Method, system and equipment for monitoring dyskinesia risk of old people and medium

Abstract

The invention provides an old people dyskinesia risk monitoring method, system and device and a medium, basic information of old people is acquired, a compliance evaluation model is established, and training parameters are adjusted based on a medical care manual intervention interface; identifying a high-risk factor combination by using an association rule, analyzing and predicting a movement function change trend of a future preset time length by combining a synchronous time sequence, and optimizing a decision weight by using a reinforcement learning module according to a medical care manual adjustment record; a dynamic risk assessment model is constructed by using a Bayesian network, cross-mechanism data cooperative training is realized by using a federal learning framework, time sequence features are extracted through feature engineering, and motion function indexes and risks of old people in a period of time in the future are predicted. And comprehensive evaluation, intelligent early warning and personalized intervention on the dyskinesia of the old people are realized. Rehabilitation training execution conditions of old people are monitored in real time, and rehabilitation prescriptions are dynamically adjusted according to execution data and subjective feedback, so that effectiveness of intervention is ensured.

Description

Technical Field

[0001] The present application relates to the technical field of movement monitoring for the elderly, and in particular to a method, system, device and medium for monitoring the risk of movement dysfunction in the elderly. Background Art

[0002] Currently, technologies used to predict fall risk or motor dysfunction in the elderly primarily rely on single-dimensional assessments (such as gait analysis or muscle strength testing). However, these methods often overlook other key factors, such as muscle tone and fine motor skills, resulting in incomplete and inaccurate risk assessments.

[0003] Related technologies often use static assessment models that are unable to dynamically adjust based on the elderly's real-time and historical data. Consequently, the models' predictive accuracy and generalization capabilities are limited, making them difficult to adapt to individual differences. Regarding personalized intervention, these technologies lack process data support, resulting in rehabilitation plans relying primarily on the experience of medical staff, making it difficult to achieve precise individualized matching. These technologies are also unable to effectively track the elderly's rehabilitation training progress, making it difficult to adjust rehabilitation plans in a timely manner, resulting in poor intervention results. Summary of the Invention

[0004] The present invention provides a method for monitoring the risk of motor dysfunction in the elderly. The method uses a multi-dimensional, quantitative assessment method combined with a machine learning algorithm to perform risk stratification, and provides intelligent early warning and rehabilitation plans based on risk levels and personalized needs, providing safer and more efficient motor dysfunction risk management services for the elderly.

[0005] Methods include: S101: Obtain basic information of the elderly and simultaneously carry out physical assessment and testing of the elderly; S102: Construct a logistic regression model to calculate the fall risk probability value and motor dysfunction index, and divide the fall risk and motor dysfunction risk into three levels. Then establish an early warning coding system to obtain risk assessment results; S103: Determine the fall risk probability or motor dysfunction index based on the early warning coding system, and push an early warning message when it exceeds the threshold; S104: Based on the risk assessment results, a standardized rehabilitation prescription is generated by matching the knowledge base of balance muscle training programs through a preset decision tree system; S105: Real-time tracking of rehabilitation training execution, establishment of a compliance evaluation model, and adjustment of training parameters based on the medical and nursing manual intervention interface; S106: Based on the data generated from steps S101 to S105, association rules are used to identify high-risk factor combinations, and synchronous time series analysis is combined to predict the trend of motor function changes over a preset period of time in the future. The reinforcement learning module is used to optimize the decision weight based on the manual adjustment records of the medical staff; S107: Trigger the re-evaluation process according to the preset period, repeat the multi-dimensional evaluation from step S101 to step S102, calculate the improvement rate of key indicators, and obtain multi-dimensional evaluation data; S108: Based on multi-dimensional assessment data, a dynamic risk assessment model is constructed using a Bayesian network. A federated learning framework is used to achieve cross-institutional data collaborative training. Time series features are extracted through feature engineering to predict the motor function indicators and risks of the elderly in the future.

[0006] It should be further explained that step S102 specifically includes: Step S1021: Standardize the basic information, integrate the balance sub-item and gait sub-item of the Tinetti scale, calculate the weighted sum according to the balance weight and gait weight, and generate a composite input vector containing demographic characteristics and functional assessment; Step S1022: constructing a fall risk prediction channel based on the composite input vector using a logistic regression algorithm, optimizing the penalty using cross-validation, and outputting a fall risk probability value and classification result; Step S1023: Based on the muscle volume, movement speed, and modified Barthel index features in the generated composite input vector, a support vector machine algorithm is used to construct a movement dysfunction prediction channel. A radial basis kernel function is used to process nonlinear relationships. Grid search is performed to optimize the parameter combination and output a three-level risk level mapped by a Softmax function. Step S1024: Cartesian product is performed on the fall risk probability value and the three-level risk level output in step S1023 to form multiple risk types, and the risk evolution slope is calculated in combination with time series analysis; Step S1025: Encapsulate the fall risk probability value, three-level risk level and trend prediction results into a JSON format data packet, push it to the electronic device, and write the risk grading results into the personal health record database.

[0007] It should be further explained that in the basic information of the method, age was normalized to the range of 0–1, gender was converted to one-hot encoding, BMI retained its original value, and fall history was marked as a binary categorical variable; The balance and gait subitems of the Tinetti scale were integrated, and a weighted sum was calculated with a weight of 0.6 for balance and 0.4 for gait to generate a composite input vector containing demographic characteristics and functional assessments.

[0008] It should be further noted that, based on the muscle volume, movement speed and modified Barthel index features in the composite input vector generated in step S1021, a support vector machine algorithm is used to construct a movement dysfunction prediction channel; The radial basis kernel function is used to deal with nonlinear relationships, and the penalty parameter C∈[0.1,10] and the kernel coefficient γ∈[0.01,1] are defined. The Softmax function is used to map the risk level into three levels. Among them, low risk: 0-0.6, medium risk: 0.6-0.8, high risk: ≥0.8, realizing the quantitative grading of movement dysfunction.

