Hand function evaluation training method and system based on multi-sensor fusion
By using multi-sensor fusion technology to calculate muscle and joint stability and grip strength, a hand function evaluation coefficient is generated, which solves the inconsistency problem of manual assessment and achieves precise matching of training intensity and improved safety.
Patent Information
- Application Number
- CN202511561330.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-02-06
AI Technical Summary
In existing technologies, hand function assessment relies on manual evaluation, which limits the consistency and reliability of the assessment results. Training programs are difficult to accurately match the patient's real-time functional status, which may result in insufficient or excessive training intensity, affecting rehabilitation outcomes and causing injury.
By using multi-sensor fusion technology, the stability of muscle groups, finger joints, and grip strength are calculated to generate objective hand function evaluation coefficients, and the training intensity is dynamically adjusted to match the patient's current state.
This achieves objectivity and repeatability in hand function assessment, ensures real-time and precise control of training intensity, and improves training efficiency and safety.
Smart Images
Figure CN121465569A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application generally relates to the technical field of rehabilitation training. More specifically, the present application relates to a multi-sensor fusion based hand function assessment training method and system. BACKGROUND
[0002] Hand function assessment training is an important rehabilitation method for patients with hand dysfunction caused by nervous system diseases. Through scientific and systematic assessment to determine the degree of dysfunction, and accordingly to develop targeted training programs, it is of great significance to restore the patient's fine motor ability and improve the quality of life.
[0003] In the prior art, hand function assessment mainly relies on therapists to conduct manual assessment using standardized scales, such as the Fugl-Meyer Assessment Scale. Therapists observe the patient's performance in completing specific tasks and score according to predetermined standards to determine the hand function status. Based on the assessment results, therapists will develop corresponding training programs and manually adjust training intensity parameters according to observed progress in subsequent training.
[0004] However, the assessment results obtained by this manual assessment and adjustment method are influenced by the subjective experience of therapists, and different therapists may have different assessments of the same patient, resulting in limited consistency and reliability of the assessment results. Moreover, the training program adjustment based on manual assessment is difficult to accurately match the real-time functional status and recovery progress of the patient, which may result in insufficient training intensity affecting rehabilitation effect, or excessive intensity causing compensatory movements and even secondary injuries. SUMMARY
[0005] To solve the technical problem that the training program adjustment of manual assessment is difficult to accurately match the real-time functional status of the patient, the present application provides solutions in the following aspects.
[0006] In a first aspect, a hand function assessment training method based on multi-sensor fusion includes: obtaining electromyographic signals of multiple muscle groups, angles of multiple finger joints, and grip strength of the patient in the previous training session, wherein the multiple muscle groups include: flexor digitorum superficialis, flexor digitorum profundus, and adductor pollicis; fitting the electromyographic signals of each muscle group at multiple time points to obtain the electromyographic signal-time curve of each muscle group, and calculating the stability of the muscle group, wherein the stability of the muscle group is related to the amplitude difference between the peak and the previous trough in the electromyographic signal-time curve of each muscle group; calculating the stability of the finger joints based on the angles of each finger joint at multiple time points; calculating a hand function evaluation coefficient, wherein the patient's grip strength at multiple time points, the peak synergy of the muscle groups, and the stability of the finger joints are all proportional to the hand function evaluation coefficient; and correcting the training intensity of the patient in the previous training session based on the hand function evaluation coefficient to obtain the training intensity of the patient in the current training session, wherein the training intensity is the training duration or the number of sets of training movements.
[0007] Preferably, the formula for calculating the stability of the muscle group is: .
[0008] in A For the stability of the muscle group, for the first k Electromyography signal-time curves of individual muscle groups. The standard deviation of the amplitude difference between all peaks and their preceding troughs. This is the mean of the amplitude differences between all peaks and their preceding troughs. Let be the standard deviation of the slope of the line containing all peaks and their preceding troughs. This is the average slope of the straight line connecting all the peaks and their preceding troughs. k It is a positive integer. norm () represents the standard normalization function.
[0009] Preferably, the formula for calculating the stability of the finger joint is: .
[0010] in S The degree of joint stability, v i For the first i Vector of angle change trend of individual finger joints v j For the first j Vector of angle change trends of individual finger joints M The number of all finger joints, For the first i The first finger joint and the first j The covariance of the vector of angular change trends between individual finger joints For the first i Standard deviation of the vector of trends in the angle changes of individual finger joints For the first j The standard deviation of the vector of the trend of angle change of each finger joint; i Vector of angle change trend of individual finger joints v i With the i The change in angle between two adjacent moments of a finger joint is related to the change in angle of the first finger joint. j Vector of angle change trend of individual finger joints v j With the j The change in angle between two adjacent moments of a finger joint is related to the change in angle between them.
[0011] Preferred, the first i Vector of angle change trend of individual finger joints v i Satisfying the formula: .
[0012] For the first i The angle of each finger joint at the first moment. For the first i The angle of each joint at the second moment For the first i The finger joint in the first n The angle at time -1 For the first i The finger joint in the first n From the perspective of time, n For the number of all times, This represents the time difference between two adjacent moments.
[0013] Preferred, the first j Vector of angle change trend of individual finger joints v j Satisfying the formula: .
[0014] For the first j The angle of each finger joint at the first moment. For the first j The angle of each joint at the second moment For the first j The finger joint in the first n The angle at time -1 For the first j The finger joint in the first n The perspective of time.
[0015] Preferably, the formula for calculating the hand function evaluation coefficient is: .
[0016] wherein G is a grip strength evaluation coefficient, the grip strength evaluation coefficient G is related to the average value and the maximum value of the grip strength of the patient at all times and 0 G <1, ω 1 is a first weight coefficient of a preset size, ω 2 is a second weight coefficient of a preset size, ω 3 is a third weight coefficient of a preset size.
[0017] Preferably, the formula for calculating the grip strength evaluation coefficient is: wherein μ F is the average value of the grip strength of the patient at all times, F max is the maximum value of the grip strength of the patient at all times, F q is a standard maximum grip strength value of a preset size, norm () represents a standard normalization function.
[0018] Preferably, the correction of the training intensity of the patient in the last training according to the hand function evaluation coefficient comprises: when the training intensity is the training duration, the formula for correcting the training duration of the patient in the last training is: wherein T is the training duration of the patient in the last training, is the corrected training duration, H 0 is a standard hand function evaluation coefficient of a preset size, and 0 H 0<1, α is a correction amplitude coefficient of a preset size, and 0 α <1; when the training intensity is the number of training action groups, the formula for correcting the number of training action groups of the patient in the last training is: , N is the number of training action groups of the patient in the last training, is the corrected number of training action groups.
[0019] Preferably, the hand function evaluation and training method based on multi-sensor fusion further comprises: in response to the hand function evaluation coefficient being less than a preset threshold, an alarm is performed.
[0020] In a second aspect, a hand function evaluation training system based on multi-sensor fusion comprises a processor and a memory, the memory storing a computer program, and the processor executes the computer program to implement a hand function evaluation training method based on multi-sensor fusion according to any one of the above summary.
[0021] The present application has the following beneficial effects: The present application calculates the muscle group stability degree, the finger joint stability degree and the grip strength evaluation coefficient, generates an accurate hand function evaluation coefficient through weighted fusion, eliminates the evaluation deviation caused by human factors, and ensures the objectivity and repeatability of the evaluation result (i.e. the hand function evaluation coefficient).
[0022] The present application determines the adjustment range of the training intensity (i.e. the training duration or the number of groups) according to the evaluation result, realizes real-time and accurate regulation of the training intensity, makes the training scheme always best match the current functional state of the patient, and thus ensures the safety of the hand function training of the patient and improves the training efficiency. BRIEF DESCRIPTION OF DRAWINGS
[0023] The above and other objects, features and advantages of the exemplary embodiments of the present application will be more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which several embodiments of the present application are shown by way of example, and wherein like reference numerals refer to like elements throughout. In the drawings: Figure 1 is a step flow chart schematically showing a hand function evaluation training method based on multi-sensor fusion according to an embodiment of the present application; Figure 2 is a structural block diagram schematically showing a hand function evaluation training system based on multi-sensor fusion according to an embodiment of the present application. DETAILED DESCRIPTION
[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.
[0025] The specific embodiments of the present application will be described in detail below with reference to the drawings.
[0026] Figure 1 is a step flow chart schematically showing a hand function evaluation training method based on multi-sensor fusion according to an embodiment of the present application;
[0027] As Figure 1As shown, a hand function evaluation training method based on multi-sensor fusion includes steps S1 to S5.
[0028] Step S1: Obtain electromyographic signals of multiple muscle groups, angles of multiple finger joints, and grip strength from the patient during the previous training session.
[0029] Among them, several muscle groups involved in movement include: flexor digitorum superficialis, flexor digitorum profundus, and adductor pollicis.
[0030] It should be noted that the flexor digitorum superficialis, flexor digitorum profundus, and adductor pollicis are the core muscle groups responsible for hand grasping. The flexor digitorum superficialis dominates the flexion of the proximal interphalangeal joints, forming the basic outline of the grasping action; the flexor digitorum profundus controls the fine flexion of the distal interphalangeal joints, enabling a stable grasp of objects; and the adductor pollicis provides the adduction and opposition force of the thumb, completing the closing phase of the grasp.
[0031] Step S2: Fit the electromyographic signals of each muscle group at multiple times to obtain the electromyographic signal-time curve of each muscle group, and calculate the stability of the muscle group.
[0032] In one embodiment, the Savitzky-Golay filtering algorithm is used to fit the electromyographic (EMG) signals of various muscle groups at multiple time points. It should be noted that the fitted EMG signal-time curve can identify the positions and amplitudes of peaks and troughs. The Savitzky-Golay smoothing algorithm is a data smoothing technique that achieves smoothing by performing polynomial fitting on data points within a certain length window. The Savitzky-Golay smoothing algorithm improves data accuracy without altering the signal trend and width, making it suitable for processing time-series data.
[0033] The stability of the muscle groups is related to the amplitude difference between the peak and the previous trough in the electromyographic signal-time curve of each muscle group.
[0034] In one embodiment, the formula for calculating the stability of the muscle group is: .
[0035] in A For the stability of the muscle group, for the first k Electromyography signal-time curves of individual muscle groups. The standard deviation of the amplitude difference between all peaks and their preceding troughs. This is the mean of the amplitude differences between all peaks and their preceding troughs. Let be the standard deviation of the slope of the line containing all peaks and their preceding troughs. This is the average slope of the straight line connecting all the peaks and their preceding troughs. k It is a positive integer.norm () represents the standard normalization function.
[0036] It should be noted that the formula The value reflects the first k The consistency of force exerted during multiple muscle contractions of a motor muscle group. A smaller value indicates more stable force exertion and better repeatability with each contraction. (Formula) The value reflects the consistency of muscle activation and relaxation rates. The smaller the value, the more regular the muscle contraction and relaxation process, and the more refined the neural control.
[0037] norm The () function converts indices of different dimensions, i.e., formulas and After normalization, the corresponding normalized values are added together to obtain the stability index of the k-th muscle group. The average of the stability indices of the three muscle groups is then used to obtain a comprehensive assessment of muscle group stability. A .
[0038] Based on this, the degree of muscle group stability A The larger the value, the more stable the neuromuscular control function of the patient's hand motor muscles, and the better the repeatability and regularity of muscle activity.
[0039] Step S3: Calculate the stability of the finger joints based on the angles of each finger joint at multiple times.
[0040] In one embodiment, the formula for calculating the stability of the finger joint is: .
[0041] in S The degree of joint stability, v i For the first i Vector of angle change trends of individual finger joints v j For the first j Vector of angle change trends of individual finger joints M The number of all finger joints, For the first i The first finger joint and the first j The covariance of the vector of angular change trends between individual finger joints For the first i Standard deviation of the vector of trends in the angle changes of individual finger joints For the first j The standard deviation of the vector of the trend of angle change of each finger joint; i Vector of angle change trend of individual finger joints v i With the iThe change in angle between two adjacent moments of a finger joint is related to the change in angle of the first finger joint. j Vector of angle change trend of individual finger joints v j With the j The change in angle between two adjacent moments of a finger joint is related to the change in angle between them.
[0042] It should be noted that the formula Indicates the first i The vector showing the trend of angle change of each finger joint and the first j The Pearson correlation coefficient between the angular change trend vectors of two finger joints is used to measure the similarity of their movement trends. The formula is as follows: The larger the value, the more significant the [value]. i The first finger joint and the first j The more similar the trends in the angle changes of each finger joint, the greater the stability of the finger joints when all joint movements are highly coordinated. S The value approaches 1. When movement is chaotic and uncoordinated, the stability of the finger joints... S A value close to 0 indicates poor joint stability. Therefore, the degree of joint stability... S The higher the value, the more stable the movement of the patient's finger joints.
[0043] In one embodiment, the first i Vector of angle change trend of individual finger joints v i Satisfying the formula: .
[0044] For the first i The angle of each finger joint at the first moment. For the first i The angle of each joint at the second moment For the first i The finger joint in the first n The angle at time -1 For the first i The finger joint in the first n From the perspective of time, n For the number of all times, This represents the time difference between two adjacent moments.
[0045] In one embodiment, the first j Vector of angle change trend of individual finger joints v j Satisfying the formula: .
[0046] For the first jThe angle of each finger joint at the first moment. For the first j The angle of each joint at the second moment For the first j The finger joint in the first n The angle at time -1 For the first j The finger joint in the first n The perspective of time.
[0047] It should be noted that the angle change trend vectors of each finger joint include... n The corresponding time n -1 angular change component.
[0048] Step S4: Calculate the hand function evaluation coefficient.
[0049] Among them, the patient's grip strength at multiple moments, the peak synergy of the muscle groups, and the stability of the finger joints are all proportional to the hand function evaluation coefficient.
[0050] In one embodiment, the formula for calculating the hand function evaluation coefficient is: .
[0051] in G This is the grip strength evaluation coefficient. G It is related to the average and maximum grip strength of the patient at all times and 0 < G <1, ω 1 is the first weighting coefficient of the preset size. ω 2 is the second weighting coefficient with a preset size. ω 3 is the third weighting coefficient with a preset size.
[0052] In one embodiment, the formula for calculating the grip strength evaluation coefficient is: ,in, μ F This represents the average grip strength of the patient at all times. F max This represents the patient's maximum grip strength at all times. F q The standard maximum grip strength value for the preset size. norm () represents the standard normalization function.
[0053] It should be noted that the formula The value reflects the efficiency and sustainability of the patient's grip strength output. (Formula) The higher the value, the higher the patient's endurance level and the stronger the patient's ability to sustain hand force. (Formula) reflects the closeness of the patient's current actual grip level to the rehabilitation expected target. Formula The closer the value of the formula to 1, the more ideal the patient's grip recovery is, and the closer the preset target is.
[0054] Step S5: correcting the training intensity of the patient in the last training according to the hand function evaluation coefficient to obtain the training intensity of the patient in the current training.
[0055] The training intensity is the training duration or the number of training actions.
[0056] In an embodiment, the correcting the training intensity of the patient in the last training according to the hand function evaluation coefficient includes: when the training intensity is the training duration, the formula for correcting the training duration of the patient in the last training is: , T is the training duration of the patient in the last training, is the corrected training duration, H 0 is a preset standard hand function evaluation coefficient, and 0 H 0<1, α is a preset correction amplitude coefficient, and 0 α <1; and when the training intensity is the number of training action groups, the formula for correcting the number of training action groups of the patient in the last training is: , N is the number of training action groups of the patient in the last training, is the corrected number of training action groups.
[0057] It should be noted that the present application compares the performance of the patient in the last training (i.e., the hand function evaluation coefficient H ) with a preset, expected standard (i.e., the standard hand function evaluation coefficient H 0), thereby dynamically adjusting the training load.
[0058] When H > H 0, it indicates that the functional state of the patient is better than expected, the calculation result of the formula is greater than 1, thereby increasing the training duration or the number of groups to promote the progress of hand function.
[0059] When H < H 0, it indicates that the current state of the patient does not meet the expectation or fatigue occurs, the calculation result of the formula is less than 1, thereby reducing the training load to prevent overtraining and injury and ensure the safety of the patient.
[0060] The correction amplitude coefficient α is used to control the "step length" of the adjustment, and the correction amplitude coefficientα The smaller the value of the adjustment conserving smoothness, the larger the value of the correction amplitude coefficient α The larger the value of the adjustment aggressive speed. In one embodiment, the value of the correction amplitude coefficient α is 0.2.
[0061] The application further comprises: in response to the hand function evaluation coefficient being less than a preset threshold, performing an alarm.
[0062] It should be noted that if the hand function evaluation coefficient is less than the preset threshold, it indicates that there is a significant mismatch between the training scheme in the last training and the actual functional state of the patient. The patient may insist on training in a state of pain or fatigue, or the training load is far beyond the current ability of the patient, resulting in an increase in compensatory movements. Therefore, an alarm is needed to inform medical staff to check the patient.
[0063] In one embodiment, the preset threshold is 0.45.
[0064] Figure 2 is a structural block diagram of a hand function evaluation training system based on multi-sensor fusion according to one of the embodiments.
[0065] The application further provides a hand function evaluation training system based on multi-sensor fusion. As shown in Figure 2 the system includes a processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, a hand function evaluation training method based on multi-sensor fusion according to the first aspect of the application is realized.
[0066] The system further includes a communication interface and other components familiar to those skilled in the art, the settings and functions of which are known in the art, and therefore will not be described here.
[0067] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.
[0068] In the description of this specification, "multiple" or "several" means at least two, such as two, three or more, unless otherwise explicitly specified.
[0069] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.
Claims
1. A hand function evaluation and training method based on multi-sensor fusion, characterized in that, include: The electromyographic signals of multiple muscle groups involved in the patient's previous training, the angles of multiple finger joints, and the grip strength were obtained. These muscle groups included the flexor digitorum superficialis, flexor digitorum profundus, and adductor pollicis. The electromyographic signals of each muscle group at multiple times are fitted to obtain the electromyographic signal-time curve of each muscle group, and the stability of the muscle group is calculated. The stability of the muscle group is related to the amplitude difference between the peak and the previous trough in the electromyographic signal-time curve of each muscle group. The stability of the finger joints is calculated based on the angles of each finger joint at multiple times. Calculate the hand function evaluation coefficient, in which the patient's grip strength at multiple moments, the peak coordination degree of the muscle groups, and the stability of the finger joints are all proportional to the hand function evaluation coefficient; The training intensity of the patient in the previous training session is adjusted based on the hand function evaluation coefficient to obtain the training intensity of the patient in the current training session, wherein the training intensity is the training duration or the number of sets of training movements.
2. The hand function evaluation and training method based on multi-sensor fusion according to claim 1, characterized in that, The formula for calculating the stability of the muscle group is: ； in A For the stability of the muscle group, for the first k Electromyography signal-time curves of individual muscle groups. The standard deviation of the amplitude difference between all peaks and their preceding troughs. This is the mean of the amplitude differences between all peaks and their preceding troughs. Let be the standard deviation of the slope of the line containing all peaks and their preceding troughs. This is the average slope of the straight line connecting all the peaks and their preceding troughs. k It is a positive integer. norm () represents the standard normalization function.
3. The hand function evaluation and training method based on multi-sensor fusion according to claim 1, characterized in that, The formula for calculating the stability of the finger joint is: ； in S The degree of joint stability, v i For the first i Vector of angle change trends of individual finger joints v j For the first j Vector of angle change trends of individual finger joints M The number of all finger joints, For the first i The first finger joint and the first j The covariance of the vector of angular change trends between individual finger joints For the first i Standard deviation of the vector of trends in the angle changes of individual finger joints For the first j The standard deviation of the vector of the trend of angle change of each finger joint; i Vector of angle change trend of individual finger joints v i With the i The change in angle between two adjacent moments of a finger joint is related to the change in angle of the first finger joint. j Vector of angle change trend of individual finger joints v j With the j The change in angle between two adjacent moments of a finger joint is related to the change in angle between them.
4. The hand function evaluation and training method based on multi-sensor fusion according to claim 3, characterized in that, No. i Vector of angle change trend of individual finger joints v i Satisfying the formula: ； For the first i The angle of each finger joint at the first moment. For the first i The angle of each joint at the second moment For the first i The finger joint in the first n The angle at time -1 For the first i The finger joint in the first n From the perspective of time, n For the number of all times, This represents the time difference between two adjacent moments.
5. The hand function evaluation and training method based on multi-sensor fusion according to claim 4, characterized in that, No. j Vector of angle change trend of individual finger joints v j Satisfying the formula: ； For the first j The angle of each finger joint at the first moment. For the first j The angle of each joint at the second moment For the first j The finger joint in the first n The angle at time -1 For the first j The finger joint in the first n The perspective of time.
6. The hand function evaluation and training method based on multi-sensor fusion according to claim 5, characterized in that, The formula for calculating the hand function evaluation coefficient is as follows: ； in G This is the grip strength evaluation coefficient. G It is related to the average and maximum grip strength of the patient at all times and 0 < G <1, ω 1 is the first weighting coefficient of the preset size. ω 2 is the second weighting coefficient with a preset size. ω 3 is the third weighting coefficient with a preset size.
7. The hand function evaluation and training method based on multi-sensor fusion according to claim 6, characterized in that, The formula for calculating the grip strength evaluation coefficient is as follows: ,in, μ F This represents the average grip strength of the patient at all times. F max This represents the maximum grip strength of the patient at all times. F q The standard maximum grip strength value for the preset size. norm () represents the standard normalization function.
8. The hand function evaluation and training method based on multi-sensor fusion according to claim 6, characterized in that, The adjustment of the patient's training intensity in the previous training session based on the aforementioned hand function evaluation coefficient includes: When the training intensity equals the training duration, the formula for adjusting the patient's training duration from the previous session is: ,in T The duration of the patient's last training session. This is the revised training duration. H 0 is the preset standard hand function evaluation coefficient, and 0 < H 0<1, α The correction magnitude coefficient is a preset value, and 0 < α <1; When the training intensity is equal to the number of training sets, the formula for adjusting the number of training sets performed by the patient in the previous training session is as follows: , N This refers to the number of sets of exercises the patient performed in the previous training session. This represents the corrected number of training sets.
9. The hand function evaluation and training method based on multi-sensor fusion according to claim 1, characterized in that, Also includes: An alarm is triggered if the hand function evaluation coefficient is less than a preset threshold.
10. A hand function assessment and training system based on multi-sensor fusion, comprising a processor and a memory, characterized in that, The memory stores a computer program, and the processor executes the computer program to implement a hand function evaluation training method based on multi-sensor fusion as described in any one of claims 1-9.