A health information management method for rehabilitation training

Abstract

The present invention relates to the technical field of rehabilitation training, and particularly relates to a health information management method for rehabilitation training. This method determines the deformed segments with non-standard movements by the morphological changes of different movement segments divided from the movement performance data; analyzes the correlation between the changes in the physiological index data of the body and the morphological changes of the movement segments for the deformed segments, quantifies different factors causing deformation through the correlation analysis, and determines the load segments; obtains the cumulative intensity from the pre-sequence deformed movement conditions of the load segments and the recovery trend of the physiological index data; adjusts the rehabilitation training of the patient from different classification results or based on the non-load conditions and the cumulative intensity of the load segments. The present invention combines the analysis of the correlation between the movement deformation conditions and the changes in the body's physiological index during the patient's rehabilitation training, classifies and analyzes the patient's movement state, and makes the adjustment of the rehabilitation training more accurate and reliable.

Description

Technical Field

[0001] The present invention relates to the technical field of rehabilitation training, and particularly relates to a health information management method for rehabilitation training. Background Art

[0002] Rehabilitation training refers to physical activities that are beneficial to the recovery or improvement of functions after injury. Except for severe injuries that require rest and treatment, general injuries do not necessarily require complete cessation of physical exercises. Appropriate and scientific physical exercises play a positive role in the rapid healing of injuries and the promotion of functional recovery. When patients are undergoing rehabilitation training, doctors need to obtain the patients' health information to formulate rehabilitation training plans. Through the health information, the recovery status of the patients and the effects of rehabilitation training can be determined. When designing the rehabilitation training plans for patients based on the obtained health information, artificial intelligence can be used to assist doctors in monitoring the patients' training. When doctors are formulating rehabilitation training, they generally provide guidance based on their own experience combined with the patients' information. However, through artificial intelligence, more scientific training can be carried out, and at the same time, the difficulty of doctors in guiding rehabilitation training can be reduced.

[0003] When guiding the rehabilitation training plan for patients, patients may perform actions inaccurately, resulting in an inability to complete the designated training plan well. However, the reasons for patients' inaccurate actions may not only be poor compliance, but also be caused by fatigue or heavy load. If the adjustment of different training plans required in various situations of non-standard actions is not considered, it may lead to slow subsequent rehabilitation of patients. Summary of the Invention

[0004] In order to solve the technical problem in the prior art that the adjustment of different training plans required in various situations of non-standard actions is not considered, which may lead to slow subsequent rehabilitation of patients, the purpose of the present invention is to provide a health information management method for rehabilitation training, and the specific technical solution adopted is as follows:

[0005] The present invention provides a health information management method for rehabilitation training, and the method includes:

[0006] Obtaining physiological index data and motor performance data at different times during the rehabilitation training of patients;

[0007] Segmenting the motor performance data of patients to obtain different motion segments; analyzing the morphological matching of the data in each motion segment with the standard motion template to obtain the morphological fluctuation index of each motion segment; screening out the deformed segments based on the size of the morphological fluctuation index of the motion segments;

[0008] Based on the numerical distribution and change rate of each physiological index data at each moment, obtain the fluctuation characteristic index of each physiological index data at each moment; on each deformation segment, analyze the correlation between the morphological fluctuation index and the fluctuation characteristic index of each physiological index data, and obtain the correlation index between each deformation segment and each physiological index data;

[0009] Based on the magnitudes of the correlation indices between each deformation segment and all physiological index data, determine the load segment; according to the time difference between the load segment and the previous deformation segment and the number of previous motion segments, as well as the correlation index between the load segment and each physiological index data combined with the changing trend in time series, obtain the cumulative intensity of the load segment;

[0010] Based on the cumulative intensity magnitude of the load segment in the deformation segments and the number of non-load segments, adjust the patient's rehabilitation training.

[0011] Furthermore, the method for obtaining the morphological fluctuation index includes:

[0012] For any one motion segment, match the motion performance data on this motion segment with the standard motion template using the DTW algorithm to obtain a number of matching pairs;

[0013] In each matching pair, take the data value difference between the mean value of the motion performance data on this motion segment and the mean value of the standard motion template as the deviation degree of each matching pair;

[0014] Normalize the sum value of the deviation degrees of all matching pairs on this motion segment to obtain the morphological fluctuation index of this motion segment.

