A real-time monitoring and evaluation method for orthopedic rehabilitation process

Abstract

The present application relates to the technical field of health assessment, in particular to a kind of orthopedics rehabilitation process real-time monitoring and evaluation method, comprising the following steps, based on gait biomechanics parameters, feature point is identified, node is optimized in combination with weight, step frequency and stride change, extract node correction parameter group, analyze knee joint and ankle joint muscle strength change, in combination with gait stability and joint activity, obtain correlation index proportion sequence.In the present application, parameters such as step frequency, stride, weight are combined with automatic identification of feature points within gait cycle, through dynamic correction of feature point determination standard, the time sequence sensitivity and individual adaptability of data are enhanced, muscle strength recovery, gait stability index is synchronized integrated and quantified in different stages, the linkage relationship between indexes can be adjusted according to real-time data, further support the accurate tracking of functional recovery path, and rehabilitation progress can be reflected by dynamic structure of multi-dimensional parameters.

Description

Technical Field

[0001] This invention relates to the field of health assessment technology, and in particular to a method for real-time monitoring and assessment of orthopedic rehabilitation processes. Background Technology

[0002] Health assessment involves multi-dimensional quantitative analysis and dynamic evaluation of human health status, including physiological signal acquisition, medical data analysis, motor function testing, and rehabilitation process assessment. This field utilizes various sensing methods and data analysis techniques to achieve continuous monitoring and periodic judgment of patients' health status, providing a scientific basis for medical intervention and rehabilitation treatment. Traditional real-time monitoring and assessment methods for orthopedic rehabilitation refer to offline collection and manual recording of functional indicators such as joint range of motion, muscle strength, and gait during the patient's rehabilitation phase through regular manual examinations, physical assessments, or the use of conventional medical instruments such as goniometers, muscle strength meters, and gait analyzers.

[0003] Existing technologies rely on manual data collection and single-point measurements using conventional instruments. Data acquisition cycles are long, and patients' daily fluctuations cannot be continuously presented. The frequency of data collection and the timeliness of data collection are insufficient to meet the dynamic changes required in the rehabilitation process. Assessment results depend more on human experience or the collection of indicators in a single time period. They lack the ability to automatically adapt to individual differences and cannot reflect the phased changes in rehabilitation and the functional linkage process. Actual assessments are prone to being delayed, biased, or even missing key turning points in the patient's rehabilitation. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a method for real-time monitoring and evaluation of orthopedic rehabilitation processes. The technical solution is as follows:

[0005] On the one hand, a method for real-time monitoring and assessment of the orthopedic rehabilitation process is provided, including the following steps:

[0006] S1: Based on gait biomechanical parameters, analyze the gait frequency sequence, compare the position of each data point in the gait cycle, calculate local maxima and minima, determine the interval and amplitude between poles, screen for abnormal data, and obtain a set of gait cycle features.

[0007] S2: Based on the gait cycle feature set, determine the changing trends of weight, cadence and stride, compare the influence of cadence and weight changes on feature point distribution, filter time-distance differences, adjust the amplitude filtering criteria, optimize feature point assignment, and obtain node correction parameter set;

[0008] S3: Based on the node correction parameter group, analyze the continuous changes in knee flexion and extension muscle strength and ankle dorsiflexion muscle strength, calculate the rate of muscle strength change in groups, determine the range of muscle strength fluctuation, screen standard groups, extract time period sequences, and obtain muscle strength change trend indicators;

[0009] S4: Based on the muscle strength change trend identifier, analyze the gait stability score and knee joint range of motion. Using a fixed period window, analyze the data curve, screen for rate direction change segments, record parameter change nodes, and obtain a set of functional transition nodes.

[0010] On the other hand, the gait cycle feature set includes gait event nodes, cycle position labels, and feature point indexes; the node correction parameter set includes node adjustment factors, distribution correction labels, and matching interval mappings; the muscle strength change trend identifier includes muscle strength growth segments, change rate labels, and trend classification numbers; and the functional transition node set includes transition interval numbers, functional transition types, and temporal feature labels.

[0011] On the other hand, the steps for obtaining the gait cycle feature set are as follows:

[0012] S101: Based on gait biomechanical parameters, compare the time position of each data point in the gait cycle, determine the trend of data change within the cycle, calculate the peaks and troughs between adjacent data points, screen out pole pairs with prominent changes, optimize the continuity of the data sequence, remove abrupt fragments, and obtain pole structure screening data.

[0013] S102: Based on the pole structure screening data, determine the distribution of each group of poles in each period, analyze the changes in the time position of poles between adjacent periods, screen poles with stable distribution, optimize the order of occurrence and distribution of poles within the period, retain the characteristic points of regular distribution, and obtain a set of periodically stable nodes.

[0014] S103: Based on the set of periodically stable nodes, analyze the distribution density of each node within the period, determine the changes in the time interval between nodes, and select a node group with continuous distribution and reasonable density within the period to obtain a gait cycle feature set.

[0015] On the other hand, the specific steps for obtaining the node correction parameter set are as follows:

[0016] S201: Based on the gait cycle feature set, analyze the cadence, weight and stride data, determine the direction of change of each parameter in each cycle, compare the trend of change of adjacent cycles, identify the segments that show a consistent trend of change, and obtain the set of linkage trend intervals by comparing the fluctuation characteristics of each parameter.

[0017] S202: Based on the aforementioned set of linkage trend intervals, compare the impact of cadence trends and weight fluctuations on the distribution of feature points, analyze the changes in the sampling positions of feature points within the difference period, determine the time distribution differences caused by cadence changes and weight fluctuations, screen the optimal combination of parameters that affect the distribution of feature points, and obtain the feature distribution influencing factor.

[0018] S203: Based on the characteristic distribution influencing factor, adjust the corresponding feature point interval screening criteria, optimize the attribution judgment of each feature point, analyze the attribution structure of the selected feature points, judge the rationality of the attribution changes within the period, correct the classification results, and obtain the node correction parameter group.

[0019] On the other hand, the specific steps for obtaining the muscle strength change trend indicator are as follows:

[0020] S301: Based on the node correction parameter group, analyze the knee joint flexor and extensor muscle strength and the ankle joint dorsiflexor muscle strength, determine the continuous changes of muscle strength data in each time period, compare the rising and falling trends of different time periods, filter continuous data in the same trend interval, and establish a continuous muscle strength grouping sequence.

[0021] S302: Based on the continuous muscle strength grouping sequence, calculate the rate of change of muscle strength in each group, analyze the fluctuation amplitude within each group, determine the difference in rate between groups, identify groups with stable rate of change, remove unstable groups, uniformly identify the remaining groups, and obtain a stable rate feature sequence.

[0022] S303: Based on the stable rate feature sequence, determine the time order of each group, analyze data segments with consistent trends, filter the same type of change intervals, number the trend types of the groups, establish the correspondence between muscle strength trend labels and time periods, and obtain muscle strength change trend identifiers.