[0009] It should be further explained that step S103 specifically includes: filtering out data that meets the trigger conditions based on the risk assessment results and the condition judgment module; wherein the trigger conditions are that the fall risk probability is greater than 0.7 or the motor dysfunction index is greater than 0.8.

[0010] It should be further explained that step S107 specifically includes: The method for calculating the improvement rate of key indicators is: Key indicator improvement rate = (current value - benchmark value corresponding to basic information) / benchmark value corresponding to basic information; Visualize the improvement rate of key indicators of basic information in the form of a radar chart; The method also builds a dynamic risk assessment model based on the historical accumulation of assessment data and combines it with a Bayesian network. It also uses a federated learning framework to achieve cross-institutional data collaborative training, ensuring data privacy while improving the model's generalization ability.

[0011] It should be further explained that in step S106, the data generated by steps S101 to S105 are used to perform association rule mining using the Apriori algorithm, wherein high-risk combinations with support ≥ 15% and confidence ≥ 80% are screened based on historical assessment data, treatment records, and outcome labels, and an interpretable decision rule library is generated; Predict motor function changes over a preset time period in the future based on a time series analysis model: Adjust the parameter combination of rehabilitation prescription based on the current status of the elderly, historical intervention records, and adjustment behavior of medical staff; The adjusted parameters are combined using a federated learning framework to aggregate multi-institutional data, retrain the association rule base and time series prediction model; generate rehabilitation prescription strategies and push them to electronic devices.

[0012] The present application also provides a system for monitoring the risk of motor dysfunction in the elderly, the system comprising: The information assessment module is used to obtain basic information about the elderly and simultaneously conduct physical assessments and tests on them; The graded warning module is used to build a logistic regression model to calculate the fall risk probability value and motor dysfunction index, and divide the fall risk and motor dysfunction risk into three levels, and then establish a warning coding system; The risk assessment module is used to determine the probability of falling risk or the motor dysfunction index based on the early warning coding system, and push warning information when the threshold is exceeded; The prescription generation module is used to generate standardized rehabilitation prescriptions based on the risk assessment results and the knowledge base of balance muscle training programs through a preset decision tree system; Parameter adjustment module, used to track the execution of rehabilitation training in real time, establish a compliance evaluation model, and adjust training parameters based on the medical and nursing manual intervention interface; The optimization processing module is used to identify high-risk factor combinations using association rules, and predict the trend of motor function changes for a preset time period in the future through synchronous time series analysis. The reinforcement learning module is used to optimize decision weights based on manual adjustments made by medical staff; The data evaluation module is used to trigger the re-evaluation process according to the preset cycle, obtain multi-dimensional evaluation, calculate the improvement rate of key indicators, and obtain multi-dimensional evaluation data; The prediction processing module is used to use multi-dimensional evaluation data, apply Bayesian networks to build a dynamic risk assessment model, adopt a federated learning framework to achieve cross-institutional data collaborative training, extract time series features through feature engineering, and predict the motor function indicators and risks of the elderly in the future.

[0013] According to another embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for monitoring the risk of motor dysfunction in the elderly when executing the program.

[0014] According to another embodiment of the present application, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for monitoring the risk of motor dysfunction in the elderly are implemented.

[0015] It can be seen from the above technical solutions that the present invention has the following advantages: The method for monitoring motor dysfunction risk in the elderly, provided in this application, enables comprehensive assessment, intelligent early warning, and personalized intervention for motor dysfunction in the elderly. By collecting basic information about the elderly and conducting a multidimensional motor function assessment, the system uses logistic regression and support vector machine models to calculate fall risk and motor dysfunction risk scores, categorizing the elderly into low, medium, and high risk levels. The system then pushes instant and predictive warnings via a mobile app, alerting the elderly and their families to the risks and enabling them to take appropriate measures. Simultaneously, the system generates personalized rehabilitation training plans based on the risk level and motor function score, and displays them to the user through the app. The system also monitors the elderly's rehabilitation training progress in real time and dynamically adjusts the rehabilitation prescription based on performance data and subjective feedback to ensure the effectiveness of the intervention. Furthermore, the system optimizes the risk assessment model through big data analysis and machine learning techniques, improving prediction accuracy and generalization. It uses a time series analysis model to predict the elderly's future motor function indicators and risk level, and adjusts rehabilitation strategies accordingly. All data is encrypted and stored, and data query and analysis functions are provided to ensure data security and accessibility, enhancing the effectiveness of risk warning and intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for the description. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 Flowchart of the method for monitoring the risk of motor dysfunction in the elderly; Figure 2 Schematic diagram of an electronic device. DETAILED DESCRIPTION

[0018] The method for monitoring the risk of motor dysfunction in the elderly, provided in this application, addresses the problem that existing technologies, which primarily rely on periodic examinations, are unable to achieve real-time monitoring of motor dysfunction in the elderly. This invention continuously tracks the motor function status of the elderly through real-time monitoring technology, dynamically adjusting assessment results and early warning strategies based on individual health changes. This makes health management for the elderly more flexible and precise, enabling effective intervention in the early stages of motor dysfunction.

[0019] The specific steps of the method for monitoring the risk of motor dysfunction in the elderly will be described in detail below. For the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are provided to facilitate a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details.

[0020] It should be understood that when used in this specification, the term "comprising" indicates the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their collections. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0021] The phrases "one embodiment" or "some embodiments" described in this application mean that the specific features, structures, or characteristics described in the embodiment are included in one or more embodiments of the application. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in other embodiments," etc. that appear in different places in this application do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized.

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] See also Figure 1 The figure is a flow chart of a method for monitoring the risk of motor dysfunction in the elderly in a specific embodiment, the method comprising: S101: Obtain basic information of the elderly.

[0024] Specifically, this embodiment collects basic information of the elderly through a human-computer interaction interface, such as name, gender, date of birth, height, weight, medical history files, such as history of diabetes / hypertension / stroke, medication records, and simultaneously implements tremor assessment, muscle strength measurement, muscle tension classification, muscle volume detection, movement speed test, fine motor assessment, joint range of motion measurement and comprehensive evaluation of balance and gait. The comprehensive evaluation of balance and gait can be the balance or gait sub-item score of the Tinetti scale, or the modified Barthel index.

[0025] S102: Based on the basic information, a logistic regression model is constructed to calculate the fall risk probability value and the motor dysfunction index, and the fall risk and motor dysfunction risk are divided into three levels. Then, an early warning coding system is established to obtain the risk assessment results.

[0026] In an embodiment of the present invention, based on step S102, a possible embodiment will be given below to illustrate its specific implementation scheme in a non-limiting manner.

[0027] S1021: Standardize the basic information, integrate the balance and gait sub-items of the Tinetti scale, calculate the weighted sum based on the balance weight and gait weight, and generate a composite input vector that includes demographic characteristics and functional assessment.

[0028] In this example, age was normalized to the range 0–1, sex was converted to a one-hot encoding, BMI was retained as its original value, and fall history was labeled as a binary categorical variable. The balance and gait sub-items of the Tinetti scale were also integrated, with a weighted sum calculated using a balance weight of 0.6 and a gait weight of 0.4. This generated a composite input vector containing demographic characteristics and functional assessments, which served as the basis for subsequent model training.

[0029] S1022: Based on the composite input vector, a logistic regression algorithm is used to construct a fall risk prediction channel, and cross-validation is used to optimize the penalty to output the fall risk probability value and classification result.

[0030] This embodiment uses a logistic regression algorithm to construct a fall risk prediction channel. Input features include age, gender, BMI, fall history, and Tinetti composite score. The linear combination results are mapped to probability values ​​in the range of 0-1 using the Sigmoid function. 0.7 is set as the critical threshold, and values ​​≥0.7 are considered high risk. Regularization terms are introduced during model training to control complexity, and cross-validation is used to optimize the penalty factor C. The final output is the fall risk probability value and classification result.

[0031] S1023: Based on the muscle volume, movement speed and modified Barthel index features in the generated composite input vector, a support vector machine algorithm is used to construct a movement dysfunction prediction channel. The radial basis kernel function is used to process nonlinear relationships. The parameter combination is optimized through grid search to output the three-level risk level mapped by the Softmax function.

[0032] In step S1023, a support vector machine algorithm is used to construct a motor dysfunction prediction channel based on the muscle volume, movement velocity, and modified Barthel index features in the composite input vector generated in step S1021. A radial basis kernel function is used to address nonlinear relationships, and a grid search is used to optimize parameter combinations, with a penalty parameter C∈[0.1,10] and a kernel coefficient γ∈[0.01,1]. The output values ​​are mapped to three risk levels using a softmax function: low risk (0-0.6), medium risk (0.6-0.8), and high risk (≥0.8), achieving a quantitative grading of motor dysfunction.

[0033] S1024: Cartesian product is performed on the fall risk probability value and the three-level risk level output in step S1023 to form multiple risk types, and the risk evolution slope is calculated in combination with time series analysis; This embodiment combines the fall risk level output from step S1022 with the motor dysfunction level output from step S1023 through a Cartesian product to form a risk type, such as high-risk fall + moderate motor dysfunction. This is combined with time series analysis to calculate the risk evolution slope. If the slope is greater than 0.15 / week, a rapid increase flag is added. Ultimately, a structured warning code is generated that includes risk type, level, and trend prediction, supporting multi-dimensional risk assessment.

[0034] S1025: Encapsulate the fall risk probability value, three-level risk level and trend prediction results into a JSON format data packet, push it to the electronic device, and write the risk grading results into the personal health record database.

[0035] This embodiment encapsulates the fall risk probability value from step S1022, the functional impairment level from step S1023, and the trend prediction results from step S1024 into a JSON-formatted data packet and pushes it to the early warning push module via the API (step S103). The risk grading results are also written into the personal health record database, serving as the key decision-making basis for generating rehabilitation prescriptions in step S104, achieving a seamless transition from risk assessment to intervention strategy.

[0036] In this embodiment, a correlation relationship is constructed between a motor dysfunction prediction channel, a grid search optimization parameter combination, and Softmax mapping to three risk levels. The three form a closed loop of model training-parameter optimization-result quantification. Among them, the original prediction value can be generated by the SVM model based on muscle volume, movement speed, and MBI score. The SVM here directly affects the generalization ability of the model. For example, a C value that is too large will lead to overfitting, and a C value that is too small will lead to underfitting. The grid search determines the optimal parameter combination by traversing the preset parameter range, C∈[0.1,10], γ∈[0.01,1], combined with the cross-validation error minimization principle. The original score output by the SVM needs to be converted into a risk level, that is, low risk / medium risk / high risk. The Softmax function maps the score to a probability distribution, and then divides it by a threshold, 0-0.6=low risk, 0.6-0.8=medium risk, ≥0.8=high risk. In this way, the parameter combination optimized by grid search determines the predictive ability of the SVM model, while Softmax converts the model output into an interpretable risk level. The three together achieve a reliable mapping from feature input to classification results.

[0037] S103: Determine the fall risk probability or motor dysfunction index based on the early warning coding system, and push an early warning message when it exceeds the threshold.

[0038] In this embodiment, when the probability of falling risk is greater than 0.7, or the motor dysfunction index is greater than 0.8, an immediate warning information including a risk heat map is pushed through the mobile terminal, and the risk evolution trend in the next 30 days is predicted based on time series analysis. Preventive recommendations are sent 72 hours in advance, and all warning data are stored in the system database in the form of text descriptions combined with three-dimensional visual charts.

[0039] S104: Based on the risk assessment results, a preset decision tree system is used to match the knowledge base of balance muscle strength training programs to generate a standardized rehabilitation prescription.

[0040] For this embodiment, a knowledge base of balance / muscle strength and other training programs is matched through a preset decision tree system to automatically generate a standardized rehabilitation prescription containing training methods, cycles, equipment parameters, intensity, duration and difficulty coefficient, and store it in the health file of the elderly.

[0041] S105: Track the execution of rehabilitation training in real time, establish a compliance evaluation model, and adjust training parameters based on the medical and nursing manual intervention interface.

[0042] In this embodiment, wearable devices such as sports bracelets and smart sportswear can be equipped for the elderly. During rehabilitation training, the wearable devices use built-in sensors to capture their motion data in real time, including the type, amplitude, and frequency of training movements, while also recording the duration and number of repetitions of training.

[0043] Based on the collected data, a rehabilitation training compliance evaluation model was constructed. The model calculated the completion degree of rehabilitation training for the elderly according to the formula: completion degree = actual implementation / prescription requirements.

[0044] For example, if a rehabilitation prescription requires 30 minutes of balance training per day, and the wearable device records 20 minutes of actual training, then the completion rate of balance training for that day is 20 ÷ 30 × 100% ≈ 66.7%. The system regularly calculates the completion rate of each rehabilitation training session and records the results.

[0045] If an elderly person's training completion rate falls short of the target, the system automatically triggers a de-escalation mechanism, reducing the training intensity by 10%. For example, if the original training intensity was 80% load, the mechanism will be triggered to adjust it to 72%. Simultaneously, the system generates a reminder message at the medical intervention interface, informing medical staff that an elderly person's training has not met the target and that the system has automatically adjusted the parameters. Medical staff can then manually adjust the training parameters based on their judgment, such as increasing or decreasing the intensity, changing the training method, or changing the training period.

[0046] S106: Based on the data generated from step S101 to step S105, association rules are used to identify high-risk factor combinations, and synchronous time series analysis is combined to predict the trend of changes in motor function for a preset time period in the future. The reinforcement learning module is used to optimize the decision weight based on the manual adjustment records of medical staff.

[0047] In this embodiment, step S106 includes the following specific methods: Step S1061: Combining parameters such as age, gender, BMI, muscle strength, balance ability, as well as low / medium / high risk and rehabilitation training parameters, a structured feature library is generated by interpolating missing values ​​and correcting outliers.

[0048] Step S1062: Use the Apriori algorithm to mine association rules on the feature library of step S1061.

[0049] Specifically, based on historical assessment data, such as muscle volume <11 points and fall risk >0.7, combined with treatment records and outcome labels, we screen for high-risk combinations with support ≥15% and confidence ≥80%, such as increased muscle tone + poor balance → a rapid increase in fall risk. This decision rule base is then generated. This rule base is regularly expanded based on the results of federated learning.

[0050] Step S1063: Predicting the motor function changes of a preset time period in the future based on the time series analysis model, using stationary data or nonlinear data to process time series features.

[0051] Obtain historical functional muscle strength, gait, training intensity, frequency and activity parameters, and generate functional change curves and risk warning labels for future preset time periods.

[0052] A reinforcement learning framework was built to optimize the decision weights for rehabilitation prescriptions. Medical staff then adjusted the parameters of the rehabilitation prescription based on the patient's current status and historical intervention records. A federated learning framework was used to aggregate data from multiple institutions and retrain the association rule base and time series prediction model. A new rehabilitation prescription strategy was generated and pushed to electronic devices, such as those used by doctors or the elderly patient's family members.

[0053] In this way, the integration of full-cycle data provides a complete feature basis for subsequent analysis, making it easier for medical staff to obtain relevant data, optimize decisions in real time through medical feedback, and solve the problem that static models cannot adapt to individual differences.

[0054] S107: Trigger the re-evaluation process according to the preset cycle, repeat the multi-dimensional evaluation from step S101 to step S102, calculate the improvement rate of key indicators, and obtain multi-dimensional evaluation data.

[0055] In this embodiment, a system timer can be used to trigger the reassessment process at a preset interval, accessing the basic information in step 101 and performing the multi-dimensional assessment in step 102. Assessment data is obtained. Based on the reassessment data and the initial baseline data, a difference calculation is performed on muscle strength scores, balance scores, gait scores, and other scores.

[0056] Key indicator improvement rate = (current value - benchmark value corresponding to basic information) / benchmark value corresponding to basic information; In this embodiment, the 3σ principle can be used to eliminate extreme values ​​caused by measurement errors to ensure the reliability of the improvement rate calculation. The improvement rate data of muscle strength, muscle tone, movement speed, fine motor skills, joint mobility, balance ability, and daily living ability are mapped to the radar chart coordinate system: different weights are assigned according to clinical importance, and the comprehensive improvement rate of each dimension is calculated by weighted average.

[0057] This embodiment builds a dynamic risk assessment model based on the historical accumulation of assessment data and combines it with a Bayesian network. It also uses a federated learning framework to achieve cross-institutional data collaborative training, ensuring data privacy while improving model generalization capabilities.

[0058] S108: Based on multi-dimensional assessment data, a dynamic risk assessment model is constructed using a Bayesian network. A federated learning framework is used to achieve cross-institutional data collaborative training. Time series features are extracted through feature engineering to predict the motor function indicators and risks of the elderly in the future.

[0059] This embodiment uses the Bayesian network to construct a dynamic risk assessment model based on the multiple evaluation data of step 107, adopts the federated learning framework to realize cross-institutional data collaborative training, extracts time series features through feature engineering, and optimizes the model parameter structure based on the correction records of medical staff, so as to realize the autonomous evolution and upgrade of the system prediction accuracy and intervention strategy.

[0060] To further enhance the reliability of the method for monitoring the risk of motor dysfunction in the elderly provided in the above embodiments, the following is a specific implementation method. In one embodiment, basic information about the elderly is collected, including their name, gender, date of birth, height, weight, and medical history, such as diabetes, hypertension, or stroke. Furthermore, the names and dosages of medications currently being taken, as well as their educational level, are recorded. This information can be manually entered by medical staff or the elderly themselves through the system's user interface. Assessment details are as follows: Objective / Subjective Assessment: Tremor: The tremor component of the Unified Parkinson's Disease Rating Scale (UPDRS) is used for assessment, recording the severity and distribution of tremor (e.g., resting tremor, postural tremor). The score is scored on a scale of 0 to 4 (0: no tremor, 4: severe tremor).

[0061] Muscle strength: Assess the strength of the major muscle groups using the Manual Muscle Measurement (MMT) criteria and enter a score between 1 and 5.

[0062] Recording method: 1 to 5 points (1: no contraction, 5: normal muscle strength).

[0063] Muscle tone: Muscle tone was assessed using the modified Ashworth Scale (MAS), which records the degree of muscle tension.

[0064] Recording method: 0 to 4 points (0: no increased muscle tone, 4: severe increased muscle tone).

[0065] Muscle mass: Muscle mass was assessed using the SARC-CalF method.

[0066] Recording method: A SARC-CalF total score ≥11 points was considered screening positive.

[0067] Motor speed: Motor speed can be assessed using a timed walking test (e.g., 10-meter walk test) or a repetitive motion test (e.g., finger tapping test).

[0068] Recording method: record walking time (seconds) or number of repetitions (times / minute).

[0069] Fine motor skills: assessed using the finger-to-nose test.

[0070] Recording method: 0-4 copies, 0 points for normal, 4 points for completely irregular drawing or incomplete.

[0071] Joint motion: Use limited range of motion assessment.

[0072] Recording method: Through joint activities, such as shoulder flexion and shoulder extension, record whether there is a restricted joint, 0 points for no restriction, 1 point for restriction.

[0073] Tinetti Balance and Gait Scale: A comprehensive assessment of the balance and gait of the elderly, recording various scores of balance and gait.

[0074] Recording method: Record the balance score and gait score (total score 28 points, 16 points and 12 points respectively). A score <24 points indicates balance dysfunction.

[0075] Modified Barthel Index (MBI): assesses the ability of elderly people to perform activities of daily living.

[0076] Recording method: Record the total MBI score (0-100 points, 0 is completely dependent, 100 is completely independent).

[0077] After the assessment is completed, all the collected information will be recorded and stored in the system, providing basic data for subsequent assessment, intervention, and prescription.

[0078] After collecting basic information and conducting subjective and objective assessments, the system focuses on assessing motor function. During this stage, healthcare professionals assess the elderly individual's muscle strength using the manual muscle testing (MMT) criteria provided by the system, entering a score between 1 and 5. The system also records results from tremor assessments, muscle tone assessments, muscle bulk assessments, motor speed assessments, fine motor skills assessments, joint range of motion assessments, the Tinetti Balance and Gait Scale, and the Modified Barthel Index (MBI). The system analyzes these inputs and, incorporating information such as the individual's age, gender, BMI, and fall history, calculates a fall risk score using a pre-trained logistic regression model and a motor dysfunction score using a support vector machine model. These scores are used to categorize elderly individuals into different risk levels—for example, low, moderate, and high fall risk, and low, moderate, and high motor dysfunction risk. This allows the system to more accurately identify elderly individuals requiring priority attention.

[0079] Based on a comprehensive assessment of the elderly's motor function and risks, the system provides intelligent early warnings. If the calculated fall risk score exceeds 0.7 or the motor dysfunction score exceeds 0.8, the system will immediately send a push notification via the mobile app to alert the elderly and their families. Furthermore, the system provides predictive early warnings. If the analysis indicates an increasing risk of falls or motor dysfunction in the future, the system will also send a notification in advance, advising the elderly to increase their exercise.

[0080] Warning types and methods: These include immediate and predictive warnings, delivered as system messages via the app. Message content includes information such as risk type, risk level, and recommendations. The presentation method involves displaying the warning message content in the app using a combination of text descriptions and charts. The generated warning message content is stored in the system database and associated with Zhang San's personal information.

[0081] In addition to providing early warnings, this embodiment also generates personalized rehabilitation prescriptions. Based on the elderly individual's risk level, motor function score (including muscle strength, muscle tone, muscle mass, movement speed, fine motor skills, joint mobility, Tinetti Balance and Gait Scale, Modified Barthel Index (MBI) assessment results), and medical history, the system matches the appropriate rehabilitation training program. For example, for elderly individuals at high risk of falls, the system recommends balance training and muscle strength training; for those at high risk of motor dysfunction, the system recommends muscle strength training and joint mobility training, and develops a corresponding relaxation training plan based on muscle tone.

[0082] Rehabilitation prescription information includes rehabilitation methods, rehabilitation cycle, rehabilitation equipment, rehabilitation frequency, treatment duration, training difficulty and other information.

[0083] Sample rehabilitation prescription content: For elderly people with decreased muscle strength: resistance training is recommended. For elderly people with increased muscle tone: relaxation training is recommended.

[0084] For elderly people with balance dysfunction: balance training is recommended.

[0085] After issuing a rehabilitation prescription, the system will track whether the elderly person is carrying out the intervention as planned. The system will record whether the elderly person performs balance training, strength training, joint mobility training, etc., as well as the duration and number of repetitions of each training session. These execution status will be displayed in the form of a chart on the app for the elderly person to view. If the system finds that the elderly person is not executing according to the plan, a message reminder will be pushed. The elderly person can also use the app to provide feedback on their subjective feelings, such as fatigue level and post-exercise feeling. Based on this execution data and subjective feedback, the system will dynamically adjust the rehabilitation prescription. For example, if the elderly person does not complete the exercise plan for three consecutive days, the system will automatically reduce the intensity or duration of the exercise by 10%. Medical staff can also manually adjust the rehabilitation prescription based on the elderly person's assessment results and subjective feedback, such as increasing or decreasing the intensity of exercise or adjusting the type of exercise.

[0086] Sample data: Execution record: May 17: Completed 40 minutes of balance and strength training. May 18: Did not train as planned. May 19: Completed 45 minutes of balance and strength training.

[0087] Subjective feedback: May 17: Pain level 1, fatigue level 3. May 19: Pain level 0, fatigue level 2.

[0088] System automatic adjustment: Since the training was not completed on May 18, the system automatically reduced the training time by 10%, from 45 minutes to 40 minutes.

[0089] Manual adjustment: Medical staff can manually adjust rehabilitation prescriptions based on the elderly’s feedback and evaluation results, for example, increasing the difficulty of training or adjusting the type of exercise.

[0090] During system operation, a large amount of motor function data for older adults will be accumulated. To better utilize this data, the system will perform multi-level data processing and model optimization. First, the system uses association analysis algorithms to identify factors closely associated with fall risk and motor dysfunction. For example, the system can analyze the correlation between indicators such as gait symmetry, balance ability, and muscle strength scores and fall risk, generating association rules. Next, the system uses cluster analysis algorithms to categorize older adults into groups with different health characteristics. For example, based on differences in muscle strength and balance ability, older adults can be divided into "high-risk fall group" and "moderate motor dysfunction group," providing a basis for personalized intervention. Furthermore, the system uses time series analysis algorithms to analyze trends in older adults' motor function over time and predict future health outcomes. For example, the system can predict changes in a particular elderly individual's balance ability over the next three months and provide early warning of potential fall risk.

[0091] In terms of model optimization, the system not only uses machine learning methods to automatically optimize the model, but also optimizes the model based on the adjustments made by medical staff to the rehabilitation prescription. Specifically, when medical staff manually adjust the rehabilitation prescription based on the elderly's assessment results and subjective feedback (such as increasing the training intensity, adjusting the type of exercise, etc.), the system will record these adjustments as new training data and use this data to retrain the fall risk assessment model and the motor dysfunction assessment model. For example, if a medical staff adjusts the intensity of a certain elderly person's balance training from 3 times a week to 5 times a week, the system will record this adjustment and use it as new input data to re-evaluate the model's prediction effect. In this way, the system can continuously learn and adapt to the decisions of medical staff, improving the accuracy and personalization of the model.

[0092] Sample data: Correlation analysis results: The correlation between gait symmetry and fall risk was 0.85, and the correlation between balance ability and fall risk was 0.78.

[0093] Cluster analysis results: The elderly were divided into "high-risk falls group" (muscle strength score <3, balance ability score <20) and "moderate motor dysfunction group" (muscle strength score 3-4, balance ability score 20-24).

[0094] Time series analysis results: It is predicted that the balance ability score of an elderly person will drop from 18 to 15 in the next three months, indicating an increased risk of falls.

[0095] Prescription Adjustment Record: The healthcare professional adjusts the balance training intensity of an elderly person from 3 times per week to 5 times per week. The system records the adjusted training data and re-trains the model.

[0096] Through model optimization based on prescription adjustments, the system can better adapt to the decisions of healthcare professionals, improving the level of personalization and prediction accuracy of the model, and thus providing more accurate risk management services for motor dysfunction for the elderly.

[0097] To ensure the effectiveness of the treatment plan, the system will conduct regular reassessments. By default, the system conducts a reassessment two weeks after the completion of the rehabilitation prescription, but healthcare professionals can adjust the frequency of assessments based on actual circumstances. The tremor assessment results, muscle strength assessment results, muscle tone assessment results, muscle volume assessment results, motor speed assessment results, fine motor assessment results, joint activity assessment results, Tinetti balance and gait scale assessment results, and modified Barthel index (mbi) assessment results of the elderly can be collected again to assess their motor function and risk level. The system will also record the compliance of the elderly during treatment. By comparing the motor function indicators such as gait symmetry, muscle strength score, and balance ability score between the initial assessment and the reassessment, and calculating the change rate of the indicators, the system can quantitatively evaluate the treatment effect. At the same time, combined with the subjective feelings of the elderly, healthcare professionals can also conduct qualitative efficacy evaluation. All these data will be recorded and stored to provide a basis for subsequent treatment adjustments.

[0098] Sample data: Reassessment results: Muscle strength score: initial assessment is 3, reassessment is 4, improvement is 33%.

[0099] Balance ability score: initial assessment is 18, reassessment is 22, improvement is 22%.

[0100] Compliance record: May 17: completed balance training, muscle strength training, duration 40 minutes.

[0101] May 18: did not follow the plan for training.

[0102] May 19: completed balance training, muscle strength training, duration 45 minutes.

[0103] Subjective feedback: May 17: pain level 1, fatigue level 3.

[0104] May 19: pain level 0, fatigue level 2.

[0105] During operation, the system not only conducts regular assessments but also optimizes algorithmic strategies based on multiple assessment results to improve prediction accuracy and intervention effectiveness. First, the system uses a time series analysis model, combining data from multiple assessments (such as tremor assessments, muscle strength assessments, and balance ability assessments) to predict the elderly's motor function indicators and risk levels over the next period of time. For example, the system can analyze the trend of changes in an elderly person's muscle strength scores over the past three months, predict muscle strength changes over the next month, and provide early warning of possible motor dysfunction risks.

[0106] In terms of algorithmic strategy optimization, the system dynamically adjusts risk assessment models and intervention strategies based on the results of multiple assessments. For example, if the system finds that an elderly person's muscle strength score continues to decline across multiple assessments, the system will automatically adjust their rehabilitation prescription to increase the intensity and frequency of muscle strength training. At the same time, the system uses machine learning algorithms to analyze common features in multiple assessment data and optimize the model's parameters and structure. For example, by analyzing the assessment data of multiple elderly people, the system can discover that certain specific factors (such as gait asymmetry and increased muscle tone) are highly correlated with fall risk, and use these factors as important input variables in the model to improve the model's predictive accuracy.

[0107] This embodiment generates personalized health trend reports based on the results of multiple assessments, helping medical staff better understand the elderly's health status and changing trends. For example, the system can generate a graph showing changes in a person's balance ability over the past six months and, combined with the predicted results, provide targeted intervention recommendations. Medical staff can manually adjust the rehabilitation plan based on these reports and promptly notify the elderly person.

[0108] Sample data: Time series analysis results: An elderly person's muscle strength score dropped from 4 to 3 in the past three months, and is predicted to continue to drop to 2.5 in the next month.

[0109] Algorithm Strategy Optimization: Based on multiple assessments, the system identified gait asymmetry and increased muscle tone as key input variables in the fall risk model. The system also automatically adjusted an elderly patient's rehabilitation prescription, increasing strength training from three times per week to five times per week.

[0110] Health Trend Report: The balance ability score of an elderly person has changed over the past six months: it has dropped from 20 to 18, and is predicted to drop to 16 in the next month.

[0111] Intervention recommendation: Increase the frequency of balance training from 2 times per week to 4 times per week.

[0112] By optimizing algorithm strategies based on multiple evaluation results, the system can more accurately predict the health trends of the elderly and provide more targeted intervention plans, significantly improving the elderly's motor function and quality of life.

[0113] All data involved in this embodiment are stored in a cloud database. In order to protect the privacy of the elderly, all sensitive data will be encrypted and stored using an encryption algorithm. At the same time, the system will regularly back up data to ensure data security. To facilitate the use of medical staff, the system provides a data query function that allows queries by name, age, risk level and other conditions. The system also provides a data analysis function that allows medical staff to view the elderly's motor function trend chart and risk change chart. To ensure data security, the system uses the RBAC mechanism to limit data access rights of different users. At the same time, to facilitate understanding and analysis, the system provides an intuitive visual interface to display data in the form of charts, graphs and reports.

[0114] The following is an embodiment of the system for monitoring the risk of motor dysfunction in the elderly provided by the embodiments of the present disclosure. The system and the method for monitoring the risk of motor dysfunction in the elderly in the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiments of the system for monitoring the risk of motor dysfunction in the elderly, please refer to the embodiments of the method for monitoring the risk of motor dysfunction in the elderly mentioned above.

[0115] The system includes: an information assessment module for obtaining basic information of the elderly and simultaneously implementing physical assessment and testing of the elderly.

[0116] The graded warning module is used to construct a logistic regression model to calculate the fall risk probability value and the motor dysfunction index, and divide the fall risk and motor dysfunction risk into three levels, and then establish a warning coding system.

[0117] The risk judgment module is used to judge the fall risk probability or movement dysfunction index based on the early warning coding system, and push early warning information when the threshold is exceeded.

[0118] The prescription generation module is used to generate standardized rehabilitation prescriptions based on the risk assessment results and the knowledge base of balance muscle strength training programs through a preset decision tree system.

[0119] The parameter adjustment module is used to track the execution of rehabilitation training in real time, establish a compliance evaluation model, and adjust training parameters based on the medical and nursing manual intervention interface.

[0120] The optimization processing module is used to use association rules to identify high-risk factor combinations, and combine synchronous time series analysis to predict the trend of changes in motor function for a preset time in the future. The reinforcement learning module is used to optimize the decision weight based on manual adjustment records of medical staff.

[0121] The data evaluation module is used to trigger the re-evaluation process according to the preset cycle, obtain multi-dimensional evaluation, calculate the improvement rate of key indicators, and obtain multi-dimensional evaluation data.

[0122] The prediction processing module is used to use multi-dimensional evaluation data, apply Bayesian networks to build a dynamic risk assessment model, adopt a federated learning framework to achieve cross-institutional data collaborative training, extract time series features through feature engineering, and predict the motor function indicators and risks of the elderly in the future.

[0123] like Figure 2 As shown, the present application also provides an electronic device, including a display module 103, a memory 102, a processor 101, and a computer program stored in the memory and executable on the processor 101. When the processor 101 executes the program, the steps of the method for monitoring the risk of motor dysfunction in the elderly are implemented.

[0124] In the embodiments of the present invention, electronic devices include, but are not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or claimed herein.

[0125] In the embodiment of the present application, the processor 101 can be implemented by using at least one of a special purpose integrated circuit, a programmable logic device, a field programmable gate array, a processor, a controller, a microcontroller, a microprocessor, and an electronic unit designed to perform the functions described herein. In some cases, such an embodiment can be implemented in a controller. For software implementation, an embodiment such as a process or function can be implemented with a separate software module that allows the execution of at least one function or operation. The software code can be implemented by a software application (or program) written in any appropriate programming language, and the software code can be stored in a memory and executed by a controller.

[0126] The display module 103 is used to display information input by the user or information provided to the user. The display module 103 may include a display panel, which may be configured in the form of a liquid crystal display, an organic light emitting diode, etc.

[0127] The memory 102 may be used to store software programs and various data. The memory 102 may include a high-speed random access memory and a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0128] The present application also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for monitoring the risk of motor dysfunction in the elderly.

[0129] The storage medium can be any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0130] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for monitoring the risk of motor dysfunction in the elderly, characterized in that: Methods include: S101: Obtain basic information of the elderly; S102: Based on the basic information, a logistic regression model is constructed to calculate the fall risk probability value and the motor dysfunction index, and the fall risk and motor dysfunction risk are divided into three risk levels. Then, an early warning coding system is established to obtain the risk assessment results; S103: Determine the fall risk probability or motor dysfunction index based on the early warning coding system, and push an early warning message when it exceeds the threshold; S104: Based on the risk assessment results, a standardized rehabilitation prescription is generated by matching the knowledge base of balance muscle training programs through a preset decision tree system; S105: Real-time tracking of rehabilitation training execution, establishment of a compliance evaluation model, and adjustment of training parameters based on the medical and nursing manual intervention interface; S106: Based on the data generated from steps S101 to S105, association rules are used to identify high-risk factor combinations, and synchronous time series analysis is combined to predict the trend of motor function changes over a preset period of time in the future. The reinforcement learning module is used to optimize the decision weight based on the manual adjustment records of the medical staff; S107: Trigger the re-evaluation process according to the preset period, repeat the multi-dimensional evaluation from step S101 to step S102, calculate the improvement rate of key indicators, and obtain multi-dimensional evaluation data; S108: Based on multi-dimensional assessment data, a dynamic risk assessment model is constructed using a Bayesian network. A federated learning framework is used to achieve cross-institutional data collaborative training. Time series features are extracted through feature engineering to predict the motor function indicators and risks of the elderly in the future.

2. The method for monitoring the risk of motor dysfunction in the elderly according to claim 1, characterized in that: Step S102 specifically includes: Step S1021: Standardize the basic information, integrate the balance sub-item and gait sub-item of the Tinetti scale, calculate the weighted sum according to the balance weight and gait weight, and generate a composite input vector containing demographic characteristics and functional assessment; Step S1022: constructing a fall risk prediction channel based on the composite input vector using a logistic regression algorithm, optimizing the penalty using cross-validation, and outputting a fall risk probability value and classification result; Step S1023: Based on the muscle volume, movement speed, and modified Barthel index features in the generated composite input vector, a support vector machine algorithm is used to construct a movement dysfunction prediction channel. A radial basis kernel function is used to process nonlinear relationships. Grid search is performed to optimize the parameter combination and output a three-level risk level mapped by a Softmax function. Step S1024: Cartesian product is performed on the fall risk probability value and the three-level risk level output in step S1023 to form multiple risk types, and the risk evolution slope is calculated in combination with time series analysis; Step S1025: Encapsulate the fall risk probability value, three-level risk level and trend prediction results into a JSON format data packet, push it to the electronic device, and write the risk grading results into the personal health record database.

3. The method for monitoring the risk of motor dysfunction in the elderly according to claim 2, characterized in that: In the basic information of the method, age was normalized to the range of 0–1, gender was converted to one-hot encoding, BMI retained its original value, and fall history was marked as a binary categorical variable; The balance and gait subitems of the Tinetti scale were integrated, and a weighted sum was calculated with a weight of 0.6 for balance and 0.4 for gait to generate a composite input vector containing demographic characteristics and functional assessments.

4. The method for monitoring the risk of motor dysfunction in the elderly according to claim 2, characterized in that: Based on the muscle volume, movement speed and modified Barthel index features in the composite input vector generated in step S1021, a support vector machine algorithm is used to construct a movement dysfunction prediction channel; The radial basis kernel function is used to deal with nonlinear relationships, and the penalty parameter C∈[0.1,10] and the kernel coefficient γ∈[0.01,1] are defined. The Softmax function is used to map the risk level into three levels. Among them, low risk: 0-0.6, medium risk: 0.6-0.8, high risk: ≥0.8, realizing the quantitative grading of movement dysfunction.

5. The method for monitoring the risk of motor dysfunction in the elderly according to claim 1, characterized in that: Step S103 specifically includes: filtering out data that meets the triggering conditions based on the risk assessment results and the condition judgment module; wherein the triggering conditions are that the fall risk probability is greater than 0.7 or the motor dysfunction index is greater than 0.

8.

6. The method for monitoring the risk of motor dysfunction in the elderly according to claim 1, characterized in that: Step S107 specifically includes: The method for calculating the improvement rate of key indicators is: Key indicator improvement rate = (current value - benchmark value corresponding to basic information) / benchmark value corresponding to basic information; Visualize the improvement rate of key indicators of basic information in the form of a radar chart; The method also builds a dynamic risk assessment model based on the historical accumulation of assessment data and combines it with a Bayesian network. It also uses a federated learning framework to achieve cross-institutional data collaborative training, ensuring data privacy while improving the model's generalization ability.

7. The method for monitoring the risk of motor dysfunction in the elderly according to claim 1, characterized in that: In step S106, the data generated by steps S101 to S105 are used to perform association rule mining using the Apriori algorithm, wherein high-risk combinations with a support degree ≥ 15% and a confidence degree ≥ 80% are screened based on historical assessment data, treatment records, and outcome labels, and an interpretable decision rule library is generated; Predict motor function changes over a preset time period in the future based on a time series analysis model: Adjust the parameter combination of rehabilitation prescription based on the current status of the elderly, historical intervention records, and adjustment behavior of medical staff; The adjusted parameters are combined using a federated learning framework to aggregate multi-institutional data, retrain the association rule base and time series prediction model; generate rehabilitation prescription strategies and push them to electronic devices.

8. A system for monitoring the risk of motor dysfunction in the elderly, characterized by: Used to implement the method for monitoring the risk of motor dysfunction in the elderly as described in any one of claims 1 to 7; The system includes: The information assessment module is used to obtain basic information about the elderly and simultaneously conduct physical assessments and tests on them; The graded warning module is used to build a logistic regression model to calculate the fall risk probability value and motor dysfunction index, and divide the fall risk and motor dysfunction risk into three levels, and then establish a warning coding system; The risk assessment module is used to determine the probability of falling risk or the motor dysfunction index based on the early warning coding system, and push warning information when the threshold is exceeded; The prescription generation module is used to generate standardized rehabilitation prescriptions based on the risk assessment results and the knowledge base of balance muscle training programs through a preset decision tree system; Parameter adjustment module, used to track the execution of rehabilitation training in real time, establish a compliance evaluation model, and adjust training parameters based on the medical and nursing manual intervention interface; The optimization processing module is used to identify high-risk factor combinations using association rules, and predict the trend of motor function changes for a preset time period in the future through synchronous time series analysis. The reinforcement learning module is used to optimize decision weights based on manual adjustments made by medical staff; The data evaluation module is used to trigger the re-evaluation process according to the preset cycle, obtain multi-dimensional evaluation, calculate the improvement rate of key indicators, and obtain multi-dimensional evaluation data; The prediction processing module is used to use multi-dimensional evaluation data, apply Bayesian networks to build a dynamic risk assessment model, adopt a federated learning framework to achieve cross-institutional data collaborative training, extract time series features through feature engineering, and predict the motor function indicators and risks of the elderly in the future.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method for monitoring the risk of motor dysfunction in the elderly as described in any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for monitoring the risk of motor dysfunction in the elderly as claimed in any one of claims 1 to 7 are implemented.

Images