[0015] Furthermore, the method for obtaining the deformation segment includes:

[0016] When the value after normalization of the morphological fluctuation index is greater than the preset deformation threshold, take the corresponding motion segment as the deformation segment.

[0017] Furthermore, the method for obtaining the fluctuation characteristic index includes:

[0018] For any kind of physiological index data, perform curve fitting on this physiological index data in time series to obtain the index curve;

[0019] Obtain the slope at each moment on the index curve; take the slope difference between each moment and the previous moment in time series as the change difference degree;

[0020] Take the product of the numerical value of this physiological index data at each moment and the change difference degree as the fluctuation characteristic index of this physiological index data at each moment.

[0021] Furthermore, the method for obtaining the correlation index includes:

[0022] For any deformation segment, each type of physiological index data is sequentially used as the analysis data; the mean value of the fluctuation characteristic indexes of the analysis data at all times on this deformation segment is normalized to obtain the physiological fluctuation degree of the analysis data on this deformation segment.

[0023] The difference between the physiological fluctuation degree of the analysis data on this deformation segment and the morphological fluctuation index is subjected to negative correlation mapping to obtain the correlation index between this deformation segment and the analysis data.

[0024] Further, the determination method of the load segment includes:

[0025] For each deformation segment, when the correlation index between the deformation segment and the physiological index data is greater than the preset correlation threshold, the deformation segment is used as the load segment.

[0026] Further, the acquisition method of the cumulative intensity includes:

[0027] For any load segment, the total number of all previous motion segments is counted as the cumulative degree of times of this load segment;

[0028] The time interval between this load segment and the previous load segment is subjected to negative correlation mapping to obtain the cumulative deformation degree of this load segment;

[0029] Combined with the correlation index between this load segment and each type of physiological index data and the change trend of the physiological index data, the cumulative degree of physical signs of this load segment is obtained;

[0030] Combined with the cumulative degree of times, the cumulative deformation degree and the cumulative degree of physical signs of this load segment, the cumulative intensity of this load segment is obtained.

[0031] Further, the acquisition method of the cumulative degree of physical signs includes:

[0032] For any type of physiological index data, after curve fitting the physiological index data on this load segment, the time series decomposition algorithm is used to obtain the trend term of the physiological index data; the mean value of all trend values in the trend term is subjected to negative correlation mapping to obtain the trend smoothness of the physiological index data.

[0033] The product of the correlation index between the physiological index data and this load segment and the trend smoothness is used as the trend probability index between the physiological index data and this load segment;

[0034] The accumulated value of the trend probability indexes between this load segment and all physiological index data is normalized to obtain the cumulative degree of physical signs of this load segment.

[0035] Further, adjusting the rehabilitation training of the patient based on the cumulative intensity magnitude of the load segment in the deformation segment and the number of non-load segments includes:

[0036] Take the proportion of the number of non-load segments in the deformation segmentation as the compliance judgment index; when the compliance judgment index is greater than the preset compliance adjustment threshold, record the current patient's rehabilitation training as action adjustment;

[0037] When the cumulative intensity of all load segments does not exceed the preset fatigue threshold, record the current patient's rehabilitation training as intensity adjustment; otherwise, record the current patient's rehabilitation training as frequency adjustment.

[0038] Further, the method for obtaining the motion segmentation includes: dividing the motion performance data in time series into different motion segments through trajectory segmentation.