[0023] On the other hand, the specific steps for obtaining the functional conversion node set are as follows:

[0024] S401: Based on the muscle strength change trend identifier, analyze the periodic changes of the gait stability score item, determine the fluctuation of stability parameters in each period, filter the time segments where the fluctuation direction changes, detect the period number corresponding to each change segment, and obtain the stability change interval group.

[0025] S402: Based on the set of stable change intervals, analyze the time series data of knee joint activity amplitude within the corresponding period, determine the maximum range of activity amplitude within each period, identify the key time point when the direction of activity amplitude change changes in the direction of change, and obtain the knee joint activity turning point.

[0026] S403: Based on the knee joint activity transition nodes, screen nodes where gait stability and knee joint range of motion change simultaneously, uniformly encode the synchronously changing nodes, establish corresponding transition types and time sequence features, and obtain a set of functional transition nodes.

[0027] On the other hand, the method also includes:

[0028] S5: Based on the set of functional transition nodes, analyze the correspondence between gait stability performance and knee joint flexibility changes, combine the synchronicity of muscle strength recovery time sequence, determine the linkage changes of each parameter, sort out the functional recovery path, and obtain the correlation index ratio sequence.

[0029] The associated indicator ratio sequence includes linkage parameter grouping, ratio allocation labels, and recovery phase structure.

[0030] On the other hand, the specific steps for obtaining the correlation index ratio sequence are as follows:

[0031] S501: Based on the set of functional conversion nodes, determine the changing trends of gait stability and knee joint flexibility in each cycle, compare the parameter fluctuations of the two in the same cycle, filter the cycle numbers with the same direction of gait stability and knee joint flexibility changes, and obtain the cycle linkage corresponding sequence.

[0032] S502: Based on the corresponding sequence of the cycle linkage, combined with the time series data of the muscle strength recovery marker in each cycle, analyze the synchronicity of the time distribution of muscle strength changes with gait stability and knee joint flexibility, determine the linkage characteristics of each parameter in each cycle, screen the cycles that show synchronous fluctuations, and obtain the parameter distribution ratio group.

[0033] S503: Based on the parameter distribution ratio group, compare the linkage ratio of each cycle, analyze the parameter grouping characteristics of each recovery stage, determine the structural changes of the linkage ratio with the recovery process, summarize the distribution characteristics of the coordinated changes of parameters within the difference cycle, and obtain the correlation index ratio sequence.

[0034] On the other hand, the gait biomechanical parameters refer to quantitative data reflecting human movement and mechanical characteristics during walking, including ground reaction force, joint angle, lower limb acceleration and plantar pressure, and the gait cycle refers to the time or data interval experienced by a complete gait.

[0035] On the other hand, the continuous change refers to the dynamic change of the collected muscle strength parameters over time, representing the process of muscle strength change. The muscle strength fluctuation range refers to the amplitude or fluctuation range of muscle strength within the target range. The extraction of time period sequence refers to serializing the time intervals corresponding to the groups that meet the conditions.

[0036] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0037] By continuously monitoring gait and physiological signals, parameters such as cadence, stride length, and weight are combined with the automatic identification of feature points within the gait cycle. By dynamically correcting the feature point judgment criteria, the temporal sensitivity and individual adaptability of the data are enhanced. Muscle strength recovery and gait stability indicators are synchronously integrated and quantified at different stages, and the linkage between indicators can be adjusted according to real-time data, further supporting the accurate tracking of functional recovery paths. Rehabilitation progress is reflected in a dynamic structure of multi-dimensional parameters. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart of the main steps of the present invention;

[0040] Figure 2 This is a flowchart of steps S1 of the present invention;

[0041] Figure 3 This is a flowchart of steps S2 of the present invention;

[0042] Figure 4 This is a flowchart of steps S3 of the present invention;

[0043] Figure 5 This is a flowchart of step S4 of the present invention;

[0044] Figure 6 This is a flowchart of steps S5 of the present invention. Detailed Implementation

[0045] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0046] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0047] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0048] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0049] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0050] This invention provides a method for real-time monitoring and evaluation of the orthopedic rehabilitation process, such as... Figure 1 As shown, it includes the following steps:

[0051] S1: Based on gait biomechanical parameters, analyze the gait frequency data sequence, compare the relative positions of each data point within the gait cycle, calculate the local maxima and minima between each data point within each cycle, determine the interval and amplitude changes between poles, screen out abnormal data that do not conform to the temporal continuity, and integrate effective feature points to obtain a gait cycle feature set.

[0052] S2: Based on the gait cycle feature set, determine the changing trends of patient weight, cadence and stride, compare the influence of cadence trend and weight fluctuation on feature point distribution, screen the differences in feature point time interval, adjust the screening criteria of amplitude range, optimize the classification of each feature point, and obtain the node correction parameter set.

[0053] S3: Based on the node-corrected parameter group, analyze the continuous changes in knee flexor and extensor strength and ankle dorsiflexor strength, calculate the rate of change of muscle strength in each group, determine the range of muscle strength fluctuation in the grouped data, screen the groups that meet the standards, and extract the time sequence to obtain the muscle strength change trend identifier.

[0054] S4: Based on the trend of muscle strength change, analyze the gait stability score and knee joint range of motion. Use a fixed period window to analyze the gait and joint activity data curves, screen the segments of rate direction change within the time period, record the nodes where parameters change, and obtain the functional transition node set.

[0055] S5: Based on the functional transition node set, analyze the correspondence between gait stability performance and knee joint flexibility changes. Combined with the temporal synchronicity of muscle strength recovery markers, determine the linkage changes of each parameter in the same cycle, sort out the functional recovery path, and obtain the proportional sequence of related indicators.

[0056] The gait cycle feature set includes gait event nodes, cycle position labels, and feature point indexes. The node correction parameter set includes node adjustment factors, distribution correction labels, and matching interval mappings. The muscle strength change trend identifier includes muscle strength growth segments, change rate labels, and trend classification numbers. The functional transition node set includes transition interval numbers, functional transition types, and temporal feature labels. The associated indicator ratio sequence includes linkage parameter grouping, ratio allocation labels, and recovery phase structure.

[0057] In S1, gait biomechanical parameters refer to quantitative data reflecting human movement and mechanical characteristics during walking, such as ground reaction force, joint angles, lower limb acceleration, and plantar pressure; gait frequency data sequence refers to the original signal sequence arranged according to the collected gait frequency (steps per unit time), giving each gait cycle data a sequential relationship; gait cycle refers to the time or data interval experienced by a complete gait (such as from the landing of one foot to the landing of the same foot again), which is the basic unit of gait analysis; relative position refers to the position of each data point in the current gait cycle corresponding to the time or acquisition sequence number; local maxima and extrema... Small points refer to the local maximum (peak) and minimum (valley) values ​​of the data curve in the gait cycle signal, reflecting key events of the movement (such as foot strike, foot lift, etc.); interval and amplitude changes refer to the time / number of data points interval between poles, as well as the numerical differences of the poles, reflecting gait structure characteristics and movement amplitude; inconsistent abnormal data refers to data points whose timing or amplitude is obviously abnormal and does not conform to normal physiological laws due to measurement errors, abnormal movements, etc.; effective feature points refer to data nodes (such as gait event points) that, after judgment, conform to the laws of gait dynamics and have analytical and discriminative significance.

