigm design of straight walking and static standing, and cannot adapt to the complex foot movement state in the task of turning on the spot. It is difficult to achieve accurate assessment of the standing orientation and turning ability of the elderly, which restricts the application depth of plantar pressure analysis technology in scenarios such as fall risk screening and sports rehabilitation assessment of the elderly. Summary of the Invention

[0005] The main objective of this application is to propose a method and system for steady-state assessment of elderly turning based on a pressure array. This aims to solve the problem that traditional plantar pressure analysis technology is difficult to accurately assess the standing orientation and turning ability of the elderly, thereby improving the application depth of plantar pressure analysis technology in scenarios such as fall risk screening and sports rehabilitation assessment for the elderly.

[0006] To achieve the above objectives, a first aspect of this application proposes a method for assessing the steady-state shift of elderly individuals based on a pressure array, the method comprising: In response to the subject standing on the pressure array and performing a turning task, the pressure array acquires the plantar pressure image of the subject; the pressure array is a platform-type pressure array with directional markings. Based on the plantar pressure images, the standing orientation of the subject relative to the pressure array is identified to obtain the standing orientation identification result of the subject; and, based on the plantar pressure images, multidimensional plantar force features of the subject when performing a turning task are extracted. The subject's turning ability is evaluated based on the standing posture orientation recognition results and the multidimensional plantar force characteristics, resulting in the subject's turning ability evaluation results. Based on the standing posture orientation recognition results and the turning ability assessment results, the subject's turning steady-state assessment results are generated.

[0007] In some embodiments, the extraction of multidimensional plantar force features of the subject performing a turning task based on the plantar pressure image includes at least two of the following: Based on the plantar pressure image, the pressure regions of the left and right feet of the subject are tracked and identified when performing a turning task; the multidimensional plantar force characteristics include the pressure regions of the left and right feet. The foot pressure image is used to track and identify the foot state of the subject when performing a turning task; the multidimensional foot force characteristics include the foot state, which includes at least one of the following: stable contact state, moving state, foot lifting state, re-landing state, transient absence state, and adhesion state of both feet. The pressure center of the subject is dynamically calculated based on the plantar pressure image, and the three-dimensional center of gravity of the subject is calculated based on the pressure center; the multidimensional plantar force characteristics include the pressure center and the three-dimensional center of gravity. The pressure distribution of the subject when performing a turning task is calculated based on the plantar pressure image; the multidimensional plantar force characteristics include the pressure distribution.

[0008] In some embodiments, the method further includes: A human coordinate system is established based on the standing posture orientation recognition results; The calculation of the pressure distribution during the subject's turning task based on the plantar pressure image includes: Based on the human body coordinate system, the plantar pressure image is divided into a forefoot region and a hindfoot region; The pressure distribution of the subject when performing a turning task is calculated based on the pressure in the forefoot region and the hindfoot region, respectively.

[0009] In some embodiments, the assessment of the subject's turning ability based on the standing posture orientation recognition result and the multidimensional plantar force characteristics to obtain the subject's turning ability assessment result includes: The standing posture orientation recognition results and the multidimensional plantar force characteristics are subjected to index standardization processing to obtain the standardized scores of the multidimensional turning ability index of the test subject. The standardized scores are weighted and summed according to preset weights to obtain the test subject’s comprehensive turning ability score. Based on the comprehensive score of the turning ability, the test subject is mapped to a turning ability level to obtain the turning ability level of the test subject, and a fall risk warning is output based on the turning ability level; the turning ability assessment result includes the turning ability level and the fall risk warning, and the turning ability level includes at least one of excellent, good, average, poor and very poor.

[0010] In some embodiments, the standardization of the standing posture orientation recognition result and the multidimensional plantar force characteristics to obtain the standardized scores of the subject's multidimensional turning ability indicators includes: Determine the multidimensional turning ability index corresponding to the standing posture orientation recognition result and the multidimensional plantar force characteristics; The multidimensional turning ability indicators are subjected to extreme value standardization processing based on a preset reference indicator distribution to obtain the standardized scores of each multidimensional turning ability indicator; the reference indicator distribution includes the multidimensional turning ability indicators of monitoring personnel of the same age group as the test subject.

[0011] In some embodiments, the method further includes: Based on the standing posture orientation recognition result and the turning ability assessment result, the state of the subject performing the turning task is identified to obtain the real-time turning state of the subject. The subject's turning ability is evaluated based on the standing posture orientation recognition result, the multidimensional plantar force characteristics, and the real-time turning state, resulting in the subject's turning ability evaluation result.

[0012] In some embodiments, the step of identifying the subject's standing orientation relative to the pressure array based on the plantar pressure image to obtain the subject's standing orientation identification result includes at least one of the following: The plantar pressure image is input into a preset standing posture orientation recognition model for standing posture orientation recognition processing, and the standing posture orientation recognition result of the subject output by the standing posture orientation recognition model is obtained. Gait analysis is performed on the subject during a turning task based on the plantar pressure image to obtain the subject's gait analysis results. Based on the gait analysis results, rule recognition processing is performed on the subject's standing orientation to obtain the subject's standing orientation recognition results. The gait analysis results include foot axis angle, toe and heel direction and / or relative position of the two feet.

[0013] To achieve the above objectives, a second aspect of this application proposes a stress array-based system for assessing the steady-state shift of elderly individuals, the system comprising: The pressure array acquisition module is used to acquire the plantar pressure image of the subject in response to the subject standing on the pressure array and performing a turning task; the pressure array is a platform-type pressure array with directional markings. The standing posture orientation recognition module is used to identify the standing posture orientation of the subject relative to the pressure array based on the plantar pressure image, and obtain the standing posture orientation recognition result of the subject. The plantar force feature extraction module is used to extract multidimensional plantar force features of the subject when performing a turning task based on the plantar pressure image. The evaluation result output module is used to evaluate the subject's turning ability based on the standing posture orientation recognition result and the multidimensional plantar force characteristics, and obtain the subject's turning ability evaluation result; and to generate the subject's steering steady-state evaluation result based on the standing posture orientation recognition result and the turning ability evaluation result.

[0014] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the stress array-based method for assessing the steady-state turning of the elderly as described in the first aspect.

[0015] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the stress array-based method for assessing the steady-state turning of the elderly as described in the first aspect.

[0016] To achieve the above objectives, a fifth aspect of the present application provides a computer program product comprising a computer program that, when executed by a processor, implements the stress array-based method for assessing the steady-state turning of the elderly as provided in the first aspect above.

[0017] The present application proposes a method, system, electronic device, computer-readable storage medium, and computer program product for assessing the turning steady state of the elderly based on a pressure array. This method acquires plantar pressure images of the subject in response to a subject performing a turning task while standing on the pressure array. The pressure array is a platform-type pressure array with directional markings. Based on the plantar pressure images, the method identifies the subject's standing orientation relative to the pressure array, obtaining a standing orientation identification result. Furthermore, it extracts multidimensional plantar force features of the subject during the turning task based on the plantar pressure images. Based on the standing orientation identification result and the multidimensional plantar force features, it assesses the subject's turning ability, obtaining a turning ability assessment result. Finally, it generates a turning steady state assessment result for the subject based on the standing orientation identification result and the turning ability assessment result.

[0018] Compared to traditional plantar pressure analysis techniques, this application embodiment acquires plantar pressure images of the subject when performing a turning task using a platform-type pressure array with directional markers. Based on these plantar pressure images, the subject's standing orientation relative to the pressure array is identified to obtain the standing orientation identification result. Furthermore, multidimensional plantar force features of the subject when performing the turning task are extracted based on the plantar pressure images. Thus, the subject's turning ability is evaluated based on the standing orientation identification result and the multidimensional plantar force features to obtain the subject's turning ability evaluation result. Finally, the subject's steering steady-state evaluation result is generated based on the standing orientation identification result and the turning ability evaluation result. Thus, the embodiments of this application can accurately identify the standing orientation of the test subject using a pressure array, and can adapt to the complex foot movement state in the task of turning in place. Based on the plantar pressure image, the turning ability can be objectively quantified. That is, the embodiments of this application can realize the accurate assessment of the standing orientation and turning ability of the elderly and other test subjects in the turning task, which solves the problem that traditional plantar pressure analysis technology is difficult to realize the accurate assessment of the standing orientation and turning ability of the elderly. This can improve the application depth of plantar pressure analysis technology in scenarios such as fall risk screening and sports rehabilitation assessment for the elderly. Attached Figure Description

[0019] Figure 1 A flowchart illustrating the steps of the pressure array-based method for assessing the steady-state turning of the elderly in some embodiments of this application; Figure 2 for Figure 1 A detailed flowchart of step S102; Figure 3 for Figure 1 A schematic diagram of another detailed step in step S102; Figure 4A flowchart illustrating the steps of the pressure array-based elderly turning steady-state assessment method provided in this application in other embodiments; Figure 5 for Figure 1 A detailed flowchart of step S103; Figure 6 A flowchart illustrating a complete embodiment of the pressure array-based method for assessing the steady-state turning of the elderly, as provided in this application. Figure 7 A schematic diagram of the system flow involved in one embodiment of the pressure array-based method for assessing the steady-state turning of the elderly provided in this application; Figure 8 A schematic diagram of the pressure array coordinate system and the human body coordinate system involved in one embodiment of the pressure array-based method for assessing the turning steady state of the elderly provided in this application; Figure 9 A schematic diagram of the state machine for the stationary turning task in one embodiment of the pressure array-based elderly turning steady-state assessment method provided in this application; Figure 10 A schematic diagram of the center of pressure (COP) trajectory index in one embodiment of the pressure array-based elderly turning steady-state assessment method provided in this application; Figure 11 A schematic diagram of the foot axis direction and forefoot / hindfoot region division involved in one embodiment of the pressure array-based elderly turning steady-state assessment method provided in this application; Figure 12 A schematic diagram of the structure of the elderly turning steady-state assessment system based on pressure array provided in the embodiments of this application; Figure 13 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0023] First, a brief explanation of the relevant technical terms involved in the embodiments of this application will be given.

[0024] Pressure array: A detection device consisting of multiple pressure sensing units arranged in a two-dimensional array, used to collect the pressure distribution on the soles of the feet when a person is standing or moving.

[0025] Pressure frame: A set of pressure data collected by the pressure array at a certain moment, which can be represented as a two-dimensional pressure matrix.

[0026] Pressure image: Data obtained by displaying or processing a two-dimensional pressure matrix in image form. The pressure value can correspond to the image grayscale or color intensity.

[0027] Center of Pressure (COP): This represents the center position of all pressure points in a pressure array after weighting by pressure magnitude. COP can be used to describe the center of gravity control of a person standing or moving.

[0028] Convolutional Neural Network (CNN): A neural network model commonly used in image recognition. In this embodiment, it can be used to identify which side of the pressure array the subject (such as an elderly person) is facing.

[0029] Standing orientation: The direction the subject faces when standing on the pressure array. For example, a square pressure array may have four sides, A, B, C, and D. The standing orientation can be expressed as facing side A, B, C, or D.

[0030] Human coordinate system: A coordinate system established based on the subject's current standing posture and orientation. The front of the body is considered "forward," the back is considered "backward," and the sides are considered "left" and "right," respectively. This coordinate system differs from the fixed coordinate system of the pressure pad.

[0031] Foot axis direction: Represented by the long axis direction of the pressure area of ​​the foot, approximately corresponding to the direction from the heel to the toe.

[0032] Forefoot area: The pressure zone of the foot near the toes, which can be divided according to the direction of the foot axis and the proportion of foot length.

[0033] The hindfoot region: the pressure zone of the foot near the heel, which can be divided according to the direction of the foot axis and the proportion of foot length.

[0034] Foot adjustment count: During a turn, the number of times a foot changes from stable contact to movement, pressure reduction, or lifting, and then re-establishes stable contact. This indicator reflects whether the subject is taking small steps or making multiple adjustments when turning.

[0035] Left-right weight asymmetry index: an index used to describe the difference in weight distribution between the left and right feet, which can be calculated from the total pressure on the left foot and the total pressure on the right foot.

[0036] Anterior-posterior pressure offset index: an index used to describe the anterior-posterior pressure offset of the human body, which can be calculated from the anterior pressure and posterior pressure.

[0037] State machine: A logical structure used to determine the current stage of a turning task. In the embodiments of this application, it may include states such as preparing to stand, turning in progress, reaching the target orientation, stabilizing and recovering, and stabilizing and ending.

[0038] Foot Progression Angle (FPA): This usually refers to the angle between the straight walking direction and the long axis of the foot. Since there is no stable forward direction during a turn in place, this application does not use FPA as a core turning indicator.

[0039] Next, the overall concept of the embodiments of this application will be briefly described.

[0040] As the population ages, assessments of basic motor skills, fall risk screening, and rehabilitation training for the elderly are becoming increasingly important components of elderly care institutions, community health services, and home-based health management. Basic motor skills in the elderly are typically related to factors such as standing stability, ability to transfer from sitting to standing, walking ability, turning ability, and dynamic balance.

[0041] Turning is a frequent action in the daily activities of the elderly, such as turning from the bedside to a wheelchair, turning from a chair to the direction of travel, and changing direction in confined spaces. Compared to walking in a straight line, turning usually requires the elderly to complete a change in body orientation, reposition their feet, shift their center of gravity, and restore postural stability in a short period of time. For elderly people with decreased balance, insufficient lower limb strength, reduced unilateral weight-bearing capacity, or neurological diseases, they are prone to taking small, dragging steps, excessive shift in center of gravity, uneven weight-bearing between the left and right sides, and instability after turning.

[0042] Traditional assessments of elderly abilities often rely on manual observation and scale scoring. Taking GB / T 42195-2022, the "Standard for Assessment of Elderly Abilities," as an example, its assessment content includes self-care ability, basic motor skills, mental state, sensory perception, and social participation. Basic motor skills are closely related to the elderly's transfer, walking, and balance control. However, in actual assessments, manual observation can usually only determine whether the elderly can complete a certain movement, whether assistance is needed, and whether the movement is noticeably unstable. It is difficult to accurately record changes in the center of pressure during turning, differences in weight-bearing between the left and right feet, anterior-posterior pressure shift, number of foot adjustments, and recovery time after turning.

[0043] A pressure array is a two-dimensional sensing device composed of multiple pressure sensing units. When a subject stands or walks on the pressure array, the system obtains a plantar pressure matrix, which represents the pressure value at each pressure point at a given moment. By continuously acquiring the pressure matrix, a sequence of plantar pressure images can be obtained, allowing for further analysis of information such as foot regions, pressure centers, pressure ratios between the left and right feet, and changes in pressure distribution. Therefore, pressure arrays provide an objective data basis for assessing the basic motor abilities of older adults.

[0044] In the field of plantar pressure analysis, common research and applications include gait analysis, balance assessment, fall risk assessment, rehabilitation training feedback, and diabetic foot pressure monitoring. Common indicators include peak plantar pressure, contact area, stride length, stride width, gait cycle, and center of pressure trajectory.

[0045] Here, the center of pressure (COP) represents the center position of all pressure points on the pressure array after weighting according to pressure magnitude. The COP trajectory in consecutive frames can reflect the center of gravity control of a person standing or moving. The formula for calculating COP is: COP_x = Σ(P_i × x_i) / ΣP_i, COP_y = Σ(P_i × y_i) / ΣP_i.

[0046] Where P_i represents the pressure value of the i-th pressure point, and x_i and y_i represent the coordinates of the pressure point in the pressure array coordinate system.

[0047] Traditional plantar pressure analysis typically focuses on straight-line walking or static standing. In straight-line walking scenarios, the foot progression angle (FPA) is often used. However, in stationary turning tasks, subjects do not have a stable direction of movement; their feet may exhibit complex states such as rotation, small steps, single-leg support, and alternating adjustments between both feet. Therefore, traditional FPA is not suitable as a core indicator for assessing stationary turning ability.

[0048] To address the aforementioned problems, this application proposes a method, system, electronic device, computer-readable storage medium, and computer program product for assessing the steady-state turning of elderly individuals based on a pressure array, aiming to achieve the following objectives: 1. Automatically identify the direction of the pressure array boundary that the elderly person is currently facing using pressure array data; 2. Automatically determine the start of the turn, the target orientation, and the regain of footing during the turning task; 3. Extract turning efficiency indicators such as turning time, number of foot adjustments, and difference in the number of adjustments between the left and right feet; 4. Extract center of gravity control indicators such as COP stabilization time, maximum COP offset, COP trajectory length, and COP swing area after turning; 5. Extract the left-right load ratio, left-right load asymmetry index, front-back pressure offset index, and pressure distribution change before and after the turn within the stability window before and after the turn. 6. Establish a human body coordinate system based on the standing posture and orientation to ensure consistent front-to-back pressure analysis under different orientations; 7. To provide objective quantitative data for assessing the basic motor abilities of the elderly, assessing their turning ability, and assisting in the screening of fall risk.

[0049] In this embodiment, a pressure array is used as the acquisition device. The elderly person stands on the pressure array, and the system continuously acquires the plantar pressure matrix. Each frame of the pressure matrix can be regarded as a pressure image. The system analyzes the pressure images, identifies the direction the elderly person is currently facing the pressure array, and extracts the pressure areas of the left and right feet, the foot movement state, the pressure center COP, and the pressure distribution.

[0050] At the start of the test, the system prompts the elderly person to stand facing a designated side, such as side A. The system collects initial static pressure data, obtaining the initial orientation, initial left and right foot regions, initial COP, initial left-right pressure ratio, and initial front-back pressure ratio. Subsequently, the system issues a turning command, such as "Please turn 90° to the right." During the turning process, the system continuously identifies the current orientation, COP, left and right foot status, and pressure distribution in each frame, and determines the test phase through a state machine.

[0051] The turning task is considered complete when the system determines that the elderly person has reached the target orientation and that COP, foot position, and pressure distribution have stabilized again. The system then outputs a turning ability assessment index.

[0052] Next, based on the overall concept of the above-described embodiments of this application, specific embodiments of the elderly turning steady-state assessment method, system, electronic device, computer-readable storage medium, and computer program product based on pressure array proposed in the embodiments of this application are presented. First, the various specific embodiments of the elderly turning steady-state assessment method based on pressure array proposed in the embodiments of this application are described in detail.

[0053] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0054] It should be noted that the stress array-based method for assessing the steady-state turning of the elderly provided in this application relates to the field of elderly capacity assessment technology. This stress array-based method for assessing the steady-state turning of the elderly provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal may be a product such as an elderly ability assessment device, a health assessment device for elderly care institutions, a fall risk screening device for elderly people in the community, a rehabilitation training assessment device, a smart plantar pressure detection platform, a smart mat, or a device that integrates / operates a pressure pad assessment system / smart gait assessment system / balance assessment system. Specific product forms may include: a pressure array assessment mat (composed of multiple pressure sensing units, used to collect plantar pressure data during the elderly's standing and turning processes), an integrated elderly ability assessment machine (including a pressure array mat, a display screen, a voice prompt module, a data processing module, and an assessment report output module), a screening device for elderly care homes / communities (used to assist in screening the elderly for standing stability, turning ability, and fall risk), a rehabilitation training assessment system (used to assess changes in the elderly's turning ability, balance ability, and weight-bearing control ability before and after rehabilitation training), and a home health monitoring device (as a smart mat or smart pressure pad, used to record the elderly's standing, turning, and center of gravity control over a long period).

[0055] Furthermore, the terminal can also be an electronic device such as a smartphone, tablet, laptop, or desktop computer. The server can be the backend server terminal device of the terminal, which can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The software can be an application, a computer program, and a storage medium carrying the computer program that implements the stress array-based method for assessing the steady-state shift of the elderly. It should be understood that, based on different design needs of practical applications, in different feasible embodiments, the terminal, server, and software of the stress array-based method for assessing the steady-state shift of the elderly provided in this application can also be other forms not listed here. The stress array-based method for assessing the steady-state shift of the elderly provided in this application does not specifically limit these aspects.

[0056] Furthermore, this application can also be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, personal computers (PCs), minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via communication networks. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0057] For ease of understanding and explanation, the following text will use the pressure array-based elderly turning steady-state assessment system (hereinafter referred to as the "system") as an example to illustrate the various specific embodiments of this application. The implementation of any of the above-described subject matter using the pressure array-based elderly turning steady-state assessment system provided in the embodiments of this application can refer to the implementation process of the pressure array-based elderly turning steady-state assessment system described below.

[0058] Please refer to Figure 1 , Figure 1 The flowchart illustrates the steps of the pressure array-based method for assessing the steady-state turning of the elderly, as provided in this application, in some embodiments. It should be understood that, although... Figure 1 The flowcharts illustrating subsequent steps show the execution order of some method steps. However, based on different design needs in practical applications, the stress array-based method for assessing the steady-state turning of the elderly provided in this application can, of course, employ a different execution order of method steps than shown in the figures. That is, Figure 1 The order of the method steps shown does not constitute a limitation on the execution logic order of the pressure array-based elderly turning steady-state assessment method provided in the embodiments of this application. Any other method based on... Figure 1 Reasonable changes to the sequence of steps shown should be included within the protection scope of the pressure array-based steady-state assessment method for elderly people provided in the embodiments of this application.

[0059] like Figure 1 As shown, in some embodiments, the method for assessing the turning to steady state of the elderly based on a pressure array provided in this application may include steps S101 to S104 as shown below.

[0060] Step S101: In response to the subject standing on the pressure array and performing a turning task, acquire the plantar pressure image of the subject collected by the pressure array; the pressure array is a platform-type pressure array with directional markings.

[0061] It should be noted that the pressure array can be square, rectangular, polygonal, or other platform-type pressure arrays with directional markings. Taking a square pressure array as an example, the four sides can be defined as side A, side B, side C, and side D, corresponding to four preset orientations. The side the test subject faces indicates the direction their body is facing. Furthermore, the turning task can be a stationary turning task, such as turning 90° to the right, turning 90° to the left, turning 180°, continuous turning, or multi-directional turning tasks, etc.

[0062] The system can perform a turning steady-state assessment (including standing orientation and turning ability assessment) on elderly subjects. During the process of the subject standing and turning on the pressure array, the system can acquire continuous pressure frames in real time through the pressure array. Since each frame of the pressure matrix can be regarded as a pressure image, the system can obtain continuous plantar pressure images of the subject.

[0063] In some embodiments, the system includes a pressure array acquisition module. The system can acquire continuous pressure frames during the subject's standing and turning processes through the pressure array acquisition module, thereby obtaining continuous plantar pressure images of the subject.

[0064] Step S102: Based on the plantar pressure image, identify the standing orientation of the subject relative to the pressure array to obtain the standing orientation identification result of the subject; and, based on the plantar pressure image, extract the multidimensional plantar force features of the subject when performing a turning task.

[0065] After acquiring continuous plantar pressure images of the subject, the system can further analyze these images to identify the subject's current orientation towards the pressure array, thus obtaining the subject's standing posture orientation. Furthermore, the system can extract features from these plantar pressure images, such as the pressure areas of the subject's left and right feet, foot movement state, pressure center COP, and pressure distribution, as multi-dimensional plantar force characteristics when the subject performs a turning task.

[0066] In some embodiments, after acquiring continuous plantar pressure images of the subject, the system can further preprocess the plantar pressure images and then use the preprocessed image data to perform subsequent operations such as standing posture orientation recognition and plantar force feature extraction.

[0067] In some embodiments, the system includes a pressure image preprocessing module. Assuming the system acquires continuous plantar pressure images of the subject as the original pressure matrix, the system can process the original pressure matrix using the pressure image preprocessing module, including: 1. Remove pressure points below the noise threshold; 2. Smooth the pressure matrix; 3. Perform threshold segmentation to extract the effective pressure region; 4. Perform connected component analysis to obtain candidate foot regions; Noise areas with insufficient filtration area or low pressure.

[0068] Step S103: Based on the standing posture orientation recognition result and the multidimensional plantar force characteristics, evaluate the subject's turning ability to obtain the subject's turning ability evaluation result.

[0069] After the system obtains the standing orientation recognition result and multidimensional plantar force characteristics of the subject, it can evaluate the subject's turning ability based on the standing orientation recognition result and multidimensional plantar force characteristics, thereby obtaining the subject's turning ability evaluation result.

[0070] In some embodiments, the assessment result of the subject's turning ability can be a comprehensive score representing the turning ability based on the standing orientation recognition result and multiple indicators corresponding to multidimensional plantar force characteristics. The system can calculate the subject's turning motion time, number of foot adjustments, difference in the number of adjustments between the left and right feet, maximum COP offset, COP trajectory length, average COP velocity, peak COP velocity, and COP swing area based on the standing orientation recognition result and multidimensional plantar force characteristics. It can also calculate the left-right weight ratio and front-back pressure ratio of the subject in the pre-turn and post-turn stability windows, and calculate the changes in left-right weight and front-back pressure before and after the turn. The calculated multiple indicators are then standardized to obtain standardized scores for each indicator. Finally, the standardized scores are weighted and summed according to preset weights to obtain the subject's comprehensive turning ability score.

[0071] Step S104: Generate the subject's turning steady-state assessment result based on the standing posture orientation recognition result and the turning ability assessment result.

[0072] After obtaining the assessment results of the subject's turning ability, the system can further generate the subject's turning steady-state assessment results based on the assessment results of the turning ability and the previously identified standing posture and orientation. For example, if the subject is at risk of falling when turning to the right, it is recommended to strengthen balance training.

[0073] In some embodiments, the turning steady-state assessment results may include a turning ability level and / or fall risk assistance prompts. The system may also directly output the turning ability level and / or fall risk assistance prompts for the test subject based on a comparison between the subject's overall turning ability score and a preset level threshold.

[0074] In this embodiment, the system acquires continuous pressure frames during the subject's standing and turning process on the pressure array, obtained in real time by the pressure array, to obtain continuous plantar pressure images of the subject. These plantar pressure images are then analyzed to identify the subject's current orientation towards the pressure array, thus obtaining the subject's standing orientation recognition result. Furthermore, features such as the pressure areas of the subject's left and right feet, foot movement state, pressure center COP, and pressure distribution are extracted from the plantar pressure images as multidimensional plantar force features when the subject performs the turning task. Subsequently, based on the standing orientation recognition result and the multidimensional plantar force features, the subject's turning ability is evaluated to obtain the subject's turning ability evaluation result. Finally, based on the turning ability evaluation result and the previously identified standing orientation recognition result, the subject's steering steady-state evaluation result is generated.

[0075] Thus, compared to traditional plantar pressure analysis techniques, the embodiments of this application can acquire plantar pressure images of the subject when performing a turning task using a platform-type pressure array with directional markers, and identify the subject's standing orientation relative to the pressure array based on the plantar pressure images to obtain the standing orientation identification result. Furthermore, multidimensional plantar force characteristics of the subject when performing a turning task are extracted based on the plantar pressure images, thereby evaluating the subject's turning ability based on the standing orientation identification result and the multidimensional plantar force characteristics to obtain the subject's turning ability evaluation result. Finally, the subject's steering steady-state evaluation result is generated based on the standing orientation identification result and the turning ability evaluation result. In other words, the embodiments of this application can accurately identify the standing orientation of the test subject using a pressure array and can adapt to the complex foot movement state in the task of turning in place. Based on the plantar pressure image, the turning ability can be objectively quantified. That is, the embodiments of this application can achieve accurate assessment of the standing orientation and turning ability of the elderly and other test subjects in the turning task, which solves the problem that traditional plantar pressure analysis technology is difficult to achieve accurate assessment of the standing orientation and turning ability of the elderly. This can improve the application depth of plantar pressure analysis technology in scenarios such as fall risk screening and sports rehabilitation assessment for the elderly.

[0076] Please refer to Figure 2 , Figure 2 for Figure 1 A detailed flowchart of step S102.

[0077] like Figure 2 As shown, in some embodiments, the step of "identifying the standing orientation of the subject relative to the pressure array based on the plantar pressure image and obtaining the standing orientation identification result of the subject" in step S102 above may include at least one of steps S201 and S202 as shown below.

[0078] Step S201: Input the plantar pressure image into a preset standing posture orientation recognition model for standing posture orientation recognition processing, and obtain the standing posture orientation recognition result of the subject output by the standing posture orientation recognition model.

[0079] It should be noted that the standing posture orientation recognition model can be trained based on a convolutional neural network. In addition, the standing posture orientation recognition model can also adopt image classification algorithms, object detection models, semantic segmentation models, foot axis angle rule algorithms or multi-frame voting algorithms.

[0080] When the system identifies the standing orientation of the subject relative to the pressure array based on the plantar pressure image, it can input the plantar pressure image into a preset standing orientation recognition model for standing orientation recognition processing, thereby obtaining the standing orientation recognition result of the subject output by the standing orientation recognition model.

[0081] Step S202: Based on the plantar pressure image, perform gait analysis on the subject performing a turning task to obtain the gait analysis result of the subject, and perform rule recognition processing on the subject's standing orientation based on the gait analysis result to obtain the subject's standing orientation recognition result; the gait analysis result includes foot axis angle, toe and heel direction and / or the relative position of the two feet.

[0082] When the system identifies the subject's standing orientation relative to the pressure array based on plantar pressure images, it can also perform gait analysis on the subject's turning task based on the plantar pressure images to obtain the subject's gait analysis results, namely, the subject's foot axis angle, toe and heel direction and / or the relative position of the two feet. Then, the system can further perform rule recognition processing on the subject's standing orientation based on these gait analysis results to obtain the subject's standing orientation recognition result.

[0083] In some embodiments, the system includes a standing orientation recognition module. When identifying the standing orientation of a subject relative to a pressure array based on plantar pressure images, the system can use this module to determine which side of the pressure array the elderly person is currently facing. One implementation of the standing orientation recognition module is using a convolutional neural network (CNN). The system inputs each frame of the pressure image into the CNN, and the network outputs the probability that the current pressure image belongs to one of four orientations: A, B, C, or D. The system selects the orientation with the highest probability as the current standing orientation and stores the orientation confidence level. For example, if the output is: "Probability of side A: 0.05, Probability of side B: 0.90, Probability of side C: 0.03, Probability of side D: 0.02", the system determines that the elderly person is currently facing side B. Another implementation of the standing orientation recognition module is based on rule recognition using foot axis angles, toe-heel directions, and the relative positions of the feet. Specifically, the system can calculate the principal axis direction for the pressure areas of the left and right feet and combine this with the initial orientation and continuous angle changes during the turning process to determine the current standing orientation.

[0084] In some embodiments, to improve stability, the system can also perform voting or smoothing on the standing orientation recognition results of multiple consecutive frames of plantar pressure images to avoid misjudgment in a single frame.

[0085] Please refer to Figure 3 , Figure 3 for Figure 1 A schematic diagram of another detailed step in step S102.

[0086] like Figure 3As shown, in some embodiments, the step of "extracting multidimensional plantar force features of the subject when performing a turning task based on the plantar pressure image" in step S102 above may include at least two of the steps S301 to S304 shown below.

[0087] Step S301: Based on the plantar pressure image, track and identify the pressure areas of the left and right feet of the subject when performing a turning task; the multidimensional plantar force characteristics include the pressure areas of the left and right feet.

[0088] When extracting multidimensional plantar force features of the subject performing a turning task from plantar pressure images, the system can segment the left and right foot pressure regions of the subject performing the turning task from the plantar pressure images and continuously track the left and right foot pressure regions in consecutive image frames. Thus, the tracked and identified left and right foot pressure regions are used as one of the multidimensional plantar force features of the subject performing the turning task.

[0089] In some embodiments, the system includes a left and right foot region recognition and tracking module. This module can segment the left and right foot pressure regions from plantar pressure images, maintaining consistency between the left and right feet across consecutive frames, thereby obtaining the left and right foot pressure regions when the subject performs a turning task.

[0090] In some embodiments, when identifying the left and right foot regions, the left and right foot region identification and tracking module is not limited to connected component analysis, but may also employ clustering algorithms, object detection models, or semantic segmentation models. Similarly, when tracking the identity of the left and right feet, the left and right foot region identification and tracking module is not limited to nearest neighbor matching, but may also employ Hungarian matching, Kalman filtering, multi-target tracking algorithms, or rule-based tracking based on temporal continuity.

[0091] In some embodiments, in the initial stage, the system can prompt the subject to stand facing a designated side, thus ensuring the front of the body is known. The system can then determine the left and right feet based on the left-right direction in the human coordinate system. Subsequently, during the turning process, the system can match the foot region in the current frame with the left and right feet in the previous frame based on features such as the center position, area, total pressure, and foot axis angle of the foot region in the plantar pressure image, thereby maintaining consistency in the identification of the left and right feet.

[0092] In some embodiments, the system may employ connected component analysis to identify the left and right foot regions. Alternatively, the system may use clustering algorithms, object detection models, or semantic segmentation models to identify the left and right foot regions.

[0093] In some embodiments, the system may employ matching methods such as nearest neighbor matching, Hungarian matching, and Kalman filter tracking to match the foot region in the current frame with the left and right feet in the previous frame.

[0094] Step S302: Based on the plantar pressure image tracking, identify the foot state of the subject when performing the turning task; the multidimensional plantar force characteristics include the foot state, which includes at least one of the following: stable contact state, moving state, foot lifting state, re-landing state, transient absence state, and adhesion state of both feet.

[0095] When extracting multidimensional plantar force features from plantar pressure images of the subject performing a turning task, the system can track and identify the foot state of the subject performing the turning task based on the plantar pressure images: stable contact state, moving state, foot lifting state, re-landing state, transient absence state, and adhesion state of both feet. Then, one or more foot states identified are used as one of the multidimensional plantar force features of the subject performing the turning task.

[0096] In some embodiments, the system includes a foot state determination module. This module can be used to determine the state of each foot of the subject in each frame of plantar pressure images. The determination criteria used by the foot state determination module may include: changes in the center position of the foot region, changes in total foot pressure, changes in the foot contact area, changes in the foot axis angle, and the matching relationship between the current foot region and the foot region in the previous frame.

[0097] In some embodiments, the system can record a foot adjustment when it identifies a subject's foot (left or right foot) from a stable contact state to a moving, decompression, or lifting state based on plantar pressure images, and then returns to a stable contact state.

[0098] Step S303: Dynamically calculate the pressure center of the subject when performing a turning task based on the plantar pressure image, and calculate the three-dimensional center of gravity of the subject based on the pressure center; the multi-dimensional plantar force characteristics include the pressure center and the three-dimensional center of gravity.

[0099] When extracting multidimensional plantar force features from plantar pressure images during a subject's turning task, the system can calculate the center of pressure (COP) at that moment based on the plantar pressure images. This calculated COP is then used as one of the multidimensional plantar force features during the turning task. Subsequently, the system can also calculate the subject's three-dimensional center of mass (COM) based on the COP, and this COM is also used as one of the multidimensional plantar force features during the turning task.

[0100] In some embodiments, the system includes a COP dynamic stability calculation module. The system can use this module to calculate the changes in the center of pressure (COP) of the test subject during a turning task based on continuous plantar pressure images. The coordinates of the COP of the center of pressure in each frame of the plantar pressure image can be calculated using the following formula: COP_x = Σ(P_i × x_i) / ΣP_i, COP_y = Σ(P_i × y_i) / ΣP_i.

[0101] Where P_i is the pressure value of the i-th pressure point, and x_i and y_i are the coordinates of the pressure point in the pressure array coordinate system.

[0102] In some embodiments, the system can use the COP dynamic stability index corresponding to the center of pressure (COP) to evaluate the subject's ability to control their center of gravity during a turning task, as well as their ability to regain stability after turning. Since turning in place involves foot movement, single-leg support, and alternating adjustments between both feet, the pressure ratios of the left and right feet and the front-to-back pressure ratio will change significantly and instantaneously. Therefore, the system does not use the maximum left-to-right pressure difference or the maximum front-to-back pressure difference during the subject's turning task as the core abnormal indicator. Instead, it primarily evaluates dynamic balance ability through the COP trajectory and COP stability.

[0103] In some embodiments, when the system calculates the change in the center of pressure (COP) of the test subject during a turning task based on continuous plantar pressure images using the COP dynamic stability calculation module, the COP dynamic stability calculation module can calculate the following indicators: Maximum COP offset: Using the average COP position during the stabilization phase before the turn as a benchmark, calculate the maximum distance between the COP and this benchmark position during the turn execution. This metric reflects the degree to which the center of gravity deviates from the initial stable position during the turn.

[0104] COP trajectory length: The COP trajectory length is obtained by summing the distances between COP coordinates of adjacent frames during the turn execution phase. This metric reflects the total magnitude of the center of gravity adjustment during the turn.

[0105] COP average speed: Calculated based on the length and duration of the COP trajectory during the turn, it reflects the average speed of the center of gravity movement during the turn.

[0106] Peak COP velocity: Calculated by taking the maximum value of the COP bit removal rate between adjacent frames over time intervals. This metric reflects whether there is a sudden, large shift in the center of gravity during a turn.

[0107] COP swing area: Calculated based on the COP point set during the turning task or the stabilization phase after the turn, such as using the area of ​​the convex hull, the area of ​​the circumscribed ellipse, or the area of ​​the confidence ellipse. This indicator reflects the range of the center of gravity swing.

[0108] COP stabilization time after turning: After the subject reaches the target orientation, COP changes are continuously monitored. The subject is considered to have regained stability when the COP velocity, COP displacement range, or COP oscillation area falls below a threshold within a consecutive preset time window. The time between reaching the target orientation and regaining stability is the COP stabilization time after turning.

[0109] The system can reflect the dynamic balance ability of the elderly during the process of turning around in place and their postural recovery ability after turning through the above indicators.

[0110] In some embodiments, the system can solve for the three-dimensional center of gravity (COM) based on a two-dimensional pressure array, the subject's body mass index (BMI), and COP. The system can then further calculate the offset angle of the center of gravity relative to the Y-axis based on this COM to assess the subject's fall risk.

[0111] For example, the system can calculate the three-dimensional center of gravity COM and the offset angle of the center of gravity relative to the Y-axis through the following process to assess the fall risk of the subject: 1. The system defines the coordinate system used for evaluation, with the force pad plane as the XY plane corresponding to the two-dimensional pressure array on the ground. The X-axis is the horizontal axis of the human body, pointing from the left foot to the right foot; the Y-axis is the front-back axis of the human body, with the toe direction as the positive Y direction and the heel direction as the negative Y direction; the Z-axis is perpendicular to the ground and pointing upwards, which is the three-dimensional vertical axis of the center of gravity in the height direction; the origin O is set as the reference point for the ground projection of the center of gravity of the human body in a standard neutral static standing position.

[0112] 2. The system collects three types of input conditions: first, the pressure data of each sensing unit on the left and right feet obtained from the two-dimensional pressure array. Secondly, the basic BMI parameter, including height. ,weight Thirdly, the left foot was initially calculated using the pressure array. Right foot Total pressure .

[0113] 3. The system calculates the two-dimensional composite centroid projection of the ground. The formula for single-leg COP is: ; .

[0114] The formula for calculating the centroid of the ground projection (XY plane) is: .

[0115] 4. The system uses BMI to calculate the three-dimensional center of gravity. Among them, the baseline static center of gravity height (BMI height derivation) of the adult trunk center of gravity standing height empirical model (adapted for the elderly population) is as follows: , For height (mm), 0.55 is suitable for trunk atrophy in the elderly; in addition, dynamic correction for vertical height (total pressure correction) is also included. The total plantar pressure is balanced by the vertical resultant force. With self-respect The difference reflects the vertical acceleration of the center of gravity and corrects the three-dimensional Z-coordinate. .

[0116] Thus, the final three-dimensional coordinates of the human body's center of gravity are: ,in, The horizontal position is left and right. For the front and rear Y-axis ground projections, The vertical height of the center of gravity above the ground (a three-dimensional key component).

[0117] 5. The system calculates the pitch angle (a key indicator of fall) of the three-dimensional center of gravity relative to the Y-axis. The Y-axis of the geometric model is the front-to-back reference axis of the human body; the three-dimensional center of gravity point... Ground reference projection point (COM projection origin when standing upright without tilt); In addition, the system construction vector points from the ground origin to the three-dimensional centroid. This yields the angle between the centroid vector and the Y-axis. (Pitch offset angle) Y-axis standard unit vector: Furthermore, the formula for the angle between the vectors used by the system (law of cosines) can be: .

[0118] Among them, dot product Vector magnitude .

[0119] The system substitutes the above dot product and vector magnitude into the simplified formula to obtain the core formula for the offset angle: .

[0120] The unit is degrees (°), and the physical meaning of angle includes: The smaller the value (around 0°), the closer the center of gravity is to the Y-axis, the more upright the body is, and the more stable it is. An increase indicates that the center of gravity is deviating from the front-to-back Y-axis, and the body is leaning forward / backward. This indicates that the center of gravity is shifted forward, or the angle of forward lean. This indicates that the center of gravity is shifted backward, and the angle of backward tilt; The larger the trunk, the more pronounced the tilt, and the higher the risk of falling forward or backward.

[0121] 6. The system is based on the offset angle. The fall risk classification for the elderly is based on the following strategies, as shown in the table below:

[0122] 7. The system can further supplement its dynamic risk assessment based on consecutive frames, thereby preventing instantaneous misjudgments. Specifically, a single frame... Exceeding the threshold only causes momentary tilting for more than 5 consecutive frames. It can be determined as a continuous loss of balance, with an angular change rate of [missing information]. The indicator shows a rapid tilting of the center of gravity, providing an immediate high-risk warning.

[0123] Step S304: Calculate the pressure distribution of the subject when performing a turning task based on the plantar pressure image; the multidimensional plantar force characteristics include the pressure distribution.

[0124] When the system extracts the multidimensional plantar force features of the subject when performing a turning task from the plantar pressure image, it can calculate the pressure distribution of the subject when performing the turning task based on the plantar pressure image, and then use the calculated pressure distribution as one of the multidimensional plantar force features of the subject when performing the turning task.

[0125] In some embodiments, the system includes a static load recovery calculation module. This module can be used to calculate the left-right load ratio, front-back pressure ratio, and their changes during the pre-turn and post-turn stabilization phases of the test subject, thereby obtaining the pressure distribution of the test subject when performing a turning task.

[0126] It should be noted that the left-right pressure ratio and front-back pressure ratio are mainly used for static standing posture analysis and recovery state analysis after turning, and not as core dynamic evaluation indicators during the turning task. The reason is that during a turn in place, actions such as foot movement, brief single-leg support, and re-landing can cause significant instantaneous changes in the left-right or front-back pressure ratio. These changes are part of the turning action itself and cannot be simply interpreted as abnormal off-center loading.

[0127] In some embodiments, the system can use a static load recovery calculation module to calculate pressure distribution indices mainly within two time windows: the pre-turn stabilization window and the post-turn stabilization window, thereby obtaining the pressure distribution of the subject when performing the turning task.

[0128] It should be noted that the pre-turn stability window refers to the period of time during which the elderly person stands stably with both feet and has minimal COP fluctuations before receiving the turning instruction. Furthermore, the post-turn stability window refers to the period of time during which the elderly person stands stably with both feet and has minimal COP fluctuations after reaching the target orientation and regaining their footing.

[0129] In some embodiments, the system can use the static load recovery calculation module to calculate the following metrics within the two stability windows mentioned above: Weight ratio for left and right before turning: calculated based on the total pressure on the left foot and the total pressure on the right foot, used to establish the initial standing posture weight baseline.

[0130] Weight-bearing ratio after turning: Calculated based on the total pressure on the left foot and the total pressure on the right foot within the stable window after turning, used to determine whether the elderly person has returned to a relatively stable weight-bearing state on both feet after completing the turn.

[0131] Changes in load before and after turning: Calculated based on the difference between the left and right load ratios before and after turning, used to determine whether there is a significant change in load distribution after turning.

[0132] Front-to-back pressure ratio before turning: Within the stable window before turning, the front and back pressure areas of the human body are divided according to the human coordinate system, and the front-to-back pressure ratio is calculated to describe whether there is a tendency to lean forward or backward in the initial standing posture of the elderly.

[0133] Front-to-back pressure ratio after turning: Within the stable window after turning, the human coordinate system is re-established based on the standing posture orientation after turning, and the front-to-back pressure ratio is calculated to determine whether the elderly person has a tendency to lean forward or backward after completing the turn.

[0134] Change in pressure before and after turning: Calculated based on the change between the ratio of pressure before and after turning and the ratio of pressure before and after turning, used to determine whether the standing posture has changed significantly relative to the initial state after turning.

[0135] In some embodiments, the stress array-based method for assessing the steady-state turning of the elderly provided in this application may further include the following steps: A human coordinate system is established based on the standing posture orientation recognition results.

[0136] The front-to-back pressure ratio is not directly divided into upper and lower areas based on the fixed coordinates of the pressure pad, but rather a human coordinate system is established based on the subject's current standing posture and orientation. The front of the body is considered the front, and the back is considered the back.

[0137] In some embodiments, step S304 above, calculating the pressure distribution of the subject performing a turning task based on the plantar pressure image, may include the following steps: Based on the human body coordinate system, the plantar pressure image is divided into a forefoot region and a hindfoot region; The pressure distribution of the subject when performing a turning task is calculated based on the pressure in the forefoot region and the hindfoot region, respectively.

[0138] The system establishes a human coordinate system based on the standing posture and orientation recognition results. It can then divide the plantar pressure image into a forefoot region and a hindfoot region based on this coordinate system. Subsequently, it calculates the pressure distribution of the subject when performing a turning task based on the pressure of each of the forefoot and hindfoot regions.

[0139] After establishing a human coordinate system based on the subject's current standing posture and orientation, the system can determine the toe and heel along the foot axis for each foot, and divide the body into forefoot and hindfoot regions according to the foot length ratio. The pressure in these forefoot and hindfoot regions can then be used to calculate the anterior-posterior pressure ratio of the human body.

[0140] For example, the 40% of foot length closest to the toes can be designated as the forefoot region, the 40% of foot length closest to the heel as the hindfoot region, and the middle 20% as the midfoot region. The midfoot region can be excluded from the calculation of the forefoot-to-hindfoot pressure ratio to reduce the impact of unstable pressure in the arch area on the results.

[0141] For example, the 50% of the foot length closest to the toe can be designated as the forefoot region, and the 50% of the foot length closest to the heel can be designated as the hindfoot region.

[0142] In some embodiments, the system can calculate the foot axis direction using principal component analysis. Alternatively, the system can also use minimum bounding rectangle, pressure-weighted principal axis analysis, or keypoint detection methods to calculate the foot axis direction.

[0143] Please refer to Figure 4 , Figure 4 The flowchart of the method for assessing the turning steady state of the elderly based on a pressure array, which is provided in the embodiments of this application, is shown in some other embodiments.

[0144] like Figure 4 As shown, in some embodiments, the method for assessing the turning steady state of the elderly based on a pressure array provided in this application may further include steps S401 and S402 as shown below.

[0145] Step S401: Based on the standing posture orientation recognition result and the turning ability assessment result, the test subject performs a turning task to identify the state and obtain the real-time turning state of the test subject.

[0146] After obtaining the assessment results of the subject's turning ability, the system can also perform state recognition of the subject's turning task based on the previously identified standing posture and orientation recognition results and the turning ability assessment results, thereby obtaining the subject's real-time turning state.

[0147] In some embodiments, the system includes a turning state machine determination module. This module can determine the stage of the turning task based on the standing posture recognition result, dynamic changes in COP, foot area movement, and pressure stability, thereby obtaining the subject's real-time turning state.

[0148] It should be noted that the turning state machine can include an initial stabilization phase, a turning execution phase, a target orientation attainment phase, a post-turn stabilization phase, and a termination phase. The initial stabilization phase is used to establish the initial COP baseline, initial left-right weight distribution, and initial front-back pressure distribution. The turning execution phase is mainly used to calculate the turning motion time, the number of foot adjustments, and the COP dynamic stability index. The target orientation attainment phase is used to record the time to reach the target orientation. The post-turn stabilization phase is used to calculate the post-turn COP stabilization time and to calculate the left-right weight distribution and front-back pressure distribution after the turn.

[0149] In some embodiments, the system can determine that the subject is in the initial stable phase when the subject is standing facing the initial direction, the COP fluctuation is small, and the position of the pressure area of ​​both feet is stable.

[0150] In some embodiments, the system can determine that the subject has entered the turning execution phase when it detects an increase in COP speed, a continuous change in COP position, movement of the foot area, or a change in orientation recognition results.

[0151] In some embodiments, the system can determine that the subject has reached the target orientation when the standing posture orientation recognition result is consistent with the target orientation for several consecutive frames, or when the error between the standing posture orientation angle and the target angle is less than a preset threshold.

[0152] In some embodiments, the system can continue to monitor COP and foot area changes after the subject reaches the target orientation. When the COP velocity is lower than a threshold, the COP displacement range is lower than a threshold, and the foot area position is stable within a consecutive preset time window, it is determined that the subject has regained stability, i.e., entered the post-turn stabilization phase.

[0153] In some embodiments, after the system completes the above-mentioned index calculation and outputs the evaluation results, it can confirm that the subject has finished performing the turning task, that is, enter the end stage.

[0154] In some embodiments, the key time points output by the state machine may include: the start time of the turn, the time to reach the target orientation, and the time to regain stability after the turn. The system can calculate the subject's turning motion time, stabilization time after the turn, and total turning time based on these time points. Wherein: Turning time = Time to reach the target orientation - Time to start turning; Stabilization time after turning = Time to regain balance after turning - Time to reach the target orientation; Total turning time = Time to regain balance after turning - Time to start turning.

[0155] Step S402: Based on the standing posture orientation recognition result, the multidimensional plantar force characteristics, and the real-time turning state, evaluate the subject's turning ability to obtain the subject's turning ability evaluation result.

[0156] After obtaining the subject's real-time turning state, the system can evaluate the subject's turning ability based on this real-time turning state, the previously identified standing orientation, and multi-dimensional plantar force characteristics, thus obtaining the subject's turning ability evaluation result. For example, when the turning ability evaluation result is a comprehensive turning ability score, the system can calculate the subject's turning motion time, number of foot adjustments, difference in the number of adjustments between the left and right feet, maximum COP offset, COP trajectory length, average COP velocity, peak COP velocity, and COP swing area based on the standing orientation, multi-dimensional plantar force characteristics, and real-time turning state; and calculate the left-right load ratio and front-back pressure ratio of the subject in the stability window before and after the turn, and calculate the changes in left-right load and front-back pressure before and after the turn; then standardize the calculated multiple indicators to obtain standardized scores for each indicator; finally, the standardized scores are weighted and summed according to preset weights to obtain the subject's comprehensive turning ability score.

[0157] It should be noted that the COP swing area is not limited to the convex hull area; it can also be the circumscribed ellipse area, the circumscribed rectangle area, or the confidence ellipse area.

[0158] Please refer to Figure 5 , Figure 5 for Figure 1 A detailed flowchart of step S103.

[0159] like Figure 5 As shown, in some embodiments, step S103 above: evaluating the subject's turning ability based on the standing posture orientation recognition result and the multidimensional plantar force characteristics to obtain the subject's turning ability evaluation result may include steps S501 to S503 as shown below.

[0160] Step S501: Perform index standardization processing on the standing posture orientation recognition result and the multidimensional foot force characteristics to obtain the standardized scores of the multidimensional turning ability index of the test subject.

[0161] In some embodiments, multidimensional turning ability indicators may include the turning motion time, stabilization time after turning, number of foot adjustments (number of small steps), difference in the number of adjustments between the left and right feet, maximum COP offset, COP trajectory length, COP swing area, changes in load on the left and right sides before and after turning, and changes in pressure before and after turning, etc.

[0162] When the system assesses the subject's turning ability based on the standing orientation recognition results and multidimensional plantar force characteristics (or when assessing the subject's turning ability based on the real-time turning state, standing orientation recognition results, and multidimensional plantar force characteristics), it first performs index standardization processing on the standing orientation recognition results and multidimensional plantar force characteristics (or performs index standardization processing on the real-time turning state, standing orientation recognition results, and multidimensional plantar force characteristics) to obtain the standardized scores of the subject's multidimensional turning ability indicators.

[0163] In some embodiments, step S501 above, which involves standardizing the standing posture recognition result and the multidimensional plantar force characteristics to obtain standardized scores for the subject's multidimensional turning ability indicators, may include the following steps: Determine the multidimensional turning ability index corresponding to the standing posture orientation recognition result and the multidimensional plantar force characteristics; The multidimensional turning ability indicators are subjected to extreme value standardization processing based on a preset reference indicator distribution to obtain the standardized scores of each multidimensional turning ability indicator; the reference indicator distribution includes the multidimensional turning ability indicators of monitoring personnel of the same age group as the test subject.

[0164] It should be noted that the reference indicator distribution can be the maximum, minimum, and target values ​​of each of the multidimensional turning ability indicators for the reference population (monitors of the same age group as the test subjects). The maximum, minimum, and target values ​​of each indicator for this reference population can be determined through one of the following methods: A large-sample test was conducted on healthy individuals of the same age group as the test subjects to establish an indicator reference database. Individualized baselines are established based on the subjects' own historical data for long-term longitudinal tracking and evaluation of rehabilitation effects; The reference range of indicators for the normal population was obtained through literature review; The scale is calibrated against authoritative clinical scales (such as the Berg Balance Scale), and the range of indicators corresponding to the full score of the scale is used as a reference benchmark.

[0165] When the system standardizes the standing posture orientation recognition results and multidimensional plantar force characteristics, it first determines the multidimensional turning ability indicators corresponding to the standing posture orientation recognition results and multidimensional plantar force characteristics. These indicators include the turning motion time, turning stability time, number of foot adjustments (number of small steps), difference in the number of left and right foot adjustments, maximum COP offset, COP trajectory length, COP swing area, left and right weight changes before and after turning, and pressure changes before and after turning.

[0166] Then, the system can use the reference index distribution of healthy people of the same age group as a benchmark to perform extreme value standardization processing on the above multidimensional turning ability index, thereby obtaining the standardized score of each multidimensional turning ability index.

[0167] In some embodiments, the system can also use the reference index distribution as a benchmark to perform zero-mean Z-score standardization (also known as standardized normalization) on the multidimensional turning capability index, thereby obtaining the standardized score of each multidimensional turning capability index.

[0168] In some embodiments, the system can also use the reference index distribution as a benchmark to perform percentile standardization on the multidimensional turning capability index, thereby obtaining the standardized score of each multidimensional turning capability index.

[0169] In some embodiments, the system can also use the reference index distribution as a benchmark to perform maximum-minimum value normalization on the multidimensional turning capability index, thereby obtaining the standardized score of each multidimensional turning capability index.

[0170] Step S502: The standardized scores are weighted and summed according to preset weights to obtain the test subject's comprehensive turning ability score.

[0171] It should be noted that the preset weights may include: turning motion time weight (0.15-0.25), post-turn stabilization time weight (0.10-0.20), foot adjustment number weight (0.10-0.20), difference in left and right foot adjustment number weight (0.05-0.15), maximum COP offset weight (0.10-0.20), COP trajectory length weight (0.05-0.15), COP swing area weight (0.05-0.15), left and right load variation weight (0.03-0.10), and front and back pressure variation weight (0.03-0.10). Furthermore, the preset weights can also be automatically learned from training samples using machine learning models (such as random forests or support vector regression), or determined using expert consultation methods (such as the Delphi method).

[0172] After the system obtains multiple standardized scores, it can further use the aforementioned weights to perform a weighted summation of these multiple standardized scores, thereby obtaining the test subject's comprehensive score for turning ability.

[0173] In some embodiments, the system may also employ nonlinear fusion methods (such as multiplicative models, fuzzy comprehensive evaluation, and analytic hierarchy process AHP) to fuse multiple standardized scores to obtain a comprehensive score of the test subject's turning ability.

[0174] In some embodiments, the system can standardize and weight the following nine indicators (turning efficiency index, turning motion time, and stabilization time after turning), foot coordination index (number of foot adjustments and differences in left and right foot adjustments), COP dynamic stability index (maximum COP offset, COP trajectory length, and COP swing area), and static load recovery index (left and right load changes and front and back pressure changes), and then output a unified comprehensive score for turning ability.

[0175] Step S503: Based on the comprehensive score of the turning ability, perform turning ability level mapping processing on the subject to obtain the turning ability level of the subject, and output a fall risk warning based on the turning ability level; the turning ability assessment result includes the turning ability level and the fall risk warning, and the turning ability level includes at least one of excellent, good, average, poor and very poor.

[0176] After obtaining the subject's overall turning ability score, the system further maps this score to a turning ability level, resulting in a rating of excellent, good, average, poor, or very poor. The system then outputs a fall risk warning based on this turning ability level. In this way, the system combines the final determined turning ability level and the fall risk warning as the subject's overall turning ability assessment result.

[0177] In some embodiments, the system includes a comprehensive scoring and evaluation level output module. The system can use this module to integrate the multi-dimensional turnaround capability indicators calculated by the aforementioned modules into an operable comprehensive evaluation result (i.e., a turnaround capability evaluation result). This module includes an indicator standardization unit, a weighted fusion unit, and a capability level mapping unit. Wherein: The indicator standardization unit is used to map the measured values ​​of each indicator to a standardized score with a unified dimension.

[0178] Because the indicators have different dimensions, direct weighted summation lacks physical meaning. Therefore, extreme value standardization is the preferred method for indicator standardization. For indicators where "smaller values ​​indicate better ability" (such as turning time, number of foot adjustments, maximum COP offset, etc.), the standardized score calculation formula can be: .

[0179] Where Xi is the measured value of the subject, and Ximax and Ximin are the maximum and minimum values ​​of the indicator in the reference population (healthy elderly people of the same age group), respectively.

[0180] Furthermore, for indicators where "the closer the value is to the target value, the better the ability" (such as the difference in the number of adjustments between the left and right feet, the change in load before and after a turn, etc.), the standardized score calculation formula can be: .

[0181] Xitarget is the target value of this indicator (usually 0, indicating complete symmetry or no change before and after).

[0182] The weighted fusion unit is used to sum the standardized scores according to preset weights to obtain the overall turning ability score: .

[0183] Where wi is the weight of the i-th indicator, satisfying Furthermore, the preset weights of each indicator can be adjusted according to the clinical application scenario. For example, the weight allocation scheme can be shown in the table below:

[0184] Furthermore, the aforementioned weight allocation can be determined based on the following principles: The turning time and the number of foot adjustments directly reflect the smoothness and efficiency of the turning action, and are therefore given high weight. The maximum offset of COP and the swing area of ​​COP directly reflect the ability to control the center of gravity during the turn, and are given high weight. Static load recovery index reflects the stable recovery capability after turning and helps to determine whether there is a continuous off-center load.

[0185] It should be noted that the above weights are the preferred options. In practical applications, the weights of each indicator can be adjusted according to the purpose of the assessment (rapid screening in the community, detailed clinical assessment, comparison before and after rehabilitation), or personalized weight configuration can be made according to the characteristics of different elderly groups (such as different age groups and different types of diseases).

[0186] The ability level mapping unit is used to output the turning ability level based on the comprehensive score. For example, the level division scheme can be shown in the table below:

[0187] Furthermore, the aforementioned grading thresholds can be adjusted according to the application scenario. For example, in community screening scenarios, the thresholds can be appropriately increased to enhance sensitivity, while in rehabilitation training scenarios, a more granular grading system can be used to more sensitively reflect the training effect.

[0188] For example, suppose an elderly person completes a 90° right turn test, and the data and standardized score obtained by the system are shown in the table below:

[0189] Then, the comprehensive score is calculated according to the preferred weight: Score = 0.20×65 + 0.15×50 + 0.15×50 + 0.10×50 + 0.15×54 + 0.10×50 + 0.10×56 + 0.05×60 + 0.05×60 = 54.1.

[0190] If the overall score corresponds to a "moderate" level of turning ability, it indicates that the elderly person has a certain risk of falling, and further evaluation or balance training intervention is recommended.

[0191] In some embodiments, in addition to using the above-mentioned rule-based scoring method to calculate and determine the test subject's overall turning ability score, the system may also use a machine learning model or an expert weight model to calculate the test subject's turning ability score.

[0192] In some embodiments, in addition to using the above five-level classification method to classify the ability level of the test subjects, the system can also use a three-level classification method (such as normal, borderline, abnormal) or a seven-level classification method, and the relevant thresholds can also be adjusted according to the actual application scenario.

[0193] Next, a complete embodiment of the stress array-based method for assessing the steady-state shift of the elderly, as provided in this application, will be presented.

[0194] It should be noted that this embodiment does not limit the specific sensor model, pressure array size, sampling frequency, orientation recognition model structure, or threshold setting. Furthermore, the relevant parameters can be configured according to the actual equipment and application scenario.

[0195] In addition, in this embodiment, the system needs to identify or calculate at least the following data for each frame of plantar pressure image: Timestamp; Original pressure matrix; Total pressure; The current standing posture is facing which side: A, B, C, or D. Orientation identification confidence; Left foot area and right foot area; The position, area, and total pressure of the left and right feet; Foot movement status is used to determine whether foot adjustments have occurred; COP coordinates; COP speed; The current transition phase.

[0196] The left-right load ratio and front-back pressure ratio can be calculated and saved in each frame, but they are mainly used for statistical analysis of the stabilization window before and after the turn. During the subject's turn task, the system mainly uses COP trajectory, COP speed, COP offset, and number of foot adjustments to evaluate dynamic turn ability.

[0197] Furthermore, in this embodiment, the turning ability evaluation index calculated by the system based on the above data can be divided into four categories: orientation completion index, turning efficiency index, COP dynamic stability index, and static load recovery index.

[0198] The orientation completion indicators include: initial orientation, target orientation, final orientation, orientation completion error, and orientation recognition confidence. These indicators are used to determine whether the elderly person has completed the turning task in the specified direction as required.

[0199] Turning efficiency indicators include: turn initiation time, turn movement time, post-turn stabilization time, total turn time, number of foot adjustments, and the difference in the number of adjustments between the left and right feet. These indicators reflect the speed, smoothness, and foot adjustment of the elderly in completing a turning movement.

[0200] COP dynamic stability indicators include: maximum COP offset, COP trajectory length, average COP velocity, peak COP velocity, COP swing area, and COP stabilization time after turning. These indicators are core dynamic stability metrics during the turning process, reflecting the elderly person's ability to control their center of gravity and recover their posture after turning.

[0201] Static weight-bearing recovery indicators include: the left-right weight-bearing ratio before turning, the left-right weight-bearing ratio after turning, the change in left-right weight-bearing before and after turning, the front-back pressure ratio before and after turning, the front-back pressure ratio after turning, and the change in front-back pressure before and after turning. These indicators are mainly used to compare the weight-bearing status of elderly people in a stable standing posture before and after turning, and to help determine whether a relatively stable standing posture has been restored after turning. These indicators do not use the instantaneous maximum left-right pressure difference or the instantaneous maximum front-back pressure difference during the turning process as the core judgment criterion.

[0202] In this embodiment, the system can sequentially execute steps 1 to 18 as shown below to achieve a steady-state assessment of the elderly person's turning and obtain a final assessment report.

[0203] 1. Establish a pressure array coordinate system and define four boundary directions: A, B, C, and D; 2. Instruct the elderly person to stand facing the designated initial direction; 3. Collect pressure data within the stable window before turning around; 4. Identify the initial stance orientation, left and right foot areas, initial COP, initial left-right weight distribution, and initial front-back pressure distribution; 5. Issue a turning command, such as turning 90° to the right or turning 180° backward; 6. Continuously acquire pressure frames during the turning process; 7. For each frame, identify the current orientation, foot region, and foot movement state, and calculate the COP coordinates and COP velocity; 8. Use a state machine to determine when to start turning, when to reach the target orientation, and when to regain balance after turning; 9. Calculate the number of foot adjustments and COP dynamic stability index during the turn execution phase; 10. Calculate the left-right load ratio and the front-back pressure ratio after turning within the stable window after turning; 11. Calculate the static load recovery index based on the pressure ratio of the stabilization window before and after the turn; 12. Output orientation completion index, turning efficiency index, COP dynamic stability index, and static load recovery index.

[0204] 13. Compare each indicator with preset thresholds, reference ranges, individual historical baselines or comprehensive scoring models to obtain evaluation results for turning efficiency, foot coordination, dynamic balance, post-turn stability recovery, and static weight-bearing recovery. 14. Based on the above evaluation results, output the elderly person's turning ability level or fall risk assistance prompts, and generate an assessment report.

[0205] 15. Input the indicators calculated in steps (4) to (10) into the comprehensive scoring module for standardization processing, and map the measured values ​​of each indicator to the standardized score with a unified dimension. 16. The standardized scores are weighted and summed according to preset weights to obtain the comprehensive score of turning ability; 17. Compare the overall score with a preset level threshold, and output the turning ability level and / or fall risk assistance prompt; 18. Generate an evaluation report based on the overall score and each sub-indicator.

[0206] Please refer to Figure 6 , Figure 6 This is a flowchart illustrating a complete embodiment of the pressure array-based method for assessing the steady-state turning of the elderly, as provided in this application.

[0207] like Figure 6 As shown, in some embodiments, when the system applies the pressure array-based steady-state assessment method for elderly turning provided in the embodiments of this application, the system first establishes a pressure array coordinate system, defines four boundary directions A, B, C, and D, prompts the elderly to stand facing the specified initial direction, collects pressure data within the stable window before turning, identifies the initial orientation, left and right foot regions, initial COP and pressure ratio. Within this window, the COP fluctuation is small and the two foot regions are stable, which is used to establish an assessment baseline.

[0208] The system then issues a turning command, guiding the elderly person to complete a 90° right turn or a 180° turn. It continuously captures pressure frames during the turning process, identifying the current orientation, foot area, and foot movement state frame by frame. It continuously calculates COP coordinates and COP velocity and determines the current turning stage. The system uses a state machine to divide the process into four stages: initial stabilization, turning execution, target orientation, and stabilization recovery. During the turning execution stage, it calculates the number of foot adjustments and the dynamic stability index of COP.

[0209] Once the system enters the stabilization window for turning, it calculates the left-right load ratio and the front-back pressure ratio to obtain static load recovery indicators. Finally, it compares these indicators with thresholds, reference ranges, historical baselines, and a comprehensive scoring model, outputting a turning ability level, fall risk assistance prompts, and an assessment report. The output results cover three core dimensions: stability, coordination, and load recovery.

[0210] Please refer to Figure 7 , Figure 7 This is a schematic diagram of the system flow involved in one embodiment of the pressure array-based method for assessing the steady-state turning of the elderly, which is provided in this application.

[0211] like Figure 7 As shown, in some embodiments, when the system performs a turnaround steady-state assessment on elderly individuals, the pressure array acquisition module continuously acquires pressure frames and pressure matrices, transmits the raw data to the pressure image preprocessing module, and sequentially performs denoising, smoothing, threshold segmentation, and connected component processing.

[0212] The preprocessed data is processed in parallel via two paths: one path is fed into the standing orientation recognition module, which uses CNN classification combined with foot axis angle rules to identify the standing orientation; simultaneously, it is fed into the left and right foot region recognition and tracking module to identify the pressure areas of the left and right feet and maintain continuity of identity; the output result is then fed into the turning state machine judgment module to determine four stages: initial stability, turning execution, reaching the target, and regaining stability. The other path of data is fed into the foot state judgment module, COP calculation module, and pressure distribution calculation module respectively: the foot state judgment module identifies stable contact, moving, decompression and foot lifting, and re-landing states; the COP calculation module outputs COP coordinates, COP velocity, and COP trajectory; the pressure distribution calculation module calculates the left and right weight-bearing ratio, the front and back pressure ratio, and the weight-bearing recovery index.

[0213] The index calculation module integrates multiple outputs to obtain the turning efficiency index, COP dynamic stability index, and static load recovery index, which are then fed into the capability evaluation module. The module combines the threshold reference range, historical baseline, and comprehensive scoring model to complete the steady-state assessment. Finally, the evaluation result output module outputs the turning capability index and auxiliary prompts.

[0214] like Figure 8 As shown, in some embodiments, the system can employ a dual-coordinate system architecture (fixed coordinate system for the pressure array and dynamic human body coordinate system) to calculate relevant pressure indicators. The pressure array coordinate system is a fixed XY two-dimensional coordinate system, with the array plane boundary divided into sides A, B, C, and D. All original pressure frames, pressure point positions, and foot pressure areas are stored using coordinate values ​​from this fixed coordinate system, serving as the reference system for the entire data acquisition process. Furthermore, the human body coordinate system is a relative coordinate system that dynamically changes with the subject's standing posture and orientation, including three directional dimensions: front, left, and right, established based on the orientation of the foot axes. When the elderly person stands facing side B, the front of the body points towards side B of the pressure array, and the left and right sides correspond to the corresponding sides of the array. The pressure areas for the left and right feet are then divided accordingly in the human body coordinate system. When calculating pressure-related indicators, the system divides the area according to the relative orientation of the front and back of the body, rather than using a fixed array coordinate direction. This eliminates indicator deviations caused by different standing orientations and avoids incomparability of evaluation indicators under different orientations. The system establishes a matching human coordinate system in real time based on the current standing posture and orientation. Under the human coordinate system, it calculates the front-to-back pressure ratio and the left-to-right load ratio, providing a unified coordinate benchmark for steady-state assessment of the entire turning process and ensuring that the assessment results of different orientation stages are consistent and comparable.

[0215] like Figure 9As shown, in some embodiments, the system can divide the entire evaluation process using a stationary turn task state machine, which sequentially includes five states: initial stabilization phase, turn execution phase, target orientation attainment phase, post-turn stabilization phase, and end phase. State transitions are completed at three time points: turn start (Ts), target attainment (Ta), and re-establishment (Tr). The initial stabilization phase corresponds to before Ts. The subject stands facing the initial direction. The system detects small COP fluctuations and stable foot areas, determining that it is in the initial stable state and establishing an evaluation baseline until changes in COP and feet are detected, triggering the turn start node. The turn execution phase corresponds to the interval from Ts to Ta. During this phase, COP speed increases and the foot area moves, causing the subject's orientation to change. The system continuously tracks the COP trajectory and foot movement until the subject reaches the target orientation. The duration of this phase is the turn motion time, calculated as Ta minus Ts. The target orientation attainment phase corresponds to time Ta. When the orientation is consistent for multiple consecutive frames, the angle deviation is less than a threshold, and the stabilization window condition is met, the subject is determined to have reached the target orientation. The post-turn stabilization phase corresponds to the interval from Ta to Tr. During this phase, COP speed decreases, foot position stabilizes, and pressure distribution gradually recovers. The system simultaneously calculates the load recovery index. The duration of this phase is the post-turn stabilization time, calculated as Tr minus Ta. The total turn time is Tr minus Ts, covering the entire process from the start of the movement to regaining footing. After the end phase (corresponding to Tr), the system completes all index calculations, outputs an evaluation report, and the state machine process terminates.

[0216] like Figure 10As shown, in some embodiments, the system can quantify the elderly person's center of gravity control ability during turning using COP trajectory indicators. All indicators are calculated and generated based on pressure center data collected in real time by the pressure array. The system uses the pressure center position during the stabilization phase before turning as the pre-turn COP reference. During the turn execution phase, a continuous COP motion trajectory is formed. The dotted area outside the trajectory is the COP swing area, covering the spatial range of the entire center of gravity fluctuation. The point on the trajectory farthest from the initial reference corresponds to the maximum COP offset Dmax, and the trajectory endpoint area corresponds to the stabilization window after turning, used to determine the moment when the subject regains stability. The system extracts six core COP trajectory indicators: Dmax is the maximum offset relative to the initial reference, representing the maximum deviation of the center of gravity; LCOP is the cumulative distance between adjacent COP points, i.e., the total length of the COP trajectory, reflecting the total path of the center of gravity movement; Vavg is the ratio of the COP trajectory length to the turning time, representing the average velocity of the center of gravity movement; Vpeak is the maximum COP velocity in adjacent frames, reflecting the peak rate of the center of gravity movement; A_COP is the convex hull or elliptical area of ​​the COP point set, corresponding to the COP swing area, quantifying the spatial coverage of the center of gravity fluctuation; and Tstable is the time to regain stability after reaching the target orientation, reflecting the steady-state recovery efficiency after turning. These indicators characterize the steady-state features of the center of gravity during the turning process from multiple dimensions, including displacement, velocity, spatial range, and recovery time, providing quantitative data support for assessing the turning ability of the elderly.

[0217] like Figure 11 As shown, in some embodiments, the system can adopt a foot region division scheme along the foot axis to provide a zoning benchmark for weight-bearing recovery analysis in turn steady-state assessment. The system uses the foot axis of the single-foot pressure region as the core direction, with the foot axis pointing to the corresponding toe. Along the foot axis, the single-foot pressure region is divided into three segments: the forefoot region, the midfoot region, and the hindfoot region. The boundaries of each segment are perpendicular to the foot axis and correspond to the physiological positions of the toe, middle, and heel of the foot, respectively. The zoning shape conforms to the elliptical contour of the single-foot pressure region.

[0218] The pressure zones for both feet are based on the direction in front of the body as a unified reference. The axis of the left and right feet should correspond to the direction in front of the body to ensure that the front and back zones of the left and right feet are consistent and to avoid deviations in zone direction caused by changes in the orientation of the standing posture.

[0219] Based on this regional division scheme, the system statistically analyzes the total pressure in the forefoot and hindfoot regions, calculating the forefoot-hindfoot pressure ratio. This index is mainly used for weight-bearing recovery analysis of the stability window before and after turning. By comparing the changes in the forefoot-hindfoot pressure distribution ratio before and after the turning action, the system quantifies the degree of plantar weight-bearing repositioning after the elderly completes a turn in place, thereby reflecting the steady-state recovery ability after turning and providing a quantitative basis for weight-bearing dimension assessment of turning ability.

[0220] Compared to traditional methods for assessing balance or fall risk in older adults based on plantar pressure, this embodiment not only collects plantar pressure data but also identifies the specific direction the elderly person faces the pressure array and uses the standing orientation identification for the assessment of the turning task. Traditional methods often focus on gait, sitting / standing, or static balance, while this embodiment establishes a complete analysis process for "standing orientation - turning process - post-turn stability".

[0221] Furthermore, this embodiment transforms the calculation of the front-to-back pressure difference from the fixed coordinate system of the pressure pad to the human body coordinate system. This way, even if the elderly person faces different sides (A, B, C, D), the system can still calculate the pressure offset according to the body's own front-to-back direction, avoiding the problem of incomparable indicators under different orientations. This embodiment also extracts the time points of the turn start, reaching the target orientation, and regaining stability through a state machine, which can further obtain the turn time and COP stabilization time, making the assessment of turnability more objective and detailed.

[0222] In addition, the assessment results of this embodiment are more intuitive and operable than those of traditional solutions: traditional technical solutions only output scattered indicator values, making it difficult for assessors to directly judge the overall turning ability level of the elderly. This embodiment, through standardized processing and weighted fusion, outputs a single comprehensive score and a clear ability level, which can be directly used for clinical decision-making and rehabilitation program development.

[0223] This embodiment solves the problem of cross-individual comparability: by standardizing with a healthy elderly population of the same age group as a reference benchmark, the influence of individual differences such as age, gender, and height on the assessment results is eliminated, making the assessment results of different elderly people horizontally comparable.

[0224] This embodiment has a multi-dimensional comprehensive reflection capability: This embodiment incorporates nine indicators from four dimensions—turning efficiency, foot coordination, COP dynamic stability, and static weight-bearing recovery—into a comprehensive score, which is more comprehensive and reliable than a single indicator evaluation.

[0225] This embodiment has strong scene adaptability: the weights and thresholds can be flexibly configured according to different scenarios such as community screening, clinical assessment, and rehabilitation training to adapt to different application needs.

[0226] The system in this embodiment can be used as a fall risk screening tool: by combining the score with the level of turning ability, the assessment results can be directly mapped to fall risk warnings, providing a convenient screening method for elderly care institutions and communities.

[0227] Please refer to Figure 12 This application also provides a pressure array-based system for assessing the steady-state turning of elderly people. The pressure array-based system for assessing the steady-state turning of elderly people provided in this application can realize the above-mentioned pressure array-based method for assessing the steady-state turning of elderly people.

[0228] like Figure 12 As shown in the embodiments of this application, the elderly turning steady-state assessment system based on a pressure array may include: The pressure array acquisition module is used to acquire the plantar pressure image of the subject in response to the subject standing on the pressure array and performing a turning task; the pressure array is a platform-type pressure array with directional markings. The standing posture orientation recognition module is used to identify the standing posture orientation of the subject relative to the pressure array based on the plantar pressure image, and obtain the standing posture orientation recognition result of the subject. The plantar force feature extraction module is used to extract multidimensional plantar force features of the subject when performing a turning task based on the plantar pressure image. The evaluation result output module is used to evaluate the subject's turning ability based on the standing posture orientation recognition result and the multidimensional plantar force characteristics, and obtain the subject's turning ability evaluation result; and to generate the subject's steering steady-state evaluation result based on the standing posture orientation recognition result and the turning ability evaluation result.

[0229] In some embodiments, the plantar force feature extraction module includes: The left and right foot region recognition and tracking module is used to track and identify the left and right foot pressure regions of the subject when performing a turning task based on the plantar pressure image; the multidimensional plantar force characteristics include the left and right foot pressure regions. The foot state determination module is used to track and identify the foot state of the subject when performing a turning task based on the plantar pressure image; the multidimensional plantar force characteristics include the foot state, which includes at least one of the following: stable contact state, moving state, foot lifting state, re-landing state, transient absence state, and adhesion state of both feet. The COP dynamic stability calculation module is used to dynamically calculate the pressure center of the subject when performing a turning task based on the plantar pressure image, and to calculate the three-dimensional center of gravity of the subject based on the pressure center; the multi-dimensional plantar force characteristics include the pressure center and the three-dimensional center of gravity. The static weight-bearing recovery calculation module is used to calculate the pressure distribution of the subject when performing a turning task based on the plantar pressure image; the multidimensional plantar force characteristics include the pressure distribution.

[0230] In some embodiments, the COP dynamic stability calculation module is further configured to divide the plantar pressure image into a forefoot region and a hindfoot region based on the human body coordinate system; calculate the pressure distribution of the subject when performing a turning task based on the pressure of the forefoot region and the hindfoot region respectively; and establish the human body coordinate system based on the standing posture orientation recognition result.

[0231] In some embodiments, the evaluation result output module is further configured to perform index standardization processing on the standing posture orientation recognition result and the multidimensional plantar force characteristics to obtain the standardized scores of each of the multidimensional turning ability indicators of the test subject; perform weighted summation processing on multiple standardized scores according to preset weights to obtain the comprehensive turning ability score of the test subject; perform turning ability level mapping processing on the test subject based on the comprehensive turning ability score to obtain the turning ability level of the test subject, and output a fall risk warning based on the turning ability level; the turning ability evaluation result includes the turning ability level and the fall risk warning, and the turning ability level includes at least one of excellent, good, average, poor, and very poor.

[0232] In some embodiments, the evaluation result output module is further configured to determine the multidimensional turning ability index corresponding to the standing posture orientation recognition result and the multidimensional plantar force characteristics; perform extreme value standardization processing on the multidimensional turning ability index based on a preset reference index distribution to obtain the standardized score of each of the multidimensional turning ability indexes; the reference index distribution includes the multidimensional turning ability index of the monitoring personnel of the same age group as the test subject.

[0233] In some embodiments, the stress array-based elderly turning steady-state assessment system provided in this application may further include: The turning state machine judgment module is used to identify the state of the subject performing the turning task based on the standing posture orientation recognition result and the turning ability evaluation result, so as to obtain the real-time turning state of the subject; and to evaluate the turning ability of the subject based on the standing posture orientation recognition result, the multi-dimensional plantar force characteristics and the real-time turning state, so as to obtain the turning ability evaluation result of the subject.

[0234] In some embodiments, the foot state judgment module is further configured to input the plantar pressure image into a preset standing orientation recognition model for standing orientation recognition processing, and obtain the standing orientation recognition result of the subject output by the standing orientation recognition model; perform gait analysis on the subject performing a turning task based on the plantar pressure image, and obtain the gait analysis result of the subject; and perform rule recognition processing on the standing orientation of the subject based on the gait analysis result, and obtain the standing orientation recognition result of the subject; the gait analysis result includes foot axis angle, toe and heel direction and / or the relative position of the two feet.

[0235] It should be noted that the specific implementation of the pressure array-based elderly turning steady-state assessment system provided in this application is basically the same as the specific implementation of the pressure array-based elderly turning steady-state assessment method described above, and will not be repeated here.

[0236] Please see Figure 13 This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method for assessing the turning steady state of the elderly based on a pressure array.

[0237] In some embodiments, the electronic device may be an elderly ability assessment device, a health assessment device for elderly care institutions, a fall risk screening device for elderly people in the community, a rehabilitation training assessment device, an intelligent plantar pressure detection platform, an intelligent floor mat, or a device that integrates / operates a pressure pad assessment system / intelligent gait assessment system / balance assessment system.

[0238] like Figure 13 As shown, the electronic device provided in this application embodiment may include: The processor 1301 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1302 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1302 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1302 and is called and executed by the processor 1301 to implement the stress array-based steady-state assessment method for elderly people according to the embodiments of this application. The input / output interface 1303 is used to implement information input and output; The communication interface 1304 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1305 transmits information between various components of the device (e.g., processor 1301, memory 1302, input / output interface 1303, and communication interface 1304); The processor 1301, memory 1302, input / output interface 1303 and communication interface 1304 are connected to each other within the device via bus 1305.

[0239] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for assessing the turning steady state of the elderly based on a pressure array.

[0240] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0241] This application also provides a computer program product, including a computer program. The steps implemented by the computer program when executed by a processor are basically the same as those in the specific embodiments of the stress array-based method for assessing the steady-state shift of the elderly, and will not be repeated here.

[0242] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0243] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0244] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0245] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0246] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0247] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding factor is divided by the following factor, or that the related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0248] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0249] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0250] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0251] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, 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. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0252] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for assessing the steady-state shift of elderly individuals based on a pressure array, characterized in that, The method includes: In response to the subject standing on the pressure array and performing a turning task, the pressure array acquires the plantar pressure image of the subject; the pressure array is a platform-type pressure array with directional markings. Based on the plantar pressure images, the standing orientation of the subject relative to the pressure array is identified to obtain the standing orientation identification result of the subject; and, based on the plantar pressure images, multidimensional plantar force features of the subject when performing a turning task are extracted. The subject's turning ability is evaluated based on the standing posture orientation recognition results and the multidimensional plantar force characteristics, resulting in the subject's turning ability evaluation results. Based on the standing posture orientation recognition results and the turning ability assessment results, the subject's turning steady-state assessment results are generated.

2. The method for assessing the steady-state turning of the elderly based on a pressure array according to claim 1, characterized in that, The extraction of multidimensional plantar force features of the subject performing a turning task based on the plantar pressure image includes at least two of the following: Based on the plantar pressure images, the pressure areas of the left and right feet of the test subject are tracked and identified when performing a turning task; The multidimensional plantar force characteristics include the pressure areas of the left and right feet; The foot pressure image is used to track and identify the foot state of the subject when performing a turning task; The multidimensional plantar force characteristics include the foot state, which includes at least one of the following: stable contact state, moving state, foot lifting state, re-landing state, transient absence state, and adhesion state of both feet. The pressure center of the subject is dynamically calculated based on the plantar pressure image, and the three-dimensional center of gravity of the subject is calculated based on the pressure center. The multidimensional plantar force characteristics include the pressure center and the three-dimensional center of gravity; The pressure distribution of the subject during the turning task was calculated based on the plantar pressure image. The multidimensional plantar force characteristics include the pressure distribution.

3. The method for assessing the steady-state turning of the elderly based on a pressure array according to claim 2, characterized in that, The method further includes: A human coordinate system is established based on the standing posture orientation recognition results; The calculation of the pressure distribution during the subject's turning task based on the plantar pressure image includes: Based on the human body coordinate system, the plantar pressure image is divided into a forefoot region and a hindfoot region; The pressure distribution of the subject when performing a turning task is calculated based on the pressure in the forefoot region and the hindfoot region, respectively.

4. The method for assessing the steady-state turning of the elderly based on a pressure array according to claim 1, characterized in that, The assessment of the subject's turning ability based on the standing posture orientation recognition result and the multidimensional plantar force characteristics yields the subject's turning ability assessment result, including: The standing posture orientation recognition results and the multidimensional plantar force characteristics are subjected to index standardization processing to obtain the standardized scores of the multidimensional turning ability index of the test subject. The standardized scores are weighted and summed according to preset weights to obtain the test subject’s comprehensive turning ability score. Based on the comprehensive score of the turning ability, the test subject is mapped to a turning ability level to obtain the turning ability level of the test subject, and a fall risk warning is output based on the turning ability level; the turning ability assessment result includes the turning ability level and the fall risk warning, and the turning ability level includes at least one of excellent, good, average, poor and very poor.

5. The method for assessing the steady-state turning of the elderly based on a pressure array according to claim 4, characterized in that, The standing posture orientation recognition result and the multidimensional plantar force characteristics are standardized to obtain the standardized scores of the subject's multidimensional turning ability indicators, including: Determine the multidimensional turning ability index corresponding to the standing posture orientation recognition result and the multidimensional plantar force characteristics; The multidimensional turning ability indicators are subjected to extreme value standardization processing based on a preset reference indicator distribution to obtain the standardized scores of each multidimensional turning ability indicator; the reference indicator distribution includes the multidimensional turning ability indicators of monitoring personnel of the same age group as the test subject.

6. The method for assessing the steady-state turning of the elderly based on a pressure array according to claim 1, characterized in that, The method further includes: Based on the standing posture orientation recognition result and the turning ability assessment result, the state of the subject performing the turning task is identified to obtain the real-time turning state of the subject. The subject's turning ability is evaluated based on the standing posture orientation recognition result, the multidimensional plantar force characteristics, and the real-time turning state, resulting in the subject's turning ability evaluation result.

7. The method for assessing the steady-state shift of elderly individuals based on a pressure array according to any one of claims 1 to 6, characterized in that, The step of identifying the subject's standing orientation relative to the pressure array based on the plantar pressure image to obtain the subject's standing orientation identification result includes at least one of the following: The plantar pressure image is input into a preset standing posture orientation recognition model for standing posture orientation recognition processing, and the standing posture orientation recognition result of the subject output by the standing posture orientation recognition model is obtained. Gait analysis is performed on the subject during a turning task based on the plantar pressure image to obtain the subject's gait analysis results. Based on the gait analysis results, rule recognition processing is performed on the subject's standing orientation to obtain the subject's standing orientation recognition results. The gait analysis results include foot axis angle, toe and heel direction and / or relative position of the two feet.

8. A stress array-based system for assessing the steady-state turning behavior of the elderly, characterized in that, The system includes: The pressure array acquisition module is used to acquire the plantar pressure image of the subject in response to the subject standing on the pressure array and performing a turning task; the pressure array is a platform-type pressure array with directional markings. The standing posture orientation recognition module is used to identify the standing posture orientation of the subject relative to the pressure array based on the plantar pressure image, and obtain the standing posture orientation recognition result of the subject. The plantar force feature extraction module is used to extract multidimensional plantar force features of the subject when performing a turning task based on the plantar pressure image. The evaluation result output module is used to evaluate the subject's turning ability based on the standing posture orientation recognition result and the multidimensional plantar force characteristics, and obtain the subject's turning ability evaluation result; and to generate the subject's steering steady-state evaluation result based on the standing posture orientation recognition result and the turning ability evaluation result.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the stress array-based method for assessing the turning steady state of the elderly as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the stress array-based method for assessing the steady-state shift of elderly individuals as described in any one of claims 1 to 7.

Images