[0039] The present invention has the following beneficial effects:

[0040] By the morphological change situation of the motion performance data divided into different motion segments, determine the deformation segments with non-standard actions. For the non-standard deformation segments, analyze the correlation between the change of the physiological index data of the patient's body and the morphological change of the motion segments during rehabilitation training, consider the different stress responses caused by fatigue pain load physiologically and poor compliance when the action is non-standard, quantify different situations after action deformation through correlation analysis, and determine the load segments to classify different influencing factors. Further, reflect the cumulative situation of the patient's body motion intensity from the deformation motion situation before the load segment and the recovery trend of the physiological index data, further classify fatigue and pain, and finally adjust the patient's rehabilitation training from different classification results of the non-load situation and the cumulative intensity of the load segment, so as to improve the accuracy of the proposed auxiliary rehabilitation training plan. The present invention combines the analysis of the correlation between the motion deformation situation and the change of the body physiological index during the patient's rehabilitation training, classifies and analyzes the patient's motion state, and makes the adjustment of the rehabilitation training more accurate and reliable. Description of the Drawings

[0041] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0042] Figure 1 It is a flowchart of a health information management method for rehabilitation training provided by an embodiment of the present invention;

[0043] Figure 2 It is a schematic diagram of different motion segments provided by an embodiment of the present invention. Detailed Embodiments

[0044] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following specifically describes, in conjunction with the accompanying drawings and preferred embodiments, a health information management method for rehabilitation training proposed according to the present invention, including its specific implementation manner, structure, features and effects, as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0046] The following specifically describes the specific solution of a health information management method for rehabilitation training provided by the present invention in conjunction with the accompanying drawings.

[0047] Please refer to Figure 1 , which shows a flowchart of a health information management method for rehabilitation training provided by an embodiment of the present invention. The method mainly includes the following steps:

[0048] S1: Obtain physiological index data and exercise performance data at different time sequences during the patient's rehabilitation training.

[0049] During the patient's rehabilitation training process, wear a device containing sensors to collect data characterizing the degree of exercise during the patient's movement as exercise performance data. For example, for patients who need knee flexion and extension rehabilitation training, they can wear an intelligent knee brace integrated with an accelerometer and a gyroscope, and can further collect the angle conditions of knee flexion and extension to reflect the range quality of the movement as exercise performance data at different time sequences. Other examples include when the shoulder abducts during shoulder rehabilitation exercise, the trajectory deviation of the arm from the vertical position to the horizontal position is collected through an integrated intelligent bracelet, etc.

[0050] The fundamental purpose of the patient during rehabilitation training is to stimulate muscle groups and cardiopulmonary function through specific movements to improve the patient's physical health level. When the patient trains through specific movements, the patient's physical index data will change to a certain extent according to the exercise situation, and the exercise effect can be reflected to a certain extent through the change of the physical index.

[0051] Therefore, during the rehabilitation training process, it is also necessary to collect the patient's physiological index data in real time through sensor devices. The physiological index data at least includes heart rate, heart rate variability, and electromyogram signals, etc. In the embodiment of the present invention, the heart rate (HR) can be collected through an intelligent watch and an electrocardiogram (ECG) monitor, and the data format is usually the number of beats per minute (bpm). For example, the heart rate data format is: [72, 75, 80, 70], indicating the heart rate at consecutive time points.

[0052] Heart rate variability (HRV) can be used to obtain RR interval data using an electrocardiogram (ECG) device. HRV is represented by indicators such as the standard deviation of the heart beat interval (SDNN) and RMSSD. For example, the SDNN value is 32 ms and the RMSSD is 22 ms. Heart rate variability represents the variation of the heart beat interval and is used to evaluate the function of the autonomic nervous system and the body's recovery status.

[0053] Electromyography (EMG) is collected using an EMG sensor. The patch of the EMG sensor is tightly installed on the muscle group that needs rehabilitation training. For example, during knee rehabilitation, the electrodes can be attached to the quadriceps on the front of the thigh or the biceps femoris on the back of the thigh. If the training involves the shoulder, the electrodes can be attached to the deltoid muscle or the rotator cuff muscles. The data is usually represented as a voltage signal (mV). By analyzing the amplitude and frequency of the electromyogram, the activity and fatigue state of the muscle can be judged. For example, the data format of a segment of EMG signal is: [0.3, 0.5, 0.7, 1.2, 1.5], which represents the electrical signal intensity of the muscle at different time points.

[0054] The reason why the exercise effect of the patient is not good during training is most likely due to the poor exercise effect caused by the patient's non-standard movements. And there are also various situations where the patient's movements are not standard. For example, when the patient's compliance is poor, the training will cause movements through the compensation of other muscles. In addition, when the patient is overly fatigued or the pain load is high, it will also cause deformation of the movements. Therefore, subsequent suggestions for assisted rehabilitation training can be proposed through the transmission, integration, and analysis of real-time physiological index data and exercise performance data.