[0058] In S2, the trend of change refers to the overall direction and shape of the increase, decrease, or fluctuation of parameters (such as weight, cadence, and stride length) over time; the cadence trend refers to the increase, decrease, or stabilization of cadence over a period of time or multiple periods; weight fluctuation refers to the change in the patient's weight during the measurement period, which is affected by rehabilitation progress, weight-bearing training, etc.; the influence of feature point distribution refers to the effect of changes in weight and cadence on the location and number of feature points in the gait cycle; the difference in time interval refers to the length of the interval between feature points on the time axis, reflecting the changes in the gait cycle structure; the screening criteria refer to the specific conditions (such as logical conditions such as interval and amplitude) used to determine whether a feature point is retained, rather than specific numerical values; the attribution judgment refers to the conclusion of whether a feature point belongs to the valid analysis object.

[0059] In S3, continuous change refers to the dynamic changes of the collected muscle strength parameters over time, reflecting the process of muscle strength change; muscle strength change rate refers to the speed at which muscle strength values ​​change over a period of time, usually calculated by the difference between adjacent data points / time interval; muscle strength fluctuation range refers to the amplitude or fluctuation range of muscle strength within a specific interval; grouping that meets the criteria refers to determining which groups of muscle strength data (such as rate and amplitude) meet the preset logical conditions after grouping the muscle strength data; extracting time interval sequences refers to serializing the time intervals corresponding to the groups that meet the conditions for subsequent analysis.

[0060] In S4, the gait stability score refers to parameters that quantify the degree of gait stability (such as gait symmetry index, coefficient of variation, etc.); knee joint range of motion refers to the range of changes in the flexion and extension angles of the knee joint within the gait cycle, reflecting the flexibility of the joint; fixed cycle window refers to segmenting the data for analysis using a preset time length or gait cycle length as the unit of analysis; activity data curve refers to the continuous curve of gait stability and knee joint range of motion changing with time or gait cycle; the segment of rate direction change refers to the time interval in the analysis curve where the rate of data change shows a positive or negative inflection (such as from rising to falling); the node of parameter change refers to the point of rate inflection, the corresponding key time point or sequence number, reflecting the node of state change; the functional transition node set is used to identify and locate the time points or intervals when gait function changes significantly during rehabilitation, that is, in the process of gait monitoring and assessment, through data analysis, the location in the curve of gait, muscle strength, joint activity and other functional parameters that shows significant changes in rate direction, amplitude structure or stage characteristics.

[0061] In S5, gait stability refers to the changes in gait stability parameters (such as coefficient of variation, symmetry, etc.) within the cycle; knee joint flexibility changes refer to the changes in knee joint range of motion (such as maximum and minimum flexion and extension angles, etc.) with the gait cycle; correspondence refers to the correlation between two sets of parameters in terms of time and numerical change trends; temporal synchronicity refers to whether the changes of different parameters (such as muscle strength, joint flexibility, etc.) remain synchronized within the same time period; linkage changes refer to the dynamic changes of multiple parameters that occur together and are interconnected over time; and functional recovery path refers to the evolutionary path or stage process of rehabilitation function advancement by sorting out the comprehensive change trends of parameters.

[0062] like Figure 2 As shown, the specific steps for obtaining the gait cycle feature set are as follows:

[0063] S101: Based on gait biomechanical parameters, compare the time position of each data point in the gait cycle, determine the trend of data change within the cycle, calculate the peaks and troughs between adjacent data points, screen out pole pairs with prominent changes, optimize the continuity of the data sequence, remove abrupt fragments, and obtain pole structure screening data.

[0064] After extracting the original ground reaction force, knee joint angle, ankle joint acceleration, and plantar pressure signal sequences, the data are arranged according to sampling time. A complete gait cycle (e.g., 0.8 seconds) is divided into 1000 sampling points, with a time interval of 0.0008 seconds between each data point. The specific time position of each sampling point within the gait cycle is analyzed sequentially. For example, the 200th point is located at 20% of the cycle. A window scan is performed on three consecutive points. For example, if the data points 198, 199, and 200 show a value significantly higher than the preceding and following values, it is considered a peak; if it is lower, it is considered a trough. If 198 is 12.1, 199 is 15.4, and 200 is 13.2, then 199 is recorded as a peak. The numerical difference between each peak and its immediate neighboring trough is then statistically analyzed. All such differences are statistically analyzed throughout the entire cycle. If the difference between most peaks and troughs is concentrated in the range of 3 to 6, a judgment benchmark of 5 is set. When a pair of extreme points... When the difference is greater than 5, it is considered a pole pair with a significant change. For example, if the peak is 18.3 and the trough is 11.8, the difference is 6.5, and this pair is retained as a significant pole pair. Then, the time interval between each pair of poles is counted. If the interval of a certain pair is less than 0.04 seconds or greater than 0.5 seconds, it indicates that there is a data jump. For example, if the interval between a trough and the previous peak is 0.02 seconds, this interval is marked as a sudden change segment. After excluding the sudden change segments according to the time axis, the remaining data is re-sorted. For points with abnormal fluctuations, the average of the five data points before and after the point is used to replace it. If the point is 20.2, the average of the five points before is 18.7, and the average of the five points after is 18.9, then the point is replaced with 18.8 to form a data curve with a smoother fluctuation trend. In the final retained pole data, each point contains the time position (e.g., 0.24 seconds), fluctuation type (peak or trough), and numerical amplitude (e.g., 16.8), forming a structured pole record sequence, which is used as screening data for subsequent processing.

[0065] S102: Based on the pole structure screening data, determine the distribution of each group of poles in each period, analyze the changes in the time position of poles between adjacent periods, screen poles with stable distribution, optimize the order of occurrence and distribution of poles within the period, retain the characteristic points of regular distribution, and obtain a set of periodically stable nodes.

