A smart assessment method for gait safety in elderly people with functional decline

Abstract

This invention relates to the field of medical intelligent detection and discloses an intelligent assessment method for gait safety in elderly patients with functional decline. It synchronously collects multi-dimensional data using an infrared three-dimensional motion capture device and an array-type pressure sensing platform. Combining static human body change coefficient calculation, dynamic postural stability assessment, and modular data fusion analysis, a weighted summation scoring model is constructed to accurately determine gait safety risk levels, decline types, and specific risk points. The method also outputs personalized exercise intervention, assistive device adaptation, and rehabilitation training programs. This invention solves the problems of existing technologies, such as single data collection dimensions, lack of synchronous fusion of static physiological parameters and dynamic postural data, and insufficient quantitative accuracy.

Description

Technical Field

[0001] This invention relates to the field of medical intelligent detection, and in particular to an intelligent assessment method for gait safety in elderly people with functional decline. Background Technology

[0002] With the increasing aging of the population, the incidence of falls, joint injuries, and other accidents among the elderly is significantly higher due to physiological decline such as muscle atrophy, bone degeneration, and decreased balance. These issues seriously affect the quality of life of the elderly and can even endanger their lives. Gait, as a core and sensitive indicator reflecting the comprehensive physical function of the elderly, including muscle strength, balance control, and joint function, is of great clinical and social significance for safety assessment. It is crucial for early risk identification, the development of personalized rehabilitation plans, and the reduction of accidental events.

[0003] Existing gait assessment methods suffer from several shortcomings: they generally suffer from limited data collection dimensions, often focusing only on skeletal movement trajectories or single indicators of plantar pressure, lacking the synchronous integration of static physiological parameters and dynamic posture, resulting in one-sided and incomplete assessment results; assessment algorithms often use general models designed for the general population, failing to construct specific analytical models that incorporate the unique physiological patterns of musculoskeletal degeneration in the elderly, resulting in insufficient quantitative precision and difficulty in accurately capturing the core characteristics of gait abnormalities in the elderly; the intervention recommendations given are often too generalized, failing to be precisely adapted to the specific risk level, type of decline, and individual differences in the assessment results, thus limiting their practicality and operability; furthermore, some technical solutions lack clear equipment parameters, missing key indicators such as sampling frequency and sensor density, leading not only to poor repeatability of assessment results in different scenarios but also increasing the difficulty of clinical promotion and application, failing to meet the urgent needs of primary healthcare institutions and elderly care facilities for efficient, accurate, and easy-to-use gait assessment technologies.

[0004] To address this, the present invention proposes an intelligent assessment method for gait safety of the elderly that features multi-dimensional synchronous data collection, precise quantitative analysis, and personalized intervention, thereby overcoming the shortcomings of existing technologies. Summary of the Invention

[0005] The present invention aims to provide an intelligent assessment method for gait safety in elderly people with functional decline, in order to solve the problems of single data collection dimension, lack of synchronous integration of static physiological parameters and dynamic body posture in the existing technology, and insufficient quantitative accuracy.

[0006] To achieve the above objectives, the present invention provides the following method:

[0007] This invention provides a gait safety intelligent assessment method for functional decline in the elderly:

[0008] S1: The gait of the elderly is recorded synchronously in multiple dimensions using an infrared three-dimensional motion capture device and an array-type pressure sensing platform. The recorded data includes the movement trajectory of key skeletal nodes, foot pressure distribution, and basic gait parameters such as cadence, stride length, and stride width. The recorded data is analyzed and evaluated from three dimensions, namely static human body shape, human body posture during movement, and modular analysis of human data. The static human body shape refers to the static characteristics of physiological parameters, the human body posture during movement refers to the dynamic changes during walking, and the modular analysis of human data is the feature extraction based on functional attributes.

[0009] S2: Collect static physiological parameters of elderly people, such as height, weight, limb length ratio, and joint mobility, as detection data of the static human body shape. After standardizing the detection data through a preset algorithm, calculate the static human body change coefficient. The static human body change coefficient is used to quantify the degree of muscle atrophy, bone degeneration, and body shape deformation in the elderly. Based on the static human body change coefficient, establish a static human body change coefficient analysis chart. Based on the trend characteristics, fluctuation patterns, and distribution of abnormal points of the static human body change coefficient analysis chart, predict static human body changes.

[0010] S3: Based on the principles of sports biomechanics and the physiological degeneration of bones and muscles in the elderly, a human motion analysis model is constructed according to the human body posture during exercise. The human motion analysis model includes joint motion analysis, stability assessment and abnormality identification functions. The human body posture during exercise is quantified through the human motion analysis model, and key indicators reflecting gait stability, coordination and standardization are extracted. The model outputs a dynamic trend chart of human body posture data changes during exercise.

[0011] S4: Collect human body data, divide the human body data into physiological function module, motor control module and balance ability module, extract feature parameters of human body data of each module, fuse the static human body change coefficient analysis chart, the dynamic human body posture data change chart and the feature parameters obtained by the sub-analysis of human body data, and combine them with preset evaluation standards to conduct gait safety assessment, identify imbalance risk, joint injury risk and fall risk, and obtain gait safety intelligent assessment results including risk level, decline type and targeted intervention suggestions.

[0012] Preferably, the sampling frequency of the infrared three-dimensional motion capture device is not less than 120Hz, and the number of skeletal nodes identified is not less than 24, covering key points of the head, neck, shoulder, elbow, wrist, hip, knee, ankle, and sole of the foot; the array-type pressure sensing platform has a sensing unit density of not less than 2 units / cm², a sampling rate of not less than 50Hz, a pressure measurement range of 0~1000N, and a measurement accuracy of ±2%FS. Its sensing density and sampling rate match the capture frequency of the infrared three-dimensional motion capture device, realizing the synchronous acquisition of sole pressure distribution and skeletal movement trajectory, and eliminating analysis bias caused by data time difference.

[0013] Preferably, the preset algorithm includes: data standardization processing and static human body change coefficient calculation: the data standardization processing involves normalizing and calibrating the original detection value of a single static physiological parameter with the mean and standard deviation of the parameter in a healthy population of the same age, wherein the mean and standard deviation of the healthy population of the same age are obtained through a survey of no less than 1,000 individuals of the same age without musculoskeletal diseases, and individual baseline differences are eliminated through numerical adjustment to unify the evaluation benchmark; the static human body change coefficient is calculated by weighting and summing the muscle atrophy coefficient, bone degeneration coefficient, and body shape deformation coefficient, wherein the muscle atrophy coefficient accounts for 40%, the bone degeneration coefficient accounts for 35%, and the... The postural deformity coefficient accounts for 25%, and its weight is determined based on the correlation analysis between various decline indicators and gait risk in clinical data statistics; the static human body change coefficient: the value ranges from 0 to 1, and the larger the value, the more severe the comprehensive decline in muscle atrophy, skeletal degeneration and postural deformity in the elderly; the muscle atrophy coefficient: it is calculated by weighting the deviation rate of the upper arm circumference, thigh circumference and calf circumference of the elderly and the standard circumference of healthy people of the same age; the skeletal degeneration coefficient: it is calculated by averaging the decline rate of the range of motion of the three major joints of the hip, knee and ankle of the elderly and the standard range of motion of healthy people of the same age; the postural deformity coefficient: it is calculated by weighting and quantifying the Cobb angle of spinal scoliosis and pelvic tilt of the elderly, with a weight ratio of 6:4.

[0014] Preferably, the static human body change coefficient analysis chart is a combination of a "time-change coefficient" line chart and an outlier scatter plot, used to visually present the static human body decline trend and abnormal fluctuations; the trend characteristics are judged by the slope of the line, with a slope greater than 0.02 per year indicating rapid decline, a slope between 0.005 and 0.02 per year indicating slow decline, and a slope less than 0.005 per year indicating stability; the fluctuation pattern is quantified by the standard deviation of the last three test data, with a standard deviation less than 0.03 indicating gentle fluctuations, and a standard deviation not less than 0.03 indicating violent fluctuations; outliers are identified by the 3σ criterion, with test points deviating from the mean ± 3σ being judged as abnormal. After excluding interference from temporary physiological states by combining clinical examinations, the outliers are incorporated into the static human body change prediction model to improve prediction accuracy.

[0015] Preferably, the joint motion analysis in the human motion analysis model targets the three major joints of hip, knee, and ankle, extracting peak joint angles, average angular velocities, and angle variation ranges to reflect the flexibility and standardization of joint movements. The stability assessment is based on the pressure center trajectory, quantifying stability by calculating the root mean square of the trajectory. The specific calculation logic is as follows: for all coordinate points of the pressure center in the horizontal plane within the sampling period, calculate the squared deviation of each coordinate point's lateral position from the average of all lateral positions and the squared deviation of its longitudinal position from the average of all longitudinal positions. Sum the squared deviations of the two types of deviations, divide the sum by the number of sampling points, and finally take the square root of the result to obtain the root mean square of the trajectory. This value directly reflects the stability of the body's center of gravity during walking. The stability judgment criteria are: the root mean square of the trajectory is less than 5 mm, the root mean square of the trajectory is between 5 and 10 mm, the trajectory is generally stable, and the root mean square of the trajectory is greater than 10 mm, indicating instability. The anomaly identification uses a dynamic threshold method to detect abrupt changes in joint movement and sudden changes in gait frequency. The dynamic threshold is based on the 95% confidence interval of gait data from healthy individuals of the same age and is calculated using statistical methods to ensure the specificity and sensitivity of anomaly identification.

[0016] Preferably, the change map of the human body posture data is a three-dimensional heatmap, with the horizontal axis representing time and the vertical axis representing joint type, including hip, knee, and ankle types. The z-axis represents the stability score, which is calculated by weighting the trajectory root mean square (RMS) score (60%) with the gait coordination index (40%), with a maximum score of 100. The gait coordination index is obtained by summing the normalized results of the step frequency variation coefficient and stride length consistency. The dynamic change trend is quantified by trend line fitting. A trend line fitting degree of not less than 0.8 indicates a significant trend, while a trend line fitting degree of less than 0.8 indicates an insignificant trend. The three-dimensional heatmap annotates the key indicator mutation points and corresponding time nodes for each detection, providing a basis for dynamic posture changes for the fusion evaluation in step S4.

[0017] Preferably, the physiological function module includes: muscle strength (grip strength detected by an electronic grip strength meter and lower limb push-off force detected by a leg press meter); bone mineral density (bone mineral density values ​​of the L2-L4 lumbar vertebrae and femoral neck, obtained by dual-energy X-ray absorptiometry); joint range of motion (hip flexion-extension angle, knee flexion-extension angle, and ankle dorsiflexion angle, obtained by a joint goniometer); the motion control module includes: gait frequency variation coefficient calculated from the gait cycle time standard deviation; stride length consistency (ratio of left-right stride length difference to average stride length); time difference of peak pressure on the left and right feet extracted from pressure sensing platform data; and gait cycle stability (time variation coefficient of 10 consecutive gait cycles); and the balance ability module includes: single-leg standing maintenance time measured by a stopwatch with the maximum value obtained from three measurements; pressure center offset distance calculated from pressure sensing platform data in a closed-eye upright state; and maximum trunk tilt angle extracted from infrared motion capture device data during walking.

[0018] Preferably, the fusion processing employs a weighted summation algorithm. The specific calculation steps are as follows: the fusion score is derived by weighting the static trend score, dynamic score, and modular comprehensive score, where the static trend score has a weight of 30%, the dynamic score has a weight of 40%, and the modular comprehensive score has a weight of 30%. These weights are determined using the analytic hierarchy process (AHP). Specifically, this involves constructing a "gait safety assessment" target layer, a "static / dynamic / modal" criterion layer, and a layer of specific indicators. Weight vectors are calculated using pairwise comparison matrices, and the final weights are determined after a consistency check (CR1). The fusion score ranges from 0 to 100 points, comprehensively assessing the gait safety level of the elderly. The static trend score is based on the slope percentage of the broken line in the static human body variation coefficient analysis graph (60%) and the last three... The standard deviation of the test data is normalized to 40%, with a maximum score of 30. The dynamic score is normalized to 70% based on the stability score and 30% based on the trend line fit, with a maximum score of 40. The comprehensive score of each module is a weighted sum of the matching degree between the characteristic parameters of the physiological function module, motor control module, and balance ability module and the standard values. The weights within each module are set according to the correlation between the parameters and gait risk. For the physiological function module, muscle strength is 40%, bone density is 35%, and joint mobility is 25%. For the motor control module, gait frequency variation coefficient is 20%, stride length consistency is 30%, peak pressure time difference is 25%, and gait cycle stability is 25%. For the balance ability module, single-leg standing time is 35%, pressure center offset distance is 35%, and trunk tilt angle is 30%, with a maximum score of 30.

[0019] Preferably, the preset assessment criteria include: Risk level: a fusion score of not less than 85 points indicates low risk, corresponding to a fall incidence rate of <5%; a fusion score between 60 and 85 points indicates medium risk, corresponding to a fall incidence rate of 5% to 20%; a fusion score below 60 points indicates high risk, corresponding to a fall incidence rate of >20%; Decline type: determined by the score ratio of each module, a score ratio of less than 25% for the physiological function module indicates physiological decline, a score ratio of less than 35% for the motor control module indicates motor control decline, a score ratio of less than 25% for the balance ability module indicates balance decline, and a score ratio of less than two or more modules indicates comprehensive decline; Risk identification: Imbalance risk corresponds to a balance ability module score below 20 points, corresponding to a single-leg standing time of 0 seconds or a pressure center offset distance >20mm; Joint injury risk corresponds to at least 3 angle abrupt changes in joint motion analysis, corresponding to a single abrupt change angle >15°; Fall risk corresponds to a trajectory root mean square greater than 8mm and step length consistency less than 80% in the stability assessment.

[0020] Preferably, the targeted intervention recommendations are formulated based on the risk level, the type of decline, and the risk identification results: Exercise intervention: Low-risk individuals will engage in brisk walking and simplified 24-form Tai Chi three times a week, 30 minutes each time (20 minutes of brisk walking and 10 minutes of Tai Chi) to maintain physiological function and balance; Medium-risk individuals will train four times a week, 40 minutes each time, with resistance training targeting weak muscle groups and balance training including single-leg standing and heel walking, alternating between the two; High-risk individuals will receive one-on-one guidance from a rehabilitation therapist, with personalized training five times a week, 30 minutes each time, focusing on low-intensity balance training and joint mobility training, gradually increasing the intensity; Assistive device intervention: Individuals at risk of joint injury will select protective devices according to the type of injured joint. For patients with hip osteoarthritis, rigid hip braces are used; for knee osteoarthritis, patellar compression knee braces are used; and for ankle osteoarthritis, elastic ankle braces are used. For patients with a risk of falls, those with a balance score of 20-25 should use a cane with a length equal to 0.7 times their height. Wheeled walking aids with a wheel diameter of 8-10cm and braking function are also used, with a walking aid width and stride width matching error of ±2cm. Rehabilitation training interventions include: for those with physiological decline, strengthening muscle strength and improving bone density; for those with motor control decline, gait standardization training is conducted, with a stride frequency controlled at 80-100 steps / min; for those with balance decline, vestibular function and core stability training is performed; for those with comprehensive decline, a composite training program is developed by the rehabilitation team, implemented in the order of balance, motor control, and physiological function training, with intensity adjusted weekly and effects assessed monthly.

[0021] The beneficial effects of this invention are as follows: This invention achieves simultaneous recording of static physiological parameters, dynamic body posture, and modular functional data, providing comprehensive evaluation dimensions and avoiding the limitations of single indicators; it specifically designs static human body change coefficients and dynamic stability assessment algorithms, combining the degenerative patterns of the elderly, making the quantitative results more consistent with actual physiological states; it formulates personalized solutions based on risk levels, decline types, and specific risk points, adapting to the functional states of different elderly individuals; it clarifies the core parameters of the equipment and algorithm logic, ensuring the consistency of assessment results in different scenarios, facilitating clinical promotion and application. Attached Figure Description

[0022] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0023] Figure 1 This is a flowchart illustrating a gait safety intelligent assessment method for functional decline in the elderly, provided as an embodiment of the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0026] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0027] Existing gait assessment methods suffer from several shortcomings: they generally suffer from limited data collection dimensions, often focusing only on skeletal movement trajectories or single indicators of plantar pressure, lacking the synchronous integration of static physiological parameters and dynamic posture, resulting in one-sided and incomplete assessment results; assessment algorithms often use general models designed for the general population, failing to construct specific analytical models that incorporate the unique physiological patterns of musculoskeletal degeneration in the elderly, resulting in insufficient quantitative precision and difficulty in accurately capturing the core characteristics of gait abnormalities in the elderly; the intervention recommendations given are often too generalized, failing to be precisely adapted to the specific risk level, type of decline, and individual differences in the assessment results, thus limiting their practicality and operability; furthermore, some technical solutions lack clear equipment parameters, missing key indicators such as sampling frequency and sensor density, leading not only to poor repeatability of assessment results in different scenarios but also increasing the difficulty of clinical promotion and application, failing to meet the urgent needs of primary healthcare institutions and elderly care facilities for efficient, accurate, and easy-to-use gait assessment technologies.

[0028] To address this, the present invention proposes an intelligent assessment method for gait safety of the elderly that features multi-dimensional synchronous data collection, precise quantitative analysis, and personalized intervention, thereby overcoming the shortcomings of existing technologies.

[0029] The present invention aims to provide an intelligent assessment method for gait safety in elderly people with functional decline, in order to solve the problems of single data collection dimension, lack of synchronous integration of static physiological parameters and dynamic body posture in the existing technology, and insufficient quantitative accuracy.

[0030] like Figure 1 As shown in the figure, a specific embodiment of the present invention provides a gait safety intelligent assessment method for functional decline in the elderly, including the following steps:

[0031] S1: The gait of the elderly is recorded synchronously in multiple dimensions using an infrared three-dimensional motion capture device and an array-type pressure sensing platform. The recorded data includes the movement trajectory of key skeletal nodes, plantar pressure distribution, and basic gait parameters such as cadence, stride length, and stride width. The recorded data is analyzed and evaluated in three dimensions, namely static human body shape, human body posture during movement, and modular analysis of human data. Static human body shape refers to the static characteristics of physiological parameters, human body posture during movement refers to the dynamic changes during walking, and modular analysis of human data involves feature extraction based on functional attributes.

[0032] In this embodiment of the invention, the sampling frequency of the infrared three-dimensional motion capture device is not less than 120Hz, and the number of skeletal nodes identified is not less than 24, covering key points of the head, neck, shoulder, elbow, wrist, hip, knee, ankle, and sole of the foot; the array-type pressure sensing platform has a sensing unit density of not less than 2 units / cm², a sampling rate of not less than 50Hz, a pressure measurement range of 0~1000N, and a measurement accuracy of ±2%FS. Its sensing density and sampling rate match the capture frequency of the infrared three-dimensional motion capture device, realizing the synchronous acquisition of sole pressure distribution and skeletal movement trajectory, and eliminating the analysis deviation caused by data time difference.

[0033] S2: Collect static physiological parameters such as height, weight, limb length ratio, and joint mobility of elderly people as static human body shape detection data. After standardizing the detection data through a preset algorithm, calculate the static human body change coefficient. The static human body change coefficient is used to quantify the degree of muscle atrophy, bone degeneration, and body shape deformation in the elderly. Based on the static human body change coefficient, establish a static human body change coefficient analysis chart. Based on the trend characteristics, fluctuation patterns, and abnormal point distribution of the static human body change coefficient analysis chart, predict static human body changes.

[0034] In this embodiment of the invention, the preset algorithm includes: data standardization processing and static human body change coefficient calculation: data standardization processing involves normalizing and calibrating the original detection value of a single static physiological parameter with the mean and standard deviation of that parameter in a healthy population of the same age. The mean and standard deviation of the healthy population of the same age are obtained through a survey of no less than 1,000 individuals without musculoskeletal diseases. Individual baseline differences are eliminated through numerical adjustment to unify the assessment benchmark. The static human body change coefficient is calculated by weighting and summing the muscle atrophy coefficient, skeletal degeneration coefficient, and body shape deformity coefficient according to their respective weights. The muscle atrophy coefficient accounts for 40%, the skeletal degeneration coefficient accounts for 35%, and the body shape deformity coefficient accounts for 25%. The weights are determined based on the correlation analysis between various decline indicators and gait risk in clinical data statistics. The static human body change coefficient ranges from 0 to 1. The larger the value, the more severe the comprehensive decline in muscle atrophy, skeletal degeneration, and body shape deformity in the elderly. The muscle atrophy coefficient is based on the upper arm circumference, thigh circumference, and calf circumference of the elderly and the standard of healthy populations of the same age. The deviation rate of circumference is calculated using a weighted average. The skeletal degeneration coefficient is calculated based on the average attenuation rate of the hip, knee, and ankle joint range of motion in the elderly compared to the standard range of motion in healthy individuals of the same age. The postural deformation coefficient is calculated based on a weighted and quantified sum of the Cobb angle of spinal scoliosis and pelvic tilt in the elderly, with a weight ratio of 6:4. The static human body change coefficient analysis chart is a combination of a "time-change coefficient" line graph and an anomaly point scatter plot, used to visually present the static human body decline trend and abnormal fluctuations. Trend characteristics are judged by the slope of the line graph: a slope greater than 0.02 per year indicates rapid decline, a slope between 0.005 and 0.02 per year indicates slow decline, and a slope less than 0.005 per year indicates stability. Fluctuation patterns are quantified by the standard deviation of the last three test data: a standard deviation less than 0.03 indicates gentle fluctuations, and a standard deviation not less than 0.03 indicates severe fluctuations. Anomalies are identified using the 3σ criterion; test points deviating from the mean ± 3σ are considered abnormal. After excluding temporary physiological interference by combining clinical examinations, anomalies are incorporated into the static human body change prediction model to improve prediction accuracy.

[0035] S3: Based on the principles of sports biomechanics and the physiological degeneration of bones and muscles in the elderly, a human motion analysis model is constructed according to the human body posture during movement. The human motion analysis model includes joint motion analysis, stability assessment and abnormality identification functions. The human motion analysis model is used to quantify the human body posture during movement, extract key indicators that reflect gait stability, coordination and standardization, and output a dynamic trend chart of human body posture data changes during movement.

[0036] In this embodiment of the invention, the joint motion analysis in the human motion analysis model targets the three major joints: hip, knee, and ankle. It extracts the peak value of joint angles, the average value of angular velocities, and the range of angle changes to reflect the flexibility and standardization of joint movement. Stability assessment is based on the pressure center trajectory. Stability is quantified by calculating the root mean square (RMS) of the trajectory. The specific calculation logic is as follows: for all coordinate points of the pressure center in the horizontal plane within the sampling period, calculate the squared deviation of each coordinate point's lateral position from the average value of all lateral positions, and the squared deviation of its longitudinal position from the average value of all longitudinal positions. Summate the two types of squared deviations, divide the sum by the number of sampling points, and finally take the square root of the result to obtain the RMS. This value directly reflects the stability of the body's center of gravity during walking. Stability judgment criteria: a RMS of less than 5mm indicates stability, a RMS between 5 and 10mm indicates general stability, and a RMS greater than 10mm indicates instability. Anomaly identification is also performed. A dynamic threshold method was used to detect abrupt joint movement changes and sudden changes in gait frequency. The dynamic threshold was calculated using statistical methods based on the 95% confidence interval of gait data from healthy individuals of the same age, ensuring the specificity and sensitivity of anomaly identification. The change in human body posture data was presented as a three-dimensional heatmap, with the horizontal axis representing time, the vertical axis representing joint type (hip, knee, and ankle), and the z-axis representing the stability score. The stability score was calculated by weighting the trajectory root mean square (RMS) score (60%) and the gait coordination index (40%), with a maximum score of 100. The gait coordination index was obtained by summing the normalized results of the gait frequency coefficient of variation and stride length consistency. The dynamic change trend was quantified by trend line fitting. A trend line fit of ≥0.8 was considered significant, while a fit of <0.8 was considered insignificant. The three-dimensional heatmap annotated the key indicator abrupt changes and corresponding time points for each detection, providing a basis for dynamic posture changes in step S4 for the fusion evaluation.

[0037] S4: Collect human body data and divide it into physiological function module, motor control module, and balance ability module. Extract the feature parameters of human body data from each module. Combine the static human body change coefficient analysis chart, the dynamic human body posture data change chart, and the feature parameters obtained from the sub-module analysis of human body data for fusion processing. Combine the preset evaluation standards to conduct gait safety assessment, identify the risk of imbalance, joint injury risk, and fall hazards, and obtain gait safety intelligent assessment results including risk level, decline type, and targeted intervention suggestions.

[0038] In this embodiment of the invention, the physiological function module includes: muscle strength comprising grip strength detected by an electronic grip strength meter and lower limb push-off force detected by a leg press meter; bone mineral density comprising bone mineral density values ​​of the L2-L4 lumbar vertebrae and femoral neck, obtained by dual-energy X-ray absorptiometry; joint range of motion comprising hip flexion-extension angle, knee flexion-extension angle, and ankle dorsiflexion angle, obtained by joint goniometer; and the motion control module includes: gait frequency variation coefficient calculated from the standard deviation of gait cycle time; stride length consistency as the ratio of the difference in left and right stride lengths to the average stride length; time difference of peak plantar pressure between the left and right feet extracted from pressure sensor platform data; and gait cycle stability as the time variation over 10 consecutive gait cycles. Coefficients; Balance Ability Module: The single-leg standing duration was timed with a stopwatch and the maximum value was obtained from three measurements. The offset distance of the pressure center in the closed-eye upright state was calculated from the pressure sensing platform data. The maximum trunk tilt angle during walking was extracted from the infrared motion capture device data. The fusion processing adopted a weighted summation algorithm. The specific calculation steps are as follows: The fusion score is obtained by superimposing the static trend score, dynamic score, and sub-module comprehensive score according to their weights. Among them, the static trend score has a weight of 30%, the dynamic score has a weight of 40%, and the sub-module comprehensive score has a weight of 30%. The weights are determined by the analytic hierarchy process. The specific steps are to construct the "gait safety assessment" target layer, the "static / dynamic / sub-module" criterion layer, and... For each specific indicator layer, a weight vector is calculated through pairwise comparison matrices, and the final weight is determined after a consistency test (CR1). The fusion score ranges from 0 to 100 points, comprehensively assessing the gait safety level of the elderly. The static trend score is calculated by normalizing the slope of the broken line in the static human body change coefficient analysis chart (60%) and the standard deviation of the last three test data (40%), with a maximum score of 30 points. The dynamic score is calculated by normalizing the stability score (70%) and the trend line fit (30%), with a maximum score of 40 points. The module-specific comprehensive score is a weighted sum of the matching degree between the characteristic parameters and standard values ​​of the physiological function module, motor control module, and balance ability module. The weights within each module are determined according to the parameters and... The correlation settings for gait risk are as follows: Physiological Function Module: Muscle Strength 40%, Bone Mineral Density 35%, Joint Range of Motion 25%; Motor Control Module: Step Frequency Coefficient of Variation 20%, Stride Consistency 30%, Peak Stress Time Difference 25%, Gait Cycle Stability 25%; Balance Ability Module: Single-Leg Standing Time 35%, Center of Stress Offset Distance 35%, Trunk Tilt Angle 30%, with a maximum score of 30 points. Preset assessment criteria include: Risk Level: A fusion score of 85 or higher indicates low risk, corresponding to a fall incidence rate of <5%; a fusion score between 60 and 85 indicates medium risk, corresponding to a fall incidence rate of 5% to 20%; a fusion score below 60 indicates high risk, corresponding to a fall incidence rate of >20%.Decline type: Determined by the percentage of scores in each module. Physiological function module score below 25% indicates physiological decline; motor control module score below 35% indicates motor control decline; balance module score below 25% indicates balance decline; two or more modules failing to meet the standard indicate comprehensive decline. Risk identification: Imbalance risk corresponds to a balance module score below 20, a single-leg standing time of 0 seconds, or a pressure center offset distance >20mm; Joint injury risk corresponds to at least 3 angle abrupt changes in joint motion analysis, with a single abrupt change angle >15°; Fall risk corresponds to a trajectory root mean square greater than 8mm and stride consistency below 80% in stability assessment. Targeted intervention recommendations are formulated based on risk level, decline type, and risk identification results: Exercise intervention: Low-risk individuals: 3 times per week brisk walking and 24-style simplified Tai Chi, 30 minutes each time (20 minutes brisk walking and 10 minutes Tai Chi) to maintain physiological function and balance; Medium-risk individuals: 4 times per week training, 40 minutes each time, resistance training targeting weak muscles. Muscle group balance training includes single-leg standing and heel walking, alternating between the two. High-risk individuals receive one-on-one guidance from rehabilitation therapists, with 5 personalized training sessions per week, 30 minutes each, focusing on low-intensity balance and joint mobility training, gradually increasing the intensity. Assistive device intervention: Individuals at risk of joint injury are provided with protective gear selected according to the type of joint injury; rigid hip braces are used for hip osteoarthritis, patellar compression knee braces for knee osteoarthritis, and elastic ankle braces for ankle osteoarthritis. For patients with a risk of falls, a balance score of 20-25 points is required to use a cane with a length equal to 0.7 times their height; wheeled walking aids with a wheel diameter of 8-10 cm and braking function are also used, with a walking aid width and stride width matching error of ±2 cm. Rehabilitation training intervention: For physiological decline, muscle strength and bone density improvement training are provided; for motor control decline, gait standardization training is conducted, with a stride frequency controlled at 80-100 steps / min; for balance decline, vestibular function and core stability training are conducted; for comprehensive decline, a composite training program is developed by the rehabilitation team, implemented in the order of balance, motor control, and physiological function training, with intensity adjusted weekly and results assessed monthly.

[0039] Example 1

[0040] I. Subject Selection

[0041] Fifty elderly individuals aged 60-80 years were selected, excluding those with severe cardiovascular and cerebrovascular diseases, severe bone and joint deformities, and cognitive impairment. Among them, there were 23 males and 27 females, with an average age of 72.3 years.

[0042] II. Equipment Preparation

[0043] Infrared 3D motion capture device: Model OptiTrack Flex 13, sampling frequency set to 120Hz, number of skeletal nodes recognized: 26 (covering the key points of claim 2);

[0044] Array-type pressure sensing platform: Model Tekscan HRMat, sensing unit density 2.5 units / cm², sampling rate set to 50Hz, pressure measurement range 0~1000N, measurement accuracy ±1.5%FS;

[0045] Auxiliary testing equipment: electronic grip strength meter (accuracy ±0.1kg), leg strength meter (range 0~500N), dual-energy X-ray bone densitometer (accuracy ±0.01g / cm²), articular goniometer (accuracy ±1°), stopwatch (accuracy ±0.01s).

[0046] III. Implementation Steps

[0047] Static physiological parameter collection: Measure the subject's height, weight, upper arm circumference, thigh circumference, calf circumference, hip / knee / ankle joint range of motion, Cobb angle of scoliosis, and pelvic tilt, and record the raw data;

[0048] Dynamic gait data acquisition: The subject walked barefoot on the pressure sensing platform at a normal walking speed 10 times, and the infrared motion capture device was activated simultaneously to record the skeletal movement trajectory, plantar pressure distribution, and parameters such as step frequency, step length, and step width.

[0049] Data processing:

[0050] Static data standardization processing is performed according to the algorithm of claim 3 to calculate the muscle atrophy coefficient, bone degeneration coefficient, body shape deformation coefficient and static human body change coefficient (average coefficient 0.32).

[0051] The root mean square of the trajectory (average value 6.8mm) is calculated according to the logic of claim 5, and joint motion indicators and abnormal points are extracted.

[0052] The fusion score was calculated using the weighted summation algorithm of claim 8 (average score 76.5 points).

[0053] Assessment results: Of the 50 subjects, 18 were low-risk (score ≥ 85 points), 25 were medium-risk (score 60-85 points), and 7 were high-risk; the types of decline included physiological decline (8), motor control decline (12), balance decline (10), and comprehensive decline (20).

[0054] Intervention implementation: A personalized intervention plan was developed for each subject according to the protocol of claim 10. After a follow-up examination after 3 months, the fall incidence rate was 0% in the low-risk group, 4% in the medium-risk group, and 8.6% in the high-risk group, which was significantly lower than before the intervention.

[0055] IV. Result Verification

[0056] Post-intervention follow-up data showed that the average static human body change coefficient of the subjects decreased to 0.28, the average root mean square of the trajectory decreased to 5.2 mm, the average fusion score increased to 82.3 points, and key indicators such as joint mobility and balance ability were improved, verifying the effectiveness and practicality of the technical solution of the present invention.

[0057] The beneficial effects of this invention are as follows: This invention achieves simultaneous recording of static physiological parameters, dynamic body posture, and modular functional data, providing comprehensive evaluation dimensions and avoiding the limitations of single indicators; it specifically designs static human body change coefficients and dynamic stability assessment algorithms, combining the degenerative patterns of the elderly, making the quantitative results more consistent with actual physiological states; it formulates personalized solutions based on risk levels, decline types, and specific risk points, adapting to the functional states of different elderly individuals; it clarifies the core parameters of the equipment and algorithm logic, ensuring the consistency of assessment results in different scenarios, facilitating clinical promotion and application.

[0058] The above descriptions are merely embodiments of the present invention. Commonly known technical solutions or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A smart assessment method for gait safety in elderly individuals with functional decline, characterized in that, The method includes: S1: The gait of the elderly is recorded synchronously in multiple dimensions using an infrared three-dimensional motion capture device and an array-type pressure sensing platform. The recorded data includes the movement trajectory of key skeletal nodes, foot pressure distribution, and basic gait parameters such as cadence, stride length, and stride width. The recorded data is analyzed and evaluated from three dimensions, namely static human body shape, human body posture during movement, and modular analysis of human data. The static human body shape refers to the static characteristics of physiological parameters, the human body posture during movement refers to the dynamic changes during walking, and the modular analysis of human data is the feature extraction based on functional attributes. S2: Collect static physiological parameters of elderly people, such as height, weight, limb length ratio, and joint mobility, as detection data of the static human body shape. After standardizing the detection data through a preset algorithm, calculate the static human body change coefficient. The static human body change coefficient is used to quantify the degree of muscle atrophy, bone degeneration, and body shape deformation in the elderly. Based on the static human body change coefficient, establish a static human body change coefficient analysis chart. Based on the trend characteristics, fluctuation patterns, and distribution of abnormal points of the static human body change coefficient analysis chart, predict static human body changes. S3: Based on the principles of sports biomechanics and the physiological degeneration of bones and muscles in the elderly, a human motion analysis model is constructed according to the human body posture during exercise. The human motion analysis model includes joint motion analysis, stability assessment and abnormality identification functions. The human body posture during exercise is quantified through the human motion analysis model, and key indicators reflecting gait stability, coordination and standardization are extracted. The model outputs a dynamic trend chart of human body posture data changes during exercise. S4: Collect human body data, divide the human body data into physiological function module, motor control module and balance ability module, extract feature parameters of human body data of each module, fuse the static human body change coefficient analysis chart, the dynamic human body posture data change chart and the feature parameters obtained by the sub-analysis of human body data, and combine them with preset evaluation standards to conduct gait safety assessment, identify imbalance risk, joint injury risk and fall risk, and obtain gait safety intelligent assessment results including risk level, decline type and targeted intervention suggestions.

2. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 1, characterized in that: The infrared three-dimensional motion capture device has a sampling frequency of no less than 120Hz and a number of skeletal nodes to be recognized of no less than 24, covering key points of the head, neck, shoulder, elbow, wrist, hip, knee, ankle and sole of the foot; The array-type pressure sensing platform has a sensing unit density of no less than 2 units / cm², a sampling rate of no less than 50Hz, a pressure measurement range of 0~1000N, and a measurement accuracy of ±2%FS. Its sensing density and sampling rate match the capture frequency of the infrared three-dimensional motion capture device, realizing the synchronous acquisition of foot pressure distribution and skeletal movement trajectory, and eliminating the analysis deviation caused by data time difference.

3. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 1, characterized in that: The preset algorithm includes: data standardization processing and static human body variation coefficient calculation. The data standardization process is as follows: the original detection value of a single static physiological parameter is normalized and calibrated with the mean and standard deviation of the parameter in a healthy population of the same age. The mean and standard deviation of the healthy population of the same age are obtained through a survey of no less than 1,000 healthy individuals of the same age without musculoskeletal diseases. Individual differences are eliminated through numerical adjustment to unify the evaluation benchmark. The static human body change coefficient is calculated by weighting and summing the muscle atrophy coefficient, skeletal degeneration coefficient, and body shape deformity coefficient according to their respective weights. The muscle atrophy coefficient accounts for 40%, the skeletal degeneration coefficient accounts for 35%, and the body shape deformity coefficient accounts for 25%. The weights are determined based on the correlation analysis between various decline indicators and gait risk from clinical data statistics. Static human body change coefficient: The value ranges from 0 to 1. The larger the value, the more severe the comprehensive decline in muscle atrophy, bone degeneration and body shape deformation in the elderly. Muscle atrophy coefficient: calculated by weighting the deviation rate of upper arm circumference, thigh circumference, and calf circumference of the elderly compared with the standard circumference of healthy people of the same age; Bone degeneration coefficient: calculated based on the average rate of decline in the range of motion of the hip, knee and ankle joints of the elderly compared with the standard range of motion of healthy people of the same age; Postural deformity coefficient: It is derived from the weighted sum and quantification of the Cobb angle of spinal scoliosis and pelvic tilt in the elderly, with a weight ratio of 6:

4.

4. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 3, characterized in that: The static human body change coefficient analysis chart is a combination of a "time-change coefficient" line chart and an anomaly point scatter plot, used to visually present the static human body's decline trend and abnormal fluctuations. Trend characteristics are determined by the slope of the line. A slope greater than 0.02 per year indicates a rapid decline, a slope between 0.005 and 0.02 per year indicates a slow decline, and a slope less than 0.005 per year indicates stability. The fluctuation pattern is quantified by the standard deviation of the most recent three test data. A standard deviation less than 0.03 indicates gentle fluctuation, while a standard deviation not less than 0.03 indicates severe fluctuation. Outliers are identified by the 3σ criterion. Test points that deviate from the mean ±3σ are judged as abnormal. After excluding the interference of temporary physiological state by combining clinical examination, the outliers are incorporated into the static human body change prediction model to improve the prediction accuracy.

5. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 1, characterized in that: The joint motion analysis in the human motion analysis model targets the three major joints of hip, knee, and ankle, extracting the peak value of joint angle, the average value of angular velocity, and the range of angle variation to reflect the flexibility and standardization of joint motion. The stability assessment is based on the pressure center trajectory. Stability is quantified by calculating the root mean square of the trajectory. The specific calculation logic is as follows: For all coordinate points of the pressure center in the horizontal plane within the sampling period, calculate the square of the deviation of the lateral position of each coordinate point from the average value of all lateral positions and the square of the deviation of the longitudinal position from the average value of all longitudinal positions. Add the squares of the two types of deviations and sum them. Divide the sum by the number of sampling points. Finally, take the square root of the result to obtain the root mean square of the trajectory. This value directly reflects the stability of the body's center of gravity during walking. Stability determination criteria: The trajectory root mean square (RMS) is less than 5 mm, which indicates stability; the trajectory root mean square (RMS) is between 5 and 10 mm, which indicates general stability; and the trajectory root mean square (RMS) is greater than 10 mm, which indicates instability. The anomaly identification uses a dynamic threshold method to detect abrupt changes in joint movement and gait frequency. The dynamic threshold is calculated using statistical methods based on the 95% confidence interval of gait data from healthy individuals of the same age, ensuring the specificity and sensitivity of anomaly identification.

6. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 5, characterized in that: The change graph of the human body posture data is a three-dimensional heat map. The horizontal axis represents time, the vertical axis represents joint type, the joint types include three categories: hip, knee and ankle, and the z-axis represents stability score. The stability score is calculated by weighting the root mean square of the trajectory (60%) and the gait coordination index (40%), with a full score of 100. The gait coordination index is obtained by summing the normalized results of the step frequency variation coefficient and stride length consistency. The dynamic change trend is quantified by trend line fitting. A trend line fitting degree of not less than 0.8 indicates a significant trend, while a trend line fitting degree of less than 0.8 indicates an insignificant trend. The three-dimensional thermal chart marks the key indicator mutation points and corresponding time nodes for each detection, providing a basis for dynamic body shape changes for the fusion evaluation in step S4.

7. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 1, characterized in that: The physiological function module includes: muscle strength (grip strength measured by an electronic grip strength meter and lower limb push-off force measured by a leg press meter); bone mineral density (bone mineral density measured by dual-energy X-ray absorptiometry for the L2-L4 lumbar vertebrae and femoral neck); and joint range of motion (hip flexion-extension angle, knee flexion-extension angle, and ankle dorsiflexion angle, measured by an articular goniometer). The motion control module: the step frequency variation coefficient is calculated from the standard deviation of gait cycle time; the step length consistency is the ratio of the difference between the left and right step lengths to the average step length; the time difference of the peak occurrence of the left and right foot plantar pressure is extracted from the pressure sensing platform data; and the gait cycle stability is the time variation coefficient of 10 consecutive gait cycles. The balance ability module: the single-leg standing time is measured by a stopwatch and the maximum value is obtained from 3 measurements; the offset distance of the pressure center in the closed-eye upright state is calculated from the pressure sensing platform data; and the maximum value of the torso tilt angle during walking is extracted from the infrared motion capture device data.

8. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 7, characterized in that: The fusion processing adopts a weighted summation algorithm. The specific calculation steps are as follows: the fusion score is obtained by superimposing the static trend score, dynamic score, and sub-module comprehensive score according to their weights, wherein the static trend score has a weight of 30%, the dynamic score has a weight of 40%, and the sub-module comprehensive score has a weight of 30%. The weights are determined by the analytic hierarchy process. The specific steps are as follows: construct the "gait safety assessment" target layer, the "static / dynamic / sub-module" criterion layer, and the specific indicator layer. Calculate the weight vector through pairwise comparison matrices. Determine the final weights after passing the consistency test CR1. The fusion score ranges from 0 to 100 points, comprehensively assessing the gait safety level of the elderly. The static trend score is calculated by normalizing the slope of the broken line in the static human body change coefficient analysis chart (60%) and the standard deviation of the last three test data (40%), with a maximum score of 30. The dynamic score is calculated by normalization based on a stability score of 70% and a trend line fit score of 30%, with a maximum score of 40. The comprehensive score for each module is a weighted sum of the matching degree between the characteristic parameters and standard values ​​of the physiological function module, motor control module, and balance ability module. The weights within each module are set according to the correlation between the parameters and gait risk. For the physiological function module, muscle strength is 40%, bone density is 35%, and joint range of motion is 25%. For the motor control module, gait frequency variation coefficient is 20%, stride length consistency is 30%, peak stress time difference is 25%, and gait cycle stability is 25%. For the balance ability module, single-leg standing time is 35%, center of pressure offset distance is 35%, and trunk tilt angle is 30%. The maximum score is 30 points.

9. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 8, characterized in that, The preset evaluation criteria include: Risk Levels: A fusion score of 85 or higher indicates low risk, corresponding to a fall incidence rate of <5%; a fusion score between 60 and 85 indicates medium risk, corresponding to a fall incidence rate of 5% to 20%; a fusion score below 60 indicates high risk, corresponding to a fall incidence rate of >20%. Type of decline: Determined by the percentage of scores in each module. If the percentage of scores in the physiological function module is less than 25%, it is physiological decline type; if the percentage of scores in the motor control module is less than 35%, it is motor control decline type; if the percentage of scores in the balance ability module is less than 25%, it is balance decline type; if the percentage of scores in two or more modules is less than the standard, it is comprehensive decline type. Risk identification: Imbalance risk corresponds to a balance ability module score of less than 20 points, corresponding to a single-leg standing time of 0 seconds or a pressure center offset distance of >20mm; Joint injury risk corresponds to no less than 3 abrupt angle changes in joint motion analysis, corresponding to a single abrupt angle of >15°; Fall risk corresponds to a trajectory root mean square greater than 8mm and step length consistency of less than 80% in stability assessment.

10. The intelligent assessment method for gait safety in elderly people with functional decline according to claim 9, characterized in that: The targeted intervention recommendations are formulated based on the risk level, the recession type, and the risk identification results: Exercise intervention: Low-risk individuals should engage in brisk walking and simplified 24-form Tai Chi three times a week, 30 minutes each time, including 20 minutes of brisk walking and 10 minutes of Tai Chi, to maintain physiological functions and balance. For those at medium risk, training is conducted 4 times a week for 40 minutes each time. Resistance training targets weak muscle groups, and balance training includes standing on one leg and walking on the heel, alternating between the two. For those at high risk, one-on-one guidance from a rehabilitation therapist is provided, with personalized training 5 times a week for 30 minutes each time. The training focuses on low-intensity balance training and joint mobility training, gradually increasing the intensity. Assistive device intervention: For individuals at risk of joint injury, protective gear should be selected according to the type of joint injury. For hip osteoarthritis, a rigid hip brace should be used; for knee osteoarthritis, a patellar compression knee brace should be used; and for ankle osteoarthritis, an elastic ankle brace should be used. For patients with a risk of falling, a balance score of 20-25 points is required to use a cane with a length equal to height × 0.

7. Wheeled walking aids with a wheel diameter of 8-10 cm and a braking function should be used, with a walking aid width and stride width matching error of ±2 cm. Rehabilitation training interventions: For physiological decline, strengthen muscle strength and improve bone density through training; for motor control decline, conduct gait standardization training with a step frequency controlled at 80-100 steps / min; for balance decline, conduct vestibular function and core stability training; for comprehensive decline, the rehabilitation team will develop a composite training program, implemented in the order of balance, motor control, and physiological function training, with the intensity adjusted weekly and the effect assessed monthly.

Images