[0055] S2: Segment the exercise performance data of the patient to obtain different exercise segments; analyze the morphological matching of each exercise segment with the standard exercise template to obtain the morphological fluctuation index of each exercise segment; screen out the deformed segments based on the size of the morphological fluctuation index of the exercise segments.

[0056] Patients usually perform repetitive training on a single movement during rehabilitation training. Therefore, registration analysis with the standard is carried out for each movement segment of each exercise to first screen out the time periods that may be deformed and then analyze the possible causes of deformation in the subsequent analysis. First, time period division is performed. In the embodiment of the present invention, the exercise performance data in time series is divided into different exercise segments through trajectory segmentation. It should be noted that trajectory segmentation is a well-known technical means for those skilled in the art and will not be elaborated and limited here. Please refer to Figure 2 , which shows a schematic diagram of different exercise segments provided by an embodiment of the present invention. The horizontal axis is time and the vertical axis is exercise characterization data, showing the situation where the data in multiple exercise segments is mapped to the same coordinate system.

[0057] The degree of fluctuation between the data in each motion segment and the standard template is reflected by the matching situation of the morphology, that is, the degree of non-standardness of the actions in each motion segment. In the embodiments of the present invention, the method for obtaining the morphological fluctuation index includes: for any motion segment, the motion performance data on the motion segment is matched with the standard motion template using the DTW algorithm to obtain a number of matching pairs. In the DTW algorithm, the obtained matching pairs represent the result of the optimal alignment between two time series. The matching pairs are the most similar parts at the time points between the two time series data, which is convenient for data morphology matching analysis. It should be noted that the process of obtaining the matching pairs by the DTW algorithm is a well-known technical means to those skilled in the art and will not be elaborated here.

[0058] Further, in each matching pair, the data value difference between the mean value of the motion performance data on the motion segment and the mean value of the standard motion template is used as the deviation degree of each matching pair. Considering that there are not only one-to-one relationships in the matching pairs, but also one-to-many or many-to-many relationships, the difference analysis is carried out between the data mean value in the motion segment during the matching and the data mean value in the corresponding matching standard template, and the motion performance fluctuation analysis of the whole matching part is carried out.

[0059] Finally, the sum value of the deviation degrees of all matching pairs on the motion segment is normalized to obtain the morphological fluctuation index of the motion segment. When the deviation degree between the data of the motion segment in the matching pair and the standard situation is larger, it indicates that the motion segment has a higher fluctuation relative to the standard template and is more likely to be a deformed segment. It should be noted that normalization is a well-known technical means to those skilled in the art, and the choice of normalization can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0060] Therefore, further screening can be preliminarily carried out through the size of the morphological fluctuation index. In the embodiments of the present invention, the method for obtaining the deformed segment by screening includes: when the value after the morphological fluctuation index is normalized is greater than the preset deformation threshold, the corresponding motion segment is used as the deformed segment. In the embodiments of the present invention, the preset deformation threshold is set to 0.6, and the specific numerical value can be adjusted by the implementer according to the specific implementation scenario and is not limited here.

[0061] The preliminary screening analysis of the existence of non-standard actions is completed, and then the factors of non-standardness are further analyzed in combination with the physiological data.

[0062] S3: According to the numerical distribution and change rate of each physiological index data at each moment, obtain the fluctuation characteristic index of each physiological index data at each moment; on each deformed segment, analyze the correlation between the morphological fluctuation index and the fluctuation characteristic index of each physiological index data to obtain the correlation index between each deformed segment and each physiological index data.

[0063] The fluctuations of different physiological index data have different representative meanings. For example, during training, the heart rate will increase with the increase of exercise intensity. However, patients usually experience pain during training. When severe pain occurs due to increased pain load, the body may trigger a stress response, which may be manifested as a sharp fluctuation or a rapid increase in heart rate. At the same time, when patients experience pain or discomfort due to excessive exercise load, the sympathetic nerve becomes active, resulting in an accelerated heart rate and a sense of fatigue. The pain or fatigue state may lead to a prolonged recovery time of the heart rate, and the heart rate recovers slowly.