[0066] For each cycle (e.g., each cycle is 0.8 seconds long), the time points at which the poles appear are read one by one and converted into percentage positions within that cycle. For example, if a pole appears at the 0.16-second mark, it is at the 20% position. The relative positions of this pole over 10 consecutive cycles are listed, such as 20.1%, 20.4%, 19.9%, 20.2%, 20.5%, 19.7%, 20.0%, 20.3%, 20.6%, and 19.8%. The maximum offset range of the calculated position is 0.9%, which is lower than the set allowable offset standard of 5%. Therefore, this pole distribution is stable and is retained as a stable point. Further checks are made to ensure that the arrangement order of the poles is consistent in each cycle. For example, if the order is ABCD in one cycle, but AB remains in another cycle... If the CD pattern is consistent, then the data for that period is discarded if the period is CBDA. The data for the remaining periods are then compared in segments, dividing the period into five time periods, for example, 0–0.16 seconds, 0.16–0.32 seconds, 0.32–0.48 seconds, 0.48–0.64 seconds, and 0.64–0.80 seconds. The extreme points are observed to see if they are concentrated in a certain segment. If a certain extreme point appears in every period between 0.16 and 0.32 seconds, it is marked as a regular distribution node for that segment. For such nodes, the period number, actual time (e.g., 0.24 seconds), relative position (e.g., 30%), and type (e.g., peak) are recorded. All such nodes are integrated to generate a set of periodically stable nodes, which is used to determine the subsequent distribution density and persistence.

[0067] S103: Based on the set of periodically stable nodes, analyze the distribution density of each node within the period, determine the change in the time interval between nodes, and select a node group with continuous distribution and reasonable density within the period to obtain the gait cycle feature set.

[0068] The system counts the number of stable nodes that appear in each complete cycle. For example, if 12 stable nodes are detected in a cycle, the cycle is divided into six equal time periods, each 0.13 seconds long. The number of nodes in each period is counted. If the first period has 2 nodes, the second has 1, the third has 3, the fourth has 2, the fifth has 2, and the sixth has 2, then the density distribution is reasonable. If a period has no nodes or more than 5 nodes, then the cycle is considered to have an abnormal density. The system then continues to analyze the time interval between any two adjacent nodes. If a node appears at 0.12 seconds and the next appears at 0.28 seconds, then the interval is 0.16 seconds. If the interval value falls within a set reasonable range (e.g., 0.08 to 0.13 seconds), then the system is considered to have a reasonable density distribution. If the interval is within 32 seconds, it is retained. If it is less than 0.05 seconds or greater than 0.4 seconds, it is judged as an abnormal interval and the corresponding node is removed. The final set of nodes must meet the following requirements: uniform distribution within the cycle, single-segment density within a reasonable range, and moderate time difference between adjacent nodes. The information in such a set of nodes includes the number of each node, the specific time point in the cycle, the percentage position, and whether it is a peak or a valley. For example, a set of nodes is located at 0.12 seconds, 0.28 seconds, 0.43 seconds, and 0.58 seconds, respectively, which are 15%, 35%, 54%, and 72%, forming a gait cycle feature set with cycle integrity, density continuity, and structural stability.

[0069] like Figure 3 As shown, the specific steps for obtaining the node correction parameter set are as follows:

[0070] S201: Based on the gait cycle feature set, analyze cadence, weight and stride data, determine the direction of change of each parameter in each cycle, compare the trend of change in adjacent cycles, identify segments that show consistent trend, and obtain the set of linked trend intervals by comparing the fluctuation characteristics of each parameter.

[0071] Step frequency, weight, and stride length data are read sequentially for each cycle. Step frequency is measured in steps per second (e.g., 1.6Hz in cycle 1, 1.8Hz in cycle 2, and 1.7Hz in cycle 3). Stride length is calculated as the distance traveled per step (e.g., 0.58m, 0.62m, and 0.60m respectively). Weight is calculated using the static weight value estimated from ground reaction force at the time of data collection. If the weights in the three cycles are 72.1kg, 71.8kg, and 71.9kg respectively, these three parameters are plotted as time series curves. The direction of change for each parameter within adjacent cycles is analyzed. The method is to subtract the previous cycle from the next cycle's value. If the difference is greater than zero, it is considered an upward trend; less than zero, a downward trend; and equal, a stable trend. For example, if the step frequency increases by 0.2Hz from cycle 1 to cycle 2 and decreases by 0.1Hz from cycle 2 to cycle 3, it indicates an upward and then downward trend. After the changing trends of the parameters are assigned to the corresponding labels, the trends of the three parameters in adjacent periods are compared. If all three are rising, falling, or stable, the period is determined to be a consistent trend segment. For example, if step frequency, stride length, and weight decrease in period 1 to period 2, it is judged as a partial trend consistency. If only step frequency and stride length are consistent, it is further judged as local linkage. The segments with a trend consistency ratio of more than 80% in all periods are selected as preliminary candidate linkage intervals. Then, the specific fluctuation range of the three parameters in each period is analyzed, and the ratio of the change value in each period to the current average value is calculated as the fluctuation index. For example, if the average stride length is 0.60 meters and it is 0.58 meters in period 1, the fluctuation ratio is 3.3%. If this ratio exceeds 10%, it is marked as a period of significant fluctuation. Then, the intersection comparison is performed with the trend consistent segments, and the periods that satisfy both trend consistency and significant fluctuation are selected as the set of linkage trend intervals.

[0072] S202: Based on the set of linked trend intervals, compare the impact of cadence trends and weight fluctuations on the distribution of feature points, analyze the changes in the sampling positions of feature points within the difference period, determine the time distribution differences caused by cadence changes and weight fluctuations, screen the optimal combination of parameters that affect the distribution of feature points, and obtain the feature distribution influencing factors.

[0073] Multiple gait cycles corresponding to each interval are selected. Data on cadence and weight changes within each interval, along with the positional sequence of gait cycle feature points within each cycle, are retrieved. The upward or downward trend of cadence is compared with the difference in the position of the feature points on the cycle axis. For example, if the cadence is 1.6Hz in cycle 1 and 1.8Hz in cycle 2, and a certain feature point appears at the 20th percentile in cycle 1 but at the 18th percentile in cycle 2, it is determined that the feature point appears earlier under high cadence. The relative time difference of the feature point before and after the cycle is recorded as -2%. If, simultaneously, the weight decreases from 72kg to 70.8kg within the same interval, it is observed whether the same feature point further advances or delays in other cycles. For example, if the point's position advances from 18% to 16.5%, the difference is further recorded. This process is used to identify cycles with small cadence changes but large weight changes. During the process, it is determined whether the feature point position shifts more sensitively. If so, it is assumed that weight has a significant impact on the feature point position. The same action is then performed on multiple feature points, and the positional change of each feature point on the time axis is calculated under changes in cadence and weight. If a certain parameter (such as cadence) causes more than 80% of the feature points to change in the same direction with an amplitude exceeding 5%, then this parameter is determined to be the main influencing factor. Then, stride length is used as a supplementary judgment dimension to analyze whether the stride length change cycle causes the key feature point position to be delayed or advanced by more than 5%. If this influence is greater than weight, then weight is replaced as the influencing factor. The influence range, number of affected feature points, and distribution location of all candidate parameters are compared, and the one that causes the most and largest amplitude of key feature point position changes is selected as the optimal influence combination in this cycle, and the output is used as the feature distribution influence factor.

[0074] S203: Based on the characteristic distribution influencing factors, adjust the corresponding characteristic point interval screening criteria, optimize the attribution judgment of each characteristic point, analyze the attribution structure of the selected characteristic points, judge the rationality of the attribution changes within the period, correct the classification results, and obtain the node correction parameter group.

[0075] The mapping relationship between the corresponding parameters and feature points is read, and the time interval information of all feature points in each cycle within the corresponding time period is extracted. The original screening criteria between feature points are used as the base value for adjustment. For example, the initial standard is that the interval between feature points should not be less than 8% of the cycle time and not more than 35% of the cycle time. If it is found that the feature points are significantly affected by the step frequency and most of them are compressed to a time interval below 6%, the lower limit is adjusted from 8% to 5%. If some feature points have an interval exceeding 40% due to the increase in stride length, the upper limit is raised to 45%. Then, all feature points in each cycle are traversed again, and the actual interval of each pair of adjacent feature points is compared with the adjusted screening criteria. Each point is compared individually. If it is not within the standard range, the point is removed or temporarily not classified. The feature points within the standard are retained to reconstruct the structure. The periodic stability of each feature point in the new classification structure is further checked. The judgment criteria are: if the feature point appears no less than three times in five consecutive periods and the classification group is consistent, the classification is considered reasonable. If the feature point frequently crosses groups or has a sudden change in classification during the classification process, a correction operation is performed to classify it into the group with the most occurrences, or to move it closer to the feature group with the smallest fluctuations in position trend. The corrected classification group of each feature point, the upper and lower limits of the adjusted time interval, and its offset value from the original position are output to form a node correction parameter group.

[0076] like Figure 4 As shown, the specific steps for obtaining the muscle strength change trend indicator are as follows:

[0077] S301: Based on the node-corrected parameter set, analyze the knee flexor and extensor strength and ankle dorsiflexor strength, determine the continuous changes in muscle strength data within each time period, compare the upward and downward trends of different time periods, filter continuous data within the same trend interval, and establish a continuous muscle strength grouping sequence.

[0078] Extract the time interval corresponding to each corrected feature point in a continuous gait cycle, and retrieve the muscle strength data corresponding to each time interval. Knee flexion and extension muscle strength data are recorded in Newtons (N). For example, if a subject's knee flexion and extension muscle strength in cycle 5 is 135, 138, 141, 143, 146 (in N), with a sampling interval of 0.5 seconds, this indicates a continuous increase in muscle strength. If the muscle strength recorded in the next interval is 146, 145, 144, 143, 141, it is considered a continuous decrease. For each data segment, compare the direction of change between every two data points chronologically. If the data increases continuously, it is considered an upward trend; if it decreases continuously, it is considered a downward trend. If there are jumps in the middle, such as 135, 138, 134, 139, 142, it is considered a discontinuous change segment. This applies to the overall muscle strength... The above trends are scanned segment by segment along the sampling time axis. Data sequences with consistent fluctuation directions within any five consecutive data points are selected. For example, if the trend is upward between 0 and 2.5 seconds and downward between 2.5 and 5.0 seconds, these two time segments are retained, and their muscle strength data are identified as continuous data within the same trend. The same operation is performed on the ankle dorsiflexion muscle strength. For example, if the sampled data is 90, 93, 95, 96, and 98 (in N), it is judged as an upward trend. If the next segment is 98, 97, 96, 95, and 93, it is a downward trend. Valid segments within the same trend are selected, and segments with abrupt changes and frequent trend reversals are removed. Only segments where the muscle strength data changes significantly within the continuous trend are retained. The start time, end time, direction of change, and joint type of the segment are identified and integrated into a continuous muscle strength group sequence.

[0079] S302: Based on the continuous muscle strength grouping sequence, calculate the rate of change of muscle strength in each group, analyze the fluctuation amplitude within each group, determine the difference in rate between groups, identify groups with stable rate of change, remove unstable groups, uniformly label the remaining groups, and obtain a stable rate feature sequence.

[0080] Extract the time and values ​​of the first and last sampling points in each muscle strength group, and calculate the change per unit time. For example, if a certain ascending group increases knee joint muscle strength from 135N to 146N in 2.5 seconds, its rate of change is 4.4N per second. After calculating the rate for each group, a rate sequence is formed. Then, calculate the fluctuation amplitude within each group, that is, calculate the difference between the maximum and minimum values ​​within that group. For example, if a group of muscle strength data is 130, 133, 131, 135, 136, the fluctuation amplitude is 6N. According to the statistical standard of muscle strength fluctuation in the entire dataset, most muscle strength fluctuations are concentrated between 3N and 7N. If the fluctuation amplitude of a group exceeds 10N, it is marked as abnormal fluctuation. Based on the constant group, the difference in the rate of change between adjacent groups is compared. For example, if the first group is 4.4 N / s and the second group is 4.6 N / s, the difference is 0.2 N / s, which is less than the stability threshold of 3 N / s. The two groups are considered to have the same rate. If the difference in the rate between the two groups exceeds 5 N / s, it is considered a rate mutation. Such mutation groups are removed, and only groups with stable rates of change are retained. Then, all stable groups are numbered and uniformly identified by their joint type, trend direction, time start and end points, and rate of change, forming a stable rate feature sequence. In the example, the upward trend rate of the knee joint is identified as P-KNEE-UP-4.4, which represents the sequence segment with an upward rate of 4.4 N / s for the knee joint.

[0081] S303: Based on the stable rate feature sequence, determine the time order of each group, analyze data segments with consistent trends, filter the same type of change intervals, number the trend types of the groups, establish the correspondence between muscle strength trend labels and time periods, and obtain muscle strength change trend identifiers.

[0082] To analyze data segments with consistent trends, use the following formula:

[0083] ;

[0084] Calculate trend consistency eigenvalues By filtering out variation intervals of the same type, assigning number to the trend types of the groups, establishing a correspondence between muscle strength trend labels and time periods, and obtaining muscle strength change trend identifiers, among which... This represents the number of data points contained in the current group. Representing the The stable rate characteristic value of each data point This represents the average value of the stable rate characteristic of all data points within the current group. Representing the The time interval between each data point and its adjacent data points;

[0085] Part 1 (Variation Consistency Detection) This section calculates the deviation of each data point's rate value from the average rate, and takes into account the effect of time intervals, obtaining an overall fluctuation "intensity value" in square root form, reflecting the overall magnitude of the trend deviation among points within the group.

[0086] Part Two (Overall Trend Level) This part is equivalent to the time-weighted average rate, which measures the overall trend level of the entire group over time.

[0087] Adding the two parts together yields the trend consistency characteristic value. It includes both local trend volatility (whether it deviates from the average) and the overall trend level (average trend performance), therefore... It can be used as a composite indicator to determine whether the trend of the data segments in a group is "uniform, stable or consistent" over time, and to measure whether the overall trend of muscle strength rate characteristics within a certain time group is consistent, serving as a numerical basis for judgment and screening.

[0088] Grouping was performed on continuously collected lower limb muscle strength data within a fixed time window, and the steady-state velocity feature value was analyzed for each data point. and its adjacent time intervals After statistical and normalization processing, calculations are performed. During implementation, data parsing is first performed on any given data segment. Let's assume this segment contains 5 valid data points, and the corresponding stable rate characteristic value is... The unit is The corresponding time interval is The unit is The average steady-state rate characteristic value during this time period is calculated as follows:

[0089] ;

[0090] Then, the standardized deviation term is calculated for each data point, i.e. After summing and taking the square root, the normalized results show that the deviations for each term are 0.0289, 0.0127, 0.034, 0.006, and 0.016, respectively, with a corresponding average of 0.0195. The square root value is:

[0091] ;

[0092] Next, the weighted rate average term is calculated, and the sum of the products of each steady rate and its corresponding time interval is calculated as follows:

[0093] ;

[0094] The sum of all time intervals is:

[0095] ;

[0096] Therefore, the weighted average rate is:

[0097] ;

[0098] The two values ​​are 0.1396 and 0.2068, respectively. Adding them together forms the trend consistency characteristic value. The characteristic value of trend consistency in this paragraph is:

[0099] ;

[0100] According to the established trend classification criteria:

[0101] like The term "stable" indicates that the data fluctuates very little and the trend of change is basically flat.

[0102] like The term "fluctuating trend type" indicates that the trend direction changes frequently but still exhibits certain regularity.

[0103] like The term "unstable and violently volatile" indicates significant trend disturbances and a lack of stable direction.

[0104] Therefore, the current calculation results It belongs to the "fluctuation trend type" segment, and the corresponding trend type number is 2. This number serves as the identification label for the trend of the current grouped data segment. By binding it with the corresponding time interval (such as 25s to 30s), the muscle strength trend label "Trend 2-25s-30s" is further generated, which constitutes the muscle strength change trend identifier.

[0105] In the formula, This indicates the number of data points within the currently processed data segment; in this example, the value is 5. This indicates that the accumulation operation is performed sequentially on each data point within the data segment. For the first The stable velocity characteristic value of each data point is derived from the joint movement velocity extracted by the real-time muscle force sensor and estimated through accelerometer readings and angle change rate. The arithmetic mean of all stable rate values ​​within this data segment is calculated by directly summing and dividing by... calculate, Indicates the first The time interval between each data point and its preceding adjacent data point is calculated from the sensor timestamp record information.

[0106] like Figure 5 As shown, the specific steps for obtaining the function conversion node set are as follows:

[0107] S401: Based on the trend of muscle strength change, analyze the periodic changes of gait stability score items, determine the fluctuation of stability parameters in each cycle, filter the time segments where the fluctuation direction changes, detect the cycle number corresponding to each change segment, and obtain the stability change interval group.

[0108] Gait cycle sequences corresponding to muscle strength trend time periods are extracted. Gait stability scores for each cycle are read sequentially. These scores can be standardized based on measured values ​​such as gait symmetry, stride length coefficient of variation, and ground reaction force fluctuation rate. For example, if the standardized stability scores for cycles 1–5 are 0.82, 0.85, 0.79, 0.76, and 0.80 respectively, they are recorded as time series. To determine the fluctuation of scores within each cycle, the difference between consecutive cycles is calculated. If the score of a cycle increases by more than 0.04 compared to the previous cycle, it is considered an upward fluctuation; if it decreases by more than 0.04, it is considered a downward fluctuation; and if the change is less than 0.02, it is considered a stable state. The fluctuation direction is categorized into three types: "upward," "downward," or "stable." Then, adjacent cycles are compared... If the direction of the cyclical fluctuation changes, for example, if the direction changes from rising to falling between cycle 2 and cycle 3, it is determined as a trend reversal. Cycle 3 is marked as the point of directional change. Each cycle sequence is continuously scanned and the location of the directional change point is determined. Trend reversal points are aggregated and statistically analyzed in units of three cycles. If the interval between two consecutive directional change points is less than three cycles within a certain period, it is considered a frequently fluctuating segment. The start and end cycle numbers of the corresponding segment are recorded. For example, if the direction changes from falling to rising and then back to falling between cycle 3 and cycle 6, then cycle 3–6 constitutes a stable change interval. All segments that meet the directional change condition are sorted and numbered according to time. Each segment is marked with the start cycle, end cycle, number of reversals, and change type sequence, forming a stable change interval group.

[0109] S402: Based on the stationary change interval group, analyze the time series data of knee joint activity amplitude within the corresponding period, determine the maximum range of activity amplitude within each period, identify the key time point when the direction of activity amplitude change changes, and obtain the knee joint activity turning point.

[0110] For each cycle number corresponding to a change segment, time-series data of knee flexion-extension angles are extracted one by one. The range of knee joint activity within each cycle is determined sequentially. This is achieved by calculating the difference between the maximum and minimum flexion-extension angles within that cycle as the range of activity for that cycle. For example, in cycle 4, the maximum flexion-extension angle is 58 degrees and the minimum is 12 degrees, so the range of activity is 46 degrees. Further, the change in range is time-localized, identifying the time points of occurrence for the maximum and minimum angles, and calculating their relative percentage within the cycle. For example, if the maximum value is at 30% of the cycle and the minimum value is at 70%, then this segment is the main activity area. Subsequently, the trend of range change is checked in multiple consecutive cycles to see if there is a directional reversal. That is, if the first two... If the range of motion continuously decreases in one cycle and begins to rise in the next cycle, then that cycle is determined to be the turning point of the direction of change. At the same time, the direction of change of the maximum and minimum angle values ​​is recorded. For example, if the angle value drops from 58 degrees to 50 degrees and then rises to 52 degrees, it indicates that cycle 5 is the turning point. Further screening is performed on the key time points when this turning point occurs. The specific data point index and corresponding time position of the curve trend direction changing from downward to upward are located in the original angle curve. For example, if the angle value starts to rise from a continuously decreasing 48 degrees to 50 degrees at 2.8 seconds, this point is recorded as the activity turning point. Each type of turning point is archived according to the cycle number, joint type, direction of change, key time point position and corresponding angle value to form a set of knee joint activity turning point data.

[0111] S403: Based on the knee joint activity transition nodes, screen nodes where gait stability and knee joint range of motion change simultaneously, uniformly encode the synchronously changing nodes, establish corresponding transition types and time-series features, and obtain a set of functional transition nodes;

[0112] Compare whether each node occurs within the same cycle or time period as the gait stability parameter change point in the original timeline. Call the timestamp of each directional turning point in the stability change interval group and compare it with the timestamp of the knee joint turning point. If the time difference between the two is no more than 0.4 seconds, it is considered a synchronous change node, marked as a "synchronous point," and a synchronization status identifier is written into the record. Continue to check whether all node pairs have the same trend change type. For example, if knee joint activity changes from decreasing to increasing, and gait stability score changes from decreasing to increasing, it is a same-direction change type. If the trend is opposite, it is recorded as a reverse change type. Assign a unique number to each type of synchronous point, such as "SYNC01," and mark it as a "same-direction" or "reverse" conversion type. Record its start and end times in the complete timeline, the muscle strength trend number involved, the stability parameter number, and the knee joint activity sequence number. Establish a structured record entry for this synchronous node. After summarizing all synchronous change nodes, a complete conversion type and temporal feature mapping table is formed as a functional conversion node set.

[0113] like Figure 6 As shown, the specific steps for obtaining the proportion series of related indicators are as follows:

[0114] S501: Based on the functional transition node set, determine the changing trends of gait stability and knee joint flexibility in each cycle, compare the parameter fluctuations of the two in the same cycle, filter the cycle numbers with the same direction of gait stability and knee joint flexibility changes, and obtain the cycle linkage corresponding sequence.

[0115] The cycle number corresponding to each node is read one by one, and the complete time series of gait stability parameters and knee joint flexibility indices within that cycle are extracted. Gait stability parameters are calculated using the standardized step length variation coefficient and support time volatility. For example, if the step length variation is 0.11 and the support time volatility is 0.14 in a certain cycle, the normalized score is 0.85. Knee joint flexibility is calculated using the difference between the maximum and minimum flexion and extension angles. For example, if the maximum angle is 58° and the minimum is 12°, the range of motion is 46°. The temporal trends of stability and flexibility are calculated for consecutive cycles. The judgment method is based on the sign of the numerical difference between adjacent periods. For example, if the stability score increases from 0.85 to 0.88, it is considered an increase; if the knee joint range of motion decreases from 46° to 42°, it is considered a decrease. The direction of change of the two parameters in each period is recorded, and the trends of the two parameters are compared side by side. If the directions are consistent, they are marked as "linked" in the results table; if they are inconsistent, they are marked as "non-linked". Continue to scan the periods corresponding to all functional transition nodes, filter out the period numbers with consistent directions of change in stability and flexibility, record the numbers, and integrate them into the corresponding sequence of periodic linkage.

[0116] S502: Based on the corresponding sequence of periodic linkage, combined with the time series data of muscle strength recovery markers in each period, analyze the synchronicity of the temporal distribution of muscle strength changes with gait stability and knee joint flexibility, determine the linkage characteristics of each parameter in each period, screen the periods that show synchronous fluctuations, and obtain the parameter distribution ratio group.

[0117] The system retrieves the corresponding muscle strength recovery marker information for each cycle. This data is provided by previously recorded muscle strength change trend labels, including key fields such as trend direction, start and end times, and rate of change. First, the muscle strength trend labels are aligned with the stability and knee joint mobility data according to the cycle number. The start and end times of each parameter are calibrated along the time axis to determine whether each parameter is within the same time period. For example, if the muscle strength trend is marked as an upward trend within a certain cycle, with the start time being from the 8th to the 12th second, and if the stability score and knee joint mobility also change within this time period and are in the same direction as the muscle strength trend, then this cycle is marked as a "synchronous fluctuation cycle." The system continues to determine the synchronicity of the time distribution and sets a synchronous time difference threshold. The value is 0.5 seconds, meaning the difference between the start and end times of the parameters must not exceed this value. If this condition is met, the time distribution is considered synchronized. The number of cycles in which gait stability, knee joint range of motion, and muscle strength markers fluctuate synchronously is counted sequentially, and the proportion of each cycle to the total number of cycles is calculated. For example, if 9 cycles out of 20 cycles meet the three synchronization conditions, the recording proportion is 45%. The synchronization status of each cycle is recorded as "yes" or "no", and the direction of change of the three parameters is recorded as "rising", "falling", or "stable". A parameter distribution proportion group is generated, which includes the cycle number, synchronization status marker, trend of the three parameters, start and end time, and synchronization judgment logic field.

[0118] S503: Based on the parameter distribution ratio group, compare the linkage ratio of each cycle, analyze the parameter grouping characteristics of each recovery stage, determine the structural changes of the linkage ratio with the recovery process, summarize the distribution characteristics of parameter synergistic changes within the difference cycle, and obtain the correlation index ratio sequence.

[0119] The cycle numbers of each cycle with a linkage status of "yes" are extracted sequentially, and the number of times the three parameters show linkage in each cycle is counted. The combination ratio of linkage between gait stability, knee joint flexibility, and muscle strength recovery in that cycle is calculated. For example, if all three parameters increase in cycle 3, it is a three-parameter linkage; if only stability and muscle strength increase in the same direction in cycle 6, it is a two-parameter linkage. Based on this, the linkage level of each cycle is divided into three categories: three-parameter linkage, two-parameter linkage, and single-parameter linkage, labeled as 3L, 2L, and 1L, respectively. The cycles are then divided into recovery stages according to time sequence, for example, the first six cycles are the early stage, the middle eight cycles are the middle stage, and the last six cycles are the late stage. The distribution ratio of each linkage level in different stages is compared. If the proportion of 3L exceeds 50% in the mid-stage and is only 20% in the early stage, it is recorded as a stage-specific proportion difference. Further, the average linkage level in each recovery stage is calculated, and the frequency and ranking of each parameter participating in linkage in that stage are listed. For example, in the late stage, the frequency of muscle strength participating in linkage is 6 times, stability is 5 times, and knee joint flexibility is 4 times, so muscle strength is the dominant linkage factor. Based on such differences, the parameter synergistic combination in different stages is labeled as a typical linkage type, such as the mid-stage is labeled as "gait-knee synergistic dominant type" and the late stage is labeled as "muscle strength dominant stability type". The structure of synergistic changes of various parameters divided by time stage and their proportion composition are output and organized into a sequence of related index proportions.

[0120] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for real-time monitoring and evaluation of orthopedic rehabilitation process, characterized in that, The method includes: S1: Based on gait biomechanical parameters, analyze the gait frequency sequence, compare the position of each data point in the gait cycle, calculate local maxima and minima, determine the interval and amplitude between poles, screen for abnormal data, and obtain a set of gait cycle features. S2: Based on the gait cycle feature set, determine the changing trends of weight, cadence and stride, compare the influence of cadence and weight changes on feature point distribution, filter time-distance differences, adjust the amplitude filtering criteria, optimize feature point assignment, and obtain node correction parameter set; S3: Based on the node correction parameter group, analyze the continuous changes in knee flexion and extension muscle strength and ankle dorsiflexion muscle strength, calculate the rate of muscle strength change in groups, determine the range of muscle strength fluctuation, screen standard groups, extract time period sequences, and obtain muscle strength change trend indicators; S4: Based on the muscle strength change trend identifier, analyze the gait stability score and knee joint range of motion, use a fixed period window to analyze the data curve, screen the rate direction change segment, record the parameter change nodes, and obtain the functional transition node set; The gait cycle feature set includes gait event nodes, cycle position labels, and feature point indexes; the node correction parameter set includes node adjustment factors, distribution correction labels, and matching interval mappings; the muscle strength change trend identifier includes muscle strength growth segments, change rate labels, and trend classification numbers; and the functional transition node set includes transition interval numbers, functional transition types, and temporal feature labels. The gait biomechanical parameters refer to quantitative data reflecting human movement and mechanical characteristics during walking, including ground reaction force, joint angles, lower limb acceleration, and plantar pressure. The gait cycle refers to the time or data interval experienced by a complete gait. The continuous change refers to the dynamic change of the collected muscle strength parameters over time, representing the process of muscle strength change. The muscle strength fluctuation range refers to the amplitude or fluctuation range of muscle strength within the target range. The extraction of time period sequence refers to serializing the time intervals corresponding to the groups that meet the conditions.

2. The method for real-time monitoring and evaluation of orthopedic rehabilitation process according to claim 1, characterized in that, The specific steps for obtaining the gait cycle feature set are as follows: S101: Based on gait biomechanical parameters, compare the time position of each data point in the gait cycle, determine the trend of data change within the cycle, calculate the peaks and troughs between adjacent data points, screen out pole pairs with prominent changes, optimize the continuity of the data sequence, remove abrupt fragments, and obtain pole structure screening data. S102: Based on the pole structure screening data, determine the distribution of each group of poles in each period, analyze the changes in the time position of poles between adjacent periods, screen poles with stable distribution, optimize the order of occurrence and distribution of poles within the period, retain the characteristic points of regular distribution, and obtain a set of periodically stable nodes. S103: Based on the set of periodically stable nodes, analyze the distribution density of each node within the period, determine the changes in the time interval between nodes, and select a node group with continuous distribution and reasonable density within the period to obtain a gait cycle feature set.

3. The method for real-time monitoring and evaluation of orthopedic rehabilitation process according to claim 1, characterized in that, The specific steps for obtaining the node correction parameter set are as follows: S201: Based on the gait cycle feature set, analyze the cadence, weight and stride data, determine the direction of change of each parameter in each cycle, compare the trend of change of adjacent cycles, identify the segments that show a consistent trend of change, and obtain the set of linkage trend intervals by comparing the fluctuation characteristics of each parameter. S202: Based on the aforementioned set of linkage trend intervals, compare the impact of cadence trends and weight fluctuations on the distribution of feature points, analyze the changes in the sampling positions of feature points within the difference period, determine the time distribution differences caused by cadence changes and weight fluctuations, screen the optimal combination of parameters that affect the distribution of feature points, and obtain the feature distribution influencing factor. S203: Based on the characteristic distribution influencing factor, adjust the corresponding feature point interval screening criteria, optimize the attribution judgment of each feature point, analyze the attribution structure of the selected feature points, judge the rationality of the attribution changes within the period, correct the classification results, and obtain the node correction parameter group.

4. The method for real-time monitoring and evaluation of orthopedic rehabilitation process according to claim 1, characterized in that, The specific steps for obtaining the muscle strength change trend indicator are as follows: S301: Based on the node correction parameter group, analyze the knee joint flexor and extensor muscle strength and the ankle joint dorsiflexor muscle strength, determine the continuous changes of muscle strength data in each time period, compare the rising and falling trends of different time periods, filter continuous data in the same trend interval, and establish a continuous muscle strength grouping sequence. S302: Based on the continuous muscle strength grouping sequence, calculate the rate of change of muscle strength in each group, analyze the fluctuation amplitude within each group, determine the difference in rate between groups, identify groups with stable rate of change, remove unstable groups, uniformly identify the remaining groups, and obtain a stable rate feature sequence. S303: Based on the stable rate feature sequence, determine the time order of each group, analyze data segments with consistent trends, filter the same type of change intervals, number the trend types of the groups, establish the correspondence between muscle strength trend labels and time periods, and obtain muscle strength change trend identifiers.

5. The method for real-time monitoring and evaluation of orthopedic rehabilitation process according to claim 1, characterized in that, The specific steps for obtaining the set of function conversion nodes are as follows: S401: Based on the muscle strength change trend identifier, analyze the periodic changes of the gait stability score item, determine the fluctuation of stability parameters in each period, filter the time segments where the fluctuation direction changes, detect the period number corresponding to each change segment, and obtain the stability change interval group. S402: Based on the set of stable change intervals, analyze the time series data of knee joint activity amplitude within the corresponding period, determine the maximum range of activity amplitude within each period, identify the key time point when the direction of activity amplitude change changes in the direction of change, and obtain the knee joint activity turning point. S403: Based on the knee joint activity transition nodes, screen nodes where gait stability and knee joint range of motion change simultaneously, uniformly encode the synchronously changing nodes, establish corresponding transition types and time sequence features, and obtain a set of functional transition nodes.

6. The method for real-time monitoring and evaluation of orthopedic rehabilitation process according to claim 1, characterized in that, The method further includes: S5: Based on the set of functional transition nodes, analyze the correspondence between gait stability performance and knee joint flexibility changes, combine the synchronicity of muscle strength recovery time sequence, determine the linkage changes of each parameter, sort out the functional recovery path, and obtain the correlation index ratio sequence. The associated indicator ratio sequence includes linkage parameter grouping, ratio allocation labels, and recovery phase structure.

7. The method for real-time monitoring and evaluation of orthopedic rehabilitation process according to claim 6, characterized in that, The specific steps for obtaining the correlation index ratio sequence are as follows: S501: Based on the set of functional conversion nodes, determine the changing trends of gait stability and knee joint flexibility in each cycle, compare the parameter fluctuations of the two in the same cycle, filter the cycle numbers with the same direction of gait stability and knee joint flexibility changes, and obtain the cycle linkage corresponding sequence. S502: Based on the corresponding sequence of the cycle linkage, combined with the time series data of the muscle strength recovery marker in each cycle, analyze the synchronicity of the time distribution of muscle strength changes with gait stability and knee joint flexibility, determine the linkage characteristics of each parameter in each cycle, screen the cycles that show synchronous fluctuations, and obtain the parameter distribution ratio group. S503: Based on the parameter distribution ratio group, compare the linkage ratio of each cycle, analyze the parameter grouping characteristics of each recovery stage, determine the structural changes of the linkage ratio with the recovery process, summarize the distribution characteristics of the coordinated changes of parameters within the difference cycle, and obtain the correlation index ratio sequence.

Images