Health risk assessment method and system fusing action assessment and occupational exposure correction

Abstract

The application discloses a health risk assessment method and system combining motion evaluation and occupational exposure correction. The application collects human kinematics and pain VAS data through inertial sensing and posture recognition, automatically divides the four stages of motion and extracts multi-dimensional motion features, quantitatively calculates the three-dimensional score of motion by comparing with the standard template, and outputs the four types of functional determination results of FN / FP / DN / DP. The motion defects are mapped to multiple types of occupational exposure indexes such as neck and shoulder to construct an interactive amplification model, the occupational risks of different parts are calculated by weighting, and the occupational risks are divided into low, medium and high levels. Relying on the training library with multi-dimensional labels, the personalized training scheme is matched according to the risk, defect and pain type, the intervention priority is dynamically adjusted through the retest data, and the closed loop of 'evaluation-classification-intervention-retest' is formed. The application solves the problems of strong subjectivity of traditional SFMA, non-combination of post load, general intervention scheme, strong evaluation standardization and strong interpretability, and is suitable for normal screening and dynamic health management of muscle and bone injuries of enterprise workers.

Description

Technical Field

[0001] This invention relates to the fields of occupational health assessment, rehabilitation assessment information processing, sports biomechanics data analysis, and musculoskeletal injury risk screening, specifically a health risk assessment method and system that integrates movement assessment and occupational exposure correction. Background Technology

[0002] Selective functional movement assessment (SFMA) is the mainstream method for musculoskeletal function screening. Traditional methods rely on physicians' manual observation and scoring, which is highly subjective and has poor consistency in repeated tests. With the implementation of motion capture and inertial sensing technologies, the industry can quantify SFMA kinematic data and classify it into four functional states: FN, FP, DN, and DP. However, existing objective SFMA only analyzes human movement defects and does not take into account occupational work conditions, so it cannot distinguish the difference in injury risk of the same functional abnormality in different positions.

[0003] Current occupational musculoskeletal injury risk assessments mainly fall into three categories: ergonomic job scoring, questionnaire-based physical sign screening, and machine vision intelligent recognition. Ergonomic assessments only evaluate job load, ignoring individual functional differences; questionnaire screening relies on subjective descriptions, making it difficult to uncover potential motor compensations; intelligent recognition algorithms have poor interpretability, can only simply determine normal or abnormal, cannot quantify risks by body part, and all rely on single static assessments, lacking the ability to dynamically monitor trends through retesting.

[0004] Existing occupational health assessment systems only output general risk levels and lack a mechanism for automatically matching training recommendations across multiple dimensions. Intervention suggestions are broad and general, and cannot be customized in terms of injury location, functional deficit, pain status, and job load to determine training movements, intensity, and frequency. Nor can they set safe training boundaries based on the SFMA pain classification.

[0005] Therefore, a health risk assessment method and system that integrates motion assessment and occupational exposure modification was proposed to address the above-mentioned problems. Summary of the Invention

[0006] 1. The technical problem to be solved by the present invention

[0007] The purpose of this invention is to propose a health risk assessment method and system based on integrated motion assessment and occupational exposure correction to solve the problems of strong subjectivity, insufficient data quantification, inadequate correspondence between job exposure and individual functional impairment, and difficulty in converting risk results into actionable training recommendations in existing occupational health function assessments.

[0008] 2. Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a health risk assessment method and system integrating motion assessment and occupational exposure modification, comprising: S1. Objective processing of selective functional movement assessment: Collect kinematic data and pain feedback data of subjects, complete the division of movement stages and extraction of movement features, combine standard movement templates to quantitatively calculate multidimensional movement scores and output FN / FP / DN / DP functional classification results; S2, Occupational Exposure Interaction Risk Calculation: The functional defects obtained in S1 are mapped to each body part to form a body part level functional impairment score. The occupational exposure information of the job is standardized and coded to obtain the body part occupational exposure score. The interaction items of the two are combined to calculate the occupational related risk score of each body part and classify the risk level. S3. Layered matching of training recommendations and dynamic updates: Based on the risk level of each body part, the type of functional impairment, the pain classification, and the source of occupational exposure, a training recommendation library is matched to generate personalized training programs. The risk level and the priority of training recommendations are dynamically adjusted in combination with retest data, forming a closed loop of assessment-risk grading-training intervention-retest update.

[0009] Preferably, the S1 selective functional action assessment objectification process specifically includes: S2.1 Data Acquisition and Preprocessing: The joint angle time-series data of the subject performing selective functional movements are acquired through at least one of the following devices: inertial sensor, motion capture, and video posture recognition. The joint angular velocity and angular acceleration are obtained by differential operation. The raw kinematic data are filtered and smoothed, outlier removal is performed, and multi-source data are synchronized in time. The VAS visual analog scale score of pain induced by each movement of the subject is acquired simultaneously as pain feedback data. S2.2 Automatic four-stage motion recognition: Set the angular velocity start threshold, static threshold and joint peak angle error threshold to divide a single complete motion into four time intervals: start stage, main motion stage, final posture stage and return stage. S2.3 Multidimensional motion feature extraction: Extracting terminal joint angle, range of motion (ROM), peak angular velocity, peak angular acceleration, total motion duration, left and right limb index differences, compensatory motion features, and index fluctuation coefficients across multiple trials; S2.4 Standard Template Comparison and Multi-Dimensional Movement Scoring: Pre-store exclusive standard templates for each selective function movement. The templates include standard joint angle range, speed range, compensation threshold, left-right difference threshold, and stability threshold. Based on the templates, calculate the movement accuracy score, movement completion score, and movement stability score respectively. S2.5 Objective Functional Scoring and Four-Category Judgment: The objective functional score of a single action is obtained by weighted summation of the accuracy, completion and stability of the action; combined with the action pain VAS score, and with the preset functional threshold Tf as the boundary, the results are divided into four categories: FN painless and normal function, FP pain with normal function, DN painless and abnormal function, and DP pain with abnormal function.

[0010] Preferably, the formula for calculating the action accuracy score is:

[0011] In the formula, It is represented as a movement deviation, which is obtained by superimposing the weighted angle deviation, speed deviation, compensation deviation, and left and right limb difference deviation. The sum of the weights of the four deviations is equal to 1. The degree of action completion is calculated by weighting three indicators: final posture achievement, action path integrity, and uninterrupted action. The formula for calculating the stability of the action is:

[0012] In the formula, It is represented as stability fluctuation, which is a weighted sum of the coefficients of variation of the final angle of multiple trials, the coefficient of variation of the action duration, and the coefficient of variation of the compensatory fluctuation.

[0013] Preferably, the occupational exposure interaction risk calculation in S2 specifically includes: S4.1 Location-based functional impairment mapping: Establish a mapping weight matrix between single selective functional movements and body parts such as the neck and shoulders, lower back, hips, knees and ankles. A unified functional impairment score for each body part is obtained by weighted averaging of the functional deficit values ​​of single movements. S4.2 Occupational Exposure Quantification Coding: For each body part, five indicators, namely, job duration, frequency of repetitive movements, external load, static posture constraints, and tool vibration exposure, are assigned a 0-3 grade and weighted normalized to obtain the occupational exposure score for the body part. S4.3 Functional Defects - Occupational Exposure Interaction Amplification Calculation: The construction site interaction item is as follows:

[0014] In the formula, The selective motor dysfunction and occupational exposure interaction score for the r-th body part is used to quantify the superimposed amplification effect of individual functional deficit and job load. For the first Functional impairment score for each body part; This is represented as the occupational exposure score for the r-th body part; S4.4 Occupational Risk Classification Calculation: The occupational risk score of the part is calculated by weighting and integrating the part functional impairment score, occupational exposure score, interactive amplification item, part pain score, and retest trend score. The score is then divided into three risk levels: low, medium, and high, based on the segment threshold, and the core risk sources of each part are marked.

[0015] Preferably, the formula for scoring the occupational risk of the aforementioned body part is:

[0016] In the formula, Let r be the occupational risk score for the r-th body part. Pain trigger score for selective functional movements related to this area. The retest trend score is given.

[0017] Preferably, the S3 hierarchical matching training suggestion and dynamic update specifically includes: S6.1 Pre-built training suggestion library: Each training suggestion in the library is bound to a matching tag, which includes the applicable body part, risk level, SFMA four-category type, functional impairment type, and corresponding occupational exposure type; each suggestion stores the training action, training intensity, training frequency, safety precautions, and recommended retesting cycle; S6.2 Multi-dimensional matching and screening mechanism: Prioritize matching target risk areas, then limit the training intensity range according to the risk level, lock the training category in combination with the type of movement defect, and finally correct the training safety boundary based on FN / FP / DN / DP pain classification; DP type only matches painless low-intensity training, FP type adds manual assessment prompts, DN type matches targeted range of motion / stability training, and FN type outputs maintenance training. S6.3 Training Suggestion Priority Ranking: Construct a priority scoring model that integrates site risk scores, functional impairment scores, pain scores, occupational exposure scores, and suggestion matching degrees into a multi-dimensional weighted calculation, and outputs multiple graded training schemes according to the scores from high to low; the expression of the priority scoring model is:

[0018] in, This represents the priority score for matching the r-th body part with the k-th training suggestion; This represents the occupational risk score for the r-th body part; Indicates a site-specific functional impairment score; Indicates pain trigger score; Indicates occupational exposure score; This indicates the degree of matching between the k-th training recommendation and the location, defect type, risk level, and occupational exposure source. , , , , These are weight parameters; S6.4 Retest Dynamic Update Logic: Store historical multi-round assessment data, compare the risk scores of each part with the current round to obtain trend values; if the risk continues to rise, increase the priority of training recommendations for the corresponding part and shorten the retest cycle; if the risk decreases and the pain is relieved, reduce the training intensity, switch to maintenance training, and automatically update the whole set of intervention plans.

[0019] Furthermore, a system for occupational health risk assessment and training recommendations based on selective functional movement assessment is proposed, including a selective functional movement objectification assessment module, an occupational exposure interaction risk calculation module, and a personalized training recommendation generation and dynamic update module. The selective functional action objectification evaluation module includes a motion data acquisition unit, a data preprocessing unit, an action stage recognition unit, a motion feature extraction unit, a standard template storage unit, an action multidimensional scoring unit, and an FN / FP / DN / DP classification output unit. The occupational exposure interactive risk calculation module includes an action-body part mapping unit, an occupational exposure input and coding unit, an interactive item calculation unit, a body part occupational risk scoring unit, and a risk level classification output unit. The personalized training suggestion generation and dynamic update module includes a training suggestion library storage unit, a multi-condition matching and filtering unit, a suggestion priority sorting unit, a retest trend analysis unit, and a training scheme output update unit.

[0020] Preferably, the motion data acquisition unit supports the access of at least one of the following devices: inertial sensor, optical motion capture, video posture recognition, and plantar pressure sensor, and synchronously acquires kinematic time-series data and the subject's pain VAS score; The standard template storage unit pre-stores all exclusive template parameters for selective functional movements. The template parameters include standard joint angles, velocity thresholds, compensation judgment thresholds, and allowable thresholds for differences between left and right limbs. The occupational exposure recording and coding unit has built-in standardized assignment rules for five types of exposure items, namely duration, repetition frequency, load, posture, and vibration, ranging from level 0 to 3. The output of the training program output update unit includes: objective scores for each action, four-category results, functional defects by body part, occupational exposure scores, interaction amplification values, risk levels by body part, explanations of risk causes, graded training actions, intensity and frequency, safety tips, retesting cycle, and risk rise / fall warnings.

[0021] Compared with existing technologies, the health risk assessment method and system that integrates motion assessment and occupational exposure correction provided by this invention differ from existing technologies in that: 1. Structured identification input mechanism based on selective functional movement assessment results: Unlike existing identification schemes that mainly rely on questionnaires, experience judgments, or general physical tests, this invention transforms the results of selective functional movement assessments into a structured feature set that can be used for risk identification, thereby providing stronger motor function basis for occupational health risk identification.

[0022] 2. Joint identification mechanism of motor function abnormalities and job identification information at the site level: Unlike existing schemes that only assess the job or only assess the individual, this invention unifies the motion assessment results and job identification information into the same site-level identification framework, which can identify "which site is at higher risk and why".

[0023] 3. Layered identification output for different body parts: Unlike existing schemes that only output overall risk results, this invention can output the risk level and risk source for different parts such as the neck and shoulders, waist and back, hips, knees and ankles, improving the precision of the identification results.

[0024] 4. Dynamic risk trend identification mechanism: Unlike existing static screening methods, this invention introduces retesting comparison and historical change logic to achieve risk trend identification and early warning output.

[0025] 5. Movement training suggestion generation mechanism based on risk grading and source of functional impairment: Unlike existing schemes that only output risk values ​​or risk levels, this invention matches movement training suggestions from a training suggestion library based on the target body part, risk level, source of functional impairment, pain classification, and occupational exposure characteristics, and sorts the suggestions by priority scoring, so that the risk identification results can be further transformed into actionable training management suggestions.

[0026] Compared with existing technologies, this solution has the following advantages: (1) The recognition results are more standardized Because it uses structured coding of selective functional action assessment results, rather than relying solely on human experience, the recognition process has better repeatability and consistency.

[0027] (2) The identification results are more interpretable Because this invention retains the chain logic of "abnormal action - part mapping - job identification - risk level", it can clearly explain the source of risk instead of just giving a vague label.

[0028] (3) More suitable for occupational health scenarios Because this invention incorporates job identification information into the identification process, the output results are more closely aligned with actual occupational scenarios, which is beneficial for enterprises to conduct health screenings, identify key populations, and perform continuous monitoring.

[0029] (4) Easier to protect patents and implement them. Compared to occupational health assessment methods that only output risk levels, this invention not only completes objective scoring of selective functional movements, site-level risk identification, and occupational exposure interaction correction, but also generates training suggestions based on risk level, source of functional impairment, pain classification, and occupational exposure characteristics. This allows the assessment results to be transformed into actionable training management outputs. This technical approach forms a continuous closed loop of "assessment—risk interpretation—training suggestion generation—retesting and updating," resulting in better applicability and practicality. Attached Figure Description Figure 1 This is a schematic diagram of the health risk assessment method and system that integrates motion assessment and occupational exposure correction according to the present invention; Figure 2 This is a schematic diagram of the method flow of an embodiment of the present invention. Detailed Implementation

[0030] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0031] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0032] Example 1, please refer to Figures 1 to 2 As shown: To address the problems mentioned in the technical solutions, this application provides a health risk assessment method and system that integrates movement assessment and occupational exposure correction. The technical solution of this invention is divided into three main parts. The first part is selective functional movement assessment objectification, which is used to convert the subject's movement performance into calculable movement accuracy, movement completion, movement stability, objective functional scores, and FN, FP, DN, and DP classification results. The second part is occupational data encoding and occupational exposure interaction, which is used to convert data such as job duration, repetition frequency, external load, postural constraints, and tool or vibration exposure into site-level occupational exposure scores, and to form an interactive risk with site-level functional impairment scores. The third part is movement training suggestion matching and generation, which is used to match training categories, recommended movements, intensity prompts, precautions, and retest cycle suggestions from a training suggestion library based on risk level, target body part, source of functional impairment, pain classification, and source of occupational exposure.

[0033] This solution specifically includes: collecting kinematic data of selective functional movements, pain feedback data, and occupational exposure information; filtering, smoothing, outlier removal, and time synchronization of the kinematic data; identifying the initiation phase, active movement phase, terminal posture phase, and return phase of selective functional movements; extracting terminal joint angles, range of motion, peak angular velocity, peak angular acceleration, movement duration, left-right differences, compensation indices, and inter-trial fluctuation indices; comparing with a standard template of selective functional movements to calculate movement accuracy, completion, and stability scores; calculating objective functional scores for selective functional movements and outputting FN, FP, DN, or DP classification results; mapping the functional impairment scores of each selective functional movement to target body parts; encoding occupational exposure information into occupational exposure scores; calculating the interaction between functional impairment and occupational exposure and outputting occupational risk scores and risk levels; finally, generating movement training suggestions based on risk levels and defect sources, and dynamically updating risk levels and suggestion priorities after retesting. The overall process is as follows: Figure 1 As shown: The specific steps are as follows: Step 1: Objectification of Selective Functional Motor Assessment 1. Data Acquisition and Basic Kinematic Calculations The system collects joint angles, body segment postures, plantar pressure, or video posture key point data of the subject when performing a target selective functional movement. Data sources may include inertial sensors, motion capture systems, video posture recognition modules, plantar pressure sensors, or combinations of the above devices.

[0034] The angle of the i-th joint at time t is denoted as The joint angular velocity is then obtained by the angle difference between adjacent sampling times:

[0035] Joint angular acceleration is obtained by the difference in angular velocity between adjacent sampling times:

[0036] In the formula, For the first Each joint The joint angle at any moment, For the first Each joint angular velocity at time t, For the first Each joint Angular acceleration at time t, This represents the sampling time interval.

[0037] 2. Selective Functional Action Stage Recognition Since most selective functional movements are not continuous periodic movements, the system does not directly use the number of cycles in periodic exercises such as running as the core evaluation indicator. Instead, it divides the movement process into four stages:

[0038] When the angular velocity of a critical joint continuously exceeds the activation threshold, the system determines that the action has begun:

[0039] Furthermore, if the duration of this state exceeds the minimum start time Tmin, then t is identified as the start point of the action; where This is expressed as the angular velocity at the initial stage; When the angular velocity of a critical joint drops below the rest threshold and the angle of the critical joint is close to the peak angle of the current motion, the system determines it to be the final attitude point.

[0040] and:

[0041] in, Let ω be the angular velocity of the key joint k at time t. The threshold for initiating an action. The final attitude static threshold. This refers to the peak angle reached by the key joints during this movement. This represents the allowable error for the joint.

[0042] 3. Standard movement template for selective functional movements The system establishes a standard action template for each selective function action:

[0043] in, T m A standard template representing the m-th selective functional action; A m Indicates the standard joint angle range; V m Indicates the standard angular velocity or range of motion speed; C m Indicates the compensation threshold; L m Indicates the threshold difference between the left and right sides; S m This represents the stability threshold.

[0044] For example, a squatting motion template may include the end angle range of the hip, knee, and ankle joints, trunk forward tilt threshold, heel lift-off judgment, knee valgus threshold, and left and right lower limb angle difference threshold; a multi-segment flexion motion template may include trunk flexion angle, hip flexion angle, knee compensation degree, finger distance to the ground, and center of gravity shift degree; a multi-segment rotation motion template may include left and right trunk rotation angle, pelvic compensation angle, and left and right rotation difference threshold.

[0045] 4. Action accuracy scoring The system compares the subject's actual movement characteristics with a standard movement template, calculates a movement deviation score, and obtains a movement accuracy score based on this.

[0046] in, ACCm Score the accuracy of the m-th selective functional movement. DEVm Scoring is given for movement deviation.

[0047] The motion deviation score consists of angular deviation, speed deviation, compensatory deviation, and left-right difference deviation.

[0048] in, D angle,m Scoring for angle deviation, D velocity,m Scoring for speed deviation, D comp,m For the compensation bias score, Scoring for left-right difference deviation; , , , The weights for angular deviation, velocity deviation, compensatory deviation, and left-right difference deviation are respectively, and they satisfy the following:

[0049] Angle deviation score can be expressed as:

[0050] in, The final angle of the subject when completing the m-th selective functional action. This is the reference terminal angle in the standard motion template. This refers to the allowable angular deviation range. When... D angle,m If the value is greater than 100, it can be counted as 100.

[0051] The speed deviation score can be expressed as:

[0052] in, The peak angular velocity of the subject when completing the m-th action. This is the reference peak angular velocity in the standard template.

[0053] The compensation bias score can be expressed as:

[0054] in, For the score of the j-th compensation item in the m-th action, This is the weight of the compensation project.

[0055] The left-right difference deviation score can be expressed as:

[0056] in, and These are the indicators corresponding to the left and right sides, respectively. The threshold for allowing left-right differences.

[0057] 5. Performance score Motor completion scoring is used to determine whether a subject has met the terminal requirements of a selective functional motor action, whether the complete motor path has been completed, and whether any interruptions occurred during the action.

[0058] in, COMP m Score the completion rate of the m-th action. E m Achieve a score for the final posture. P m Score the completeness of the motion path. N m Scoring is given for actions that are not interrupted.

[0059] Final stance achievement score:

[0060] in, ROM actual,m The actual range of motion achieved by the subject. ROM ref,m This is the reference range of motion required by the standard template. When E m If the value is greater than 100, it is counted as 100.

[0061] The completeness score for the motion path is:

[0062] in, The number of action stages actually completed. The number of stages required for a standard action.

[0063] The score for uninterrupted movement is determined based on whether the movement is paused, adjusted, or stopped. Continuous completion is scored as 100 points, slight pauses as 75 points, significant pauses or readjustments as 50 points, and stopping due to pain, instability, or inability to continue as 0 points.

[0064] 6. Movement stability score Movement stability scores are used to reflect the consistency of the same selective functional movement across multiple trials, and whether there are significant fluctuations, instability, or compensatory changes during the movement:

[0065] in, STAB m To score the stability of the m-th action, STABDEV m The stability fluctuation score is given.

[0066] The stability fluctuation score is:

[0067] in, Score the terminal angle variation. Scoring for motion time variation, For compensatory fluctuation scoring; , , These are the angle-stabilized weights, time-stabilized weights, and compensatory-stabilized weights, respectively.

[0068] Terminal angle variation score:

[0069] The motion time variation score is:

[0070] in, SD The standard deviation of the final angle after multiple trials. This represents the average of the final angles from multiple trials. The standard deviation of the time to complete the action across multiple trials. This is the average time taken to complete the action across multiple trials.

[0071] 7. Selective functional motor objective function rating and FN / FP / DN / DP classification The system calculates the objective function score for selective function actions based on action accuracy score, action completion score, and action stability score:

[0072] in, Selective Function Actions_SCORE m The objective functional score for the m-th selective functional action.

[0073] The system further collects pain trigger information, and the pain trigger score is recorded as follows:

[0074] in, Visual analog scale (VAS) scores were assigned to the m-th action to elicit pain, ranging from 0 to 10.

[0075] Based on the objective function score of selective functional movements and pain triggering, the system outputs FN, FP, DN, or DP classification results:

[0076] in, T f The threshold for normal function is preferably 80 points, but it can also be calibrated based on expert-annotated samples, standard action libraries, or historical test data.

[0077] Step 2: Calculation of Business Exposure and Interaction Risks 1. Location-based functional impairment score The system further maps the individual selective functional movement scores to body parts, forming functional impairment scores for target body parts such as the neck and shoulders, lower back, hips, knees, and ankles.

[0078] The single-item selective motor function impairment score is:

[0079] The site-specific functional impairment score for the r-th body part is:

[0080] in, For the first Functional impairment scores for each body part For the first Selective functional actions and the first Mapping weights between body parts The number of selective function actions involved in the calculation.

[0081] For example, neck rotation and shoulder movement patterns can be primarily mapped to the neck and shoulders; multi-segment flexion, multi-segment extension, and multi-segment rotation can be primarily mapped to the lower back; squatting, single-leg standing, and lower limb-related movements can be primarily mapped to the hip, knee, and ankle.

[0082] 2. Occupational Exposure Score The system encodes job exposure information into five indicators: duration, repetition frequency, external load, postural constraints, and tool or vibration exposure.

[0083] in, Score the duration of exposure. For repeated exposure scoring, External load exposure score. For posture constraint exposure scoring, Rate the exposure to tools or vibration. The above indicators can be rated on a scale of 0 to 3, where 0 represents no exposure, 1 represents low exposure, 2 represents moderate exposure, and 3 represents high exposure.

[0084] The occupational exposure score for the r-th body part is:

[0085] in, , , , , The weights are respectively for duration, repetition frequency, external load, posture constraints, and tool or vibration exposure, and satisfy the following:

[0086] In a preferred embodiment, , , , , .

[0087] 3. Interaction between selective motor dysfunction and occupational exposure To reflect the occupational relevance of the same selective motor dysfunction under different job conditions, the system calculates the interaction between site-specific functional impairment scores and occupational exposure scores:

[0088] in, For the first The interaction score between selective motor dysfunction and occupational exposure for a specific body part. This interaction item is used to characterize that when a body part has both selective motor dysfunction and high occupational exposure, the occupational risk of that part is amplified; when the body part has selective motor dysfunction but low occupational exposure, its occupational risk is not significantly amplified.

[0089] 4. Occupation-related risk score and risk level The system integrates site-specific functional impairment scores, occupational exposure scores, interaction amplification scores, pain trigger scores, and retest trend scores to calculate an occupational-related risk score.

[0090] in, For the first Occupational risk scores for each body part Pain trigger score for selective functional movements related to this area. The retest trend score is given.

[0091]

[0092]

[0093] in, This represents the change in the risk score of the r-th body part relative to the previous period during the nth retest. Scoring the retest trend; This sets a preset risk increase reference threshold. No historical data is available for the initial assessment, so the trend score for subsequent assessments is recorded as 0. If the risk score increases compared to the previous period, the trend score is calculated by normalizing the increase; if the risk score remains unchanged or decreases, the trend score can be recorded as 0, and a separate risk stabilization or risk decrease indication will be output.

[0094] Risk levels are determined according to the following rules:

[0095] The above thresholds can be calibrated based on expert annotations, sample data, job type, or historical test results.

[0096] Step 3: Generate motion training suggestions based on risk grading 1. Construction of the training suggestion library After obtaining the occupational risk score and risk level for the r-th target body part, the system further matches corresponding movement training suggestions from a pre-set training suggestion library based on the site-level functional impairment score, occupational exposure score, interaction amplification score, pain trigger score, and selective functional movement classification results. This training suggestion generation process does not alter the aforementioned objective functional score and occupational risk score for selective functional movements; rather, it serves as an auxiliary output layer after risk identification, prompting assessors or subjects to select appropriate suggestions for improving activity levels, stability training, motor control training, compensatory correction, or occupational load management.

[0097] The k-th training suggestion in the training suggestion library can be represented as:

[0098] in, This represents the k-th training suggestion; Indicates the applicable risk level; Indicates the applicable body part; This indicates the applicable selective functional action classification results; Indicates the applicable type of functional impairment; This indicates recommended training exercises or training categories; Indicates training intensity; Indicates training frequency; This indicates important points to note.

[0099] 2. Suggested matching rules The system first selects training suggestions based on the target body part, then selects the training intensity based on the low-risk, medium-risk, or high-risk level, then determines the training category based on the source of functional impairment such as insufficient movement accuracy, insufficient movement completion, insufficient movement stability, significant left-right differences, or significant compensation, and finally makes safety corrections based on the FN, FP, DN, or DP classification results.

[0100] When the classification result is FN, the system can output maintenance training or routine training suggestions; when the classification result is DN, the system can output targeted suggestions for improving range of motion, stability training, or motor control training; when the classification result is FP, the system indicates that the functional performance is acceptable but there are pain triggers, and the training stimulation should be reduced and manual confirmation is recommended; when the classification result is DP, the system prioritizes outputting low-intensity training suggestions within the pain-free range and prompts for further manual evaluation if necessary.

[0101] 3. Suggested priority scoring The system ranks the training suggestions according to the following priority scores:

[0102] in, This represents the priority score for matching the r-th body part with the k-th training suggestion; This represents the occupational risk score for the r-th body part; Indicates a site-specific functional impairment score; Indicates pain trigger score; Indicates occupational exposure score; This indicates the degree of matching between the k-th training recommendation and the location, defect type, risk level, and occupational exposure source. , , , , These are the weight parameters. The system can output one or more training suggestions in descending order of priority.

[0103] 4. Training suggestion output content The system's output of movement training suggestions should include at least the training category, recommended movements, training intensity tips, training frequency, precautions, and retesting cycle recommendations. For example, when the neck and shoulder risk level is medium and the main sources of defects are limited neck rotation, limited shoulder mobility patterns, and compensatory shrugs, the system can match painless neck rotation exercises, thoracic spine extension and rotation exercises, and scapular retraction and depressor control exercises, and combine these with occupational load management suggestions for short periods of activity every 45 to 60 minutes, based on prolonged desk exposure.

[0104] 5. Updates on retesting trends and dynamic adjustments to recommendations After each retest, the system recalculates the occupational risk score for the target body part and calculates a dynamic trend value based on the current risk score and the trend risk score of the previous period. When the risk score increases twice consecutively, or increases by more than a preset percentage compared to the previous period, the system outputs a risk increase warning; when the risk score decreases twice consecutively, and the pain trigger score decreases simultaneously, the system outputs a risk decrease warning.

[0105] In implementations that include generating movement training suggestions, the system can also update the priority of output training suggestions based on changes in risk scores and movement scores after retesting; when the risk score of a certain part increases continuously, the priority of training suggestions for that part is increased and the retesting cycle is shortened; when the risk score decreases and the pain trigger score decreases simultaneously, the intensity of training suggestions is reduced or converted into maintenance training suggestions.

[0106] 6. Final output of the system The final output of the system should include at least the following: accuracy score, completion score, stability score, and objective function score for each selective functional movement; FN, FP, DN, or DP classification results for each selective functional movement; functional impairment scores for target body parts such as the neck and shoulders, lower back, hips, knees, and ankles; occupational exposure scores for each target body part; amplification scores for the interaction between functional impairment and occupational exposure; occupational risk scores and low, medium, and high risk levels for each target body part; explanations of the main sources of risk; retesting trend indicators and whether further manual assessment is recommended; and movement training suggestions, training intensity indicators, precautions, and retesting cycle recommendations that match the risk areas, sources of functional impairment, pain classification, and occupational exposure characteristics.

[0107] Example 2: Based on Embodiment 1, but with some differences, the health risk assessment method and system based on fusion of motion assessment and occupational exposure correction proposed in this invention will be described below with reference to specific examples and accompanying drawings. The specific content is as follows: like Figure 2As shown, taking long-term desk workers as an example, the system collects selective functional movement data such as neck rotation, shoulder movement patterns, multi-segment flexion, multi-segment extension, multi-segment rotation, squatting, and single-leg standing, and records pain feedback and job exposure information. After movement stage recognition and standard template comparison, the system obtained an objective function score of 68.31 for right neck rotation accompanied by pain triggering, and output DP classification; a shoulder movement pattern score of 74.50 for shoulder movement patterns accompanied by pain triggering, and output DP classification; and a squatting score of 85.40 for squatting without pain, and output FN classification.

[0108] The system further mapped the functional impairments in right cervical rotation and shoulder movement patterns to the neck and shoulder area, resulting in a neck and shoulder functional impairment score of 29.5. After encoding occupational exposure information such as prolonged desk work, frequent use of keyboard and mouse, and postural constraints, the system obtained an occupational exposure score of 65.0 for the neck and shoulder area, and calculated an interactive amplification score of 19.18. Based on this, the system outputs an occupational risk score of 33.1 for the neck and shoulder area, indicating a slightly high risk profile and recommending a retest.

[0109] In the training suggestion generation module, the system identified the subject's main problems as limited neck rotation, limited right shoulder movement, shoulder compensation, and prolonged desk posture constraints. Because the classification results included dynamic positioning (DP), the system prioritized matching training suggestions within a low-stimulation, pain-free range. These included neck rotation exercises within a pain-free range, thoracic spine extension and rotation exercises, scapular retraction and depressor control exercises, and short-term activity suggestions during work breaks. The system also suggested a follow-up assessment of neck rotation and shoulder movement patterns in two weeks. If the neck and shoulder risk score increased to medium risk after the follow-up assessment, the system increased the priority of the aforementioned training suggestions and prompted further manual evaluation.

[0110] Compared with traditional methods, this invention can additionally output quantitative scores, site risks, occupational exposure interaction items, training suggestions, and retest update results, as shown in Table 1;

[0111] This embodiment uses office workers who spend long hours at their desks as test subjects, and completes the entire process of "objective assessment of selective functional movements - calculation of occupational exposure interaction risks - matching of personalized training programs - dynamic updates through retesting". Through quantitative scoring, body part mapping, interaction risk calculation, hierarchical training matching, and dynamic retesting adjustment of the entire data output, it intuitively demonstrates the five core advantages of this technical solution: First, the evaluation is standardized, abandoning the traditional subjective judgment by humans, and relying on inertial sensing and template comparison to output objective action scores and FN / FP / DN / DP classifications, resulting in stronger consistency in retesting; Second, the risks are explainable. Through three-dimensional quantitative indicators of functional defects, occupational exposure, and interactive amplification, the dual sources of risk for a single body part can be clearly deconstructed, thus solving the problem of unclear risk causes in traditional solutions. Third, it adapts to occupational health scenarios, binding individual musculoskeletal function defects with occupational loads such as job duration, repetitive movements, and static postures to accurately distinguish the damage differences of the same movement defects under different working conditions. Fourth, the intervention plan is refined and stratified, setting safe training boundaries based on pain type. For people with abnormal pain in DP, only pain-free, low-intensity movements are pushed, which is different from the traditional general rest advice and is more feasible. Fifth, it has the ability to monitor dynamically and continuously, automatically adjust training priority and retest cycle based on retest trends, and provide early warnings of risk increases and decreases, breaking the limitations of traditional single static screening and building a closed loop for normalized dynamic health management of musculoskeletal injuries for enterprise employees.

[0112] Please refer to the above work process. Figures 1 to 2 .

[0113] It should be noted that the term "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0114] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A health risk assessment method integrating motion assessment and occupational exposure modification, characterized in that, include: S1. Objective processing of selective functional movement assessment: Collect subjects’ kinematic data and pain feedback data, complete the division of movement stages and extraction of movement features, combine standard movement templates to quantitatively calculate multidimensional movement scores and output the functional classification results of each movement; S2, Occupational Exposure Interaction Risk Calculation: The functional defects obtained in S1 are mapped to each body part to form a body part level functional impairment score. The occupational exposure information of the job is standardized and coded to obtain the body part occupational exposure score. The interaction items of the two are combined to calculate the occupational related risk score of each body part and classify the risk level. S3. Layered matching of training recommendations and dynamic updates: Based on the risk level of each body part, the type of functional impairment, the pain classification, and the source of occupational exposure, a training recommendation library is matched to generate personalized training programs. The risk level and the priority of training recommendations are dynamically adjusted in combination with retest data, forming a closed loop of assessment-risk grading-training intervention-retest update.

2. The health risk assessment method integrating motion assessment and occupational exposure modification according to claim 1, characterized in that, The S1 selective functional action assessment objectification process specifically includes: S2.1 Data Acquisition and Preprocessing: The joint angle time-series data of the subject performing selective functional movements are acquired through at least one of the following devices: inertial sensor, motion capture, and video posture recognition. The joint angular velocity and angular acceleration are obtained by differential operation. The raw kinematic data are filtered and smoothed, outlier removal is performed, and multi-source data are synchronized in time. The VAS visual analog scale score of pain induced by each movement of the subject is acquired simultaneously as pain feedback data. S2.2 Automatic four-stage motion recognition: Set the angular velocity start threshold, static threshold and joint peak angle error threshold to divide a single complete motion into four time intervals: start stage, main motion stage, final posture stage and return stage. S2.3 Multidimensional motion feature extraction: Extracting terminal joint angle, range of motion (ROM), peak angular velocity, peak angular acceleration, total motion duration, left and right limb index differences, compensatory motion features, and index fluctuation coefficients across multiple trials; S2.4 Standard Template Comparison and Multi-Dimensional Movement Scoring: Pre-store exclusive standard templates for each selective function movement. The templates include standard joint angle range, speed range, compensation threshold, left-right difference threshold, and stability threshold. Based on the templates, calculate the movement accuracy score, movement completion score, and movement stability score respectively. S2.5 Objective Functional Scoring and Four-Category Judgment: The objective functional score of a single action is obtained by weighted summation of the accuracy, completion and stability of the action; combined with the action pain VAS score, and with the preset functional threshold Tf as the boundary, the results are divided into four categories: FN painless and normal function, FP pain with normal function, DN painless and abnormal function, and DP pain with abnormal function.

3. The health risk assessment method integrating motion assessment and occupational exposure modification according to claim 2, characterized in that, The formula for calculating the accuracy score of the action is as follows: In the formula, It is represented as a movement deviation, which is obtained by superimposing the weighted angle deviation, speed deviation, compensation deviation, and left and right limb difference deviation. The sum of the weights of the four deviations is equal to 1. The motion completion rate is calculated by weighting three indicators: final pose achievement, motion path integrity, and uninterrupted motion. The motion stability calculation formula is: In the formula, It is represented as stability fluctuation, which is a weighted sum of the coefficients of variation of the final angle of multiple trials, the coefficient of variation of the action duration, and the coefficient of variation of the compensatory fluctuation.

4. The health risk assessment method integrating motion assessment and occupational exposure modification according to claim 1, characterized in that, The occupational exposure interaction risk calculation in S2 specifically includes: S4.1 Location-based functional impairment mapping: Establish a mapping weight matrix between single selective functional movements and body parts such as the neck and shoulders, lower back, hips, knees and ankles. A unified functional impairment score for each body part is obtained by weighted averaging of the functional deficit values ​​of single movements. S4.2 Occupational Exposure Quantification Coding: For each body part, five indicators, namely, job duration, frequency of repetitive movements, external load, static posture constraints, and tool vibration exposure, are assigned a 0-3 grade and weighted normalized to obtain the occupational exposure score for the body part. S4.3 Functional Defects - Occupational Exposure Interaction Amplification Calculation: The construction site interaction item is as follows: In the formula, The selective motor dysfunction and occupational exposure interaction score for the r-th body part is used to quantify the superimposed amplification effect of individual functional deficit and job load. For the first Functional impairment score for each body part; This is represented as the occupational exposure score for the r-th body part; S4.4 Occupational Risk Classification Calculation: The occupational risk score of the part is calculated by weighting and integrating the part functional impairment score, occupational exposure score, interactive amplification item, part pain score, and retest trend score. The score is then divided into three risk levels: low, medium, and high, based on the segment threshold, and the core risk sources of each part are marked.

5. The health risk assessment method integrating motion assessment and occupational exposure modification according to claim 4, characterized in that, The formula for scoring occupational-related risks for the aforementioned body part is: In the formula, Let r be the occupational risk score for the r-th body part. Pain trigger score for selective functional movements related to this area. The retest trend score is given.

6. The health risk assessment method integrating motion assessment and occupational exposure modification according to claim 1, characterized in that, The S3 hierarchical matching training suggestions and dynamic updates specifically include: S6.1 Pre-built training suggestion library: Each training suggestion in the library is bound to a matching tag, which includes the applicable body part, risk level, SFMA four-category type, functional impairment type, and corresponding occupational exposure type; each suggestion stores the training action, training intensity, training frequency, safety precautions, and recommended retesting cycle; S6.2 Multi-dimensional matching and screening mechanism: Prioritize matching target risk areas, then limit the training intensity range according to the risk level, lock the training category in combination with the type of movement defect, and finally correct the training safety boundary based on FN / FP / DN / DP pain classification; DP type only matches painless low-intensity training, FP type adds manual assessment prompts, DN type matches targeted range of motion / stability training, and FN type outputs maintenance training. S6.3 Training Suggestion Priority Ranking: Construct a priority scoring model that integrates site risk scores, functional impairment scores, pain scores, occupational exposure scores, and suggestion matching degrees into a multi-dimensional weighted calculation, and outputs multiple graded training schemes according to the scores from high to low; the expression of the priority scoring model is: in, This represents the priority score for matching the r-th body part with the k-th training suggestion; This represents the occupational risk score for the r-th body part; Indicates a site-specific functional impairment score; Indicates pain trigger score; Indicates occupational exposure score; This indicates the degree of matching between the k-th training recommendation and the location, defect type, risk level, and occupational exposure source. , , , , These are weight parameters; S6.4 Retest Dynamic Update Logic: Store historical multi-round assessment data, compare the risk scores of each part with the current round to obtain trend values; if the risk continues to rise, increase the priority of training recommendations for the corresponding part and shorten the retest cycle; if the risk decreases and the pain is relieved, reduce the training intensity, switch to maintenance training, and automatically update the whole set of intervention plans.

7. A system for occupational health risk assessment and training recommendation generation based on selective functional movement assessment, characterized in that, The method for performing any one of claims 1 to 6 includes a selective functional movement objectification assessment module, an occupational exposure interaction risk calculation module, and a personalized training suggestion generation and dynamic update module. The selective functional action objectification evaluation module includes a motion data acquisition unit, a data preprocessing unit, an action stage recognition unit, a motion feature extraction unit, a standard template storage unit, an action multidimensional scoring unit, and an FN / FP / DN / DP classification output unit. The occupational exposure interactive risk calculation module includes an action-body part mapping unit, an occupational exposure input and coding unit, an interactive item calculation unit, a body part occupational risk scoring unit, and a risk level classification output unit. The personalized training suggestion generation and dynamic update module includes a training suggestion library storage unit, a multi-condition matching and filtering unit, a suggestion priority sorting unit, a retest trend analysis unit, and a training scheme output update unit.

8. The occupational health risk assessment and training suggestion generation system based on selective functional movement assessment according to claim 7, characterized in that, The motion data acquisition unit supports the connection of at least one of the following devices: inertial sensor, optical motion capture, video posture recognition, and plantar pressure sensor, and synchronously acquires kinematic time-series data and the subject's pain VAS score; The standard template storage unit pre-stores all exclusive template parameters for selective functional movements. The template parameters include standard joint angles, velocity thresholds, compensation judgment thresholds, and allowable thresholds for differences between left and right limbs. The occupational exposure recording and coding unit has built-in standardized assignment rules for five types of exposure items, namely duration, repetition frequency, load, posture, and vibration, ranging from level 0 to 3. The output of the training program output update unit includes: objective scores for each action, four-category results, functional defects by body part, occupational exposure scores, interaction amplification values, risk levels by body part, explanations of risk causes, graded training actions, intensity and frequency, safety tips, retesting cycle, and risk rise / fall warnings.

Images