[0064] For exercise performance data, if a patient feels pain during a certain movement, the patient may automatically reduce the amplitude of the movement. For example, during knee flexion and extension training, patients who feel pain in the knee may subconsciously reduce the bending angle of the knee joint. At the same time, pain may lead to a slowdown in exercise frequency and unnatural pauses, etc.

[0065] Therefore, for each physiological index data, it is necessary to pay attention to the fluctuation changes and the numerical values. By comprehensively reflecting the instability degree of the physiological index data in terms of numerical value and the degree of previous change at each moment, the fluctuation characteristics of each moment of the physiological index data are reflected. Preferably, in the embodiments of the present invention, the method for obtaining the fluctuation characteristic index of each physiological index data at each moment includes:

[0066] First, for any physiological index data, the physiological index data in time series is curve-fitted to obtain an index curve, which represents the overall change of the data. The slope of each moment on the index curve can be further obtained. By taking the difference in slope between each moment and the previous moment in time series as the change difference degree, it reflects whether there are large fluctuations in the change state of the data in time series, such as a rapid increase, etc. In the embodiments of the present invention, the fitting can be performed by the method of spline interpolation. Curve fitting is a well-known technical means for those skilled in the art and will not be elaborated and limited here.

[0067] Finally, the product of the numerical value of the physiological index data at each moment and the change difference degree is used as the fluctuation characteristic index of the physiological index data at each moment. The size of the numerical value also reflects the physical training state. For example, the maintenance of an elevated heart rate can reflect that the training has produced effects. However, when the heart rate rises too rapidly, it may be caused by excessive exercise volume or a high pain load. Therefore, through the numerical value and the change difference degree, the situation of the body state affected by exercise is reflected.

[0068] Furthermore, in each deformation segment, the correlation between the physiological index data and the exercise performance data of the body can be analyzed and quantified. The correlation analysis and judgment can be used to observe whether the fluctuations of the physiological index data are related to the movements. For example, when the heart rate fluctuates too rapidly, the patient's movements generally also fluctuate, reflecting that the patient's current exercise state is a high-load situation.

[0069] Preferably, in the embodiments of the present invention, the method for obtaining the correlation index includes:

[0070] For any deformation segment, each type of physiological index data is sequentially used as the analysis data, and correlation analysis is performed on each type of physiological index data on the deformation segment in sequence. The mean value of the fluctuation characteristic indexes of the analysis data at all times on the deformation segment is normalized to obtain the physiological fluctuation degree of the analysis data on the deformation segment. By means of the mean value, the fluctuation characteristic indexes corresponding to all times in the time period are comprehensively obtained, and the physiological fluctuation degree is obtained, which represents the degree of high fluctuation change of the analysis data on the deformation segment.

[0071] Furthermore, the difference between the physiological fluctuation degree of the analysis data on the deformation segment and the morphological fluctuation index is negatively correlated and mapped to obtain the correlation index between the deformation segment and the analysis data. When the physiological fluctuation degree is larger, it indicates that the high fluctuation change generated by the physiological index data represented by the analysis data on the deformation segment is more significant. And when the morphological fluctuation index is larger, it indicates that the deformation of the action in the time period is more significant. Therefore, by analyzing the consistency of the significant situations, the correlation degree is reflected. When the difference between the physiological fluctuation degree and the morphological fluctuation index is smaller, it indicates that the situation of continuous significant change is higher, so the correlation index is larger.

[0072] In the embodiments of the present invention, the negative correlation mapping can select the form of negative exponential power. The negative correlation mapping is a well-known technical means to those skilled in the art, and the inverse proportion or negative linear function can also be used, etc., which will not be elaborated and limited here.

[0073] S4: Determine the load segment based on the magnitudes of the correlation indexes between each deformation segment and all physiological index data; according to the time difference between the load segment and the previous deformation segment and the number of the previous motion segments, as well as the correlation indexes between the load segment and each type of physiological index data in combination with the change trend in time sequence, obtain the cumulative intensity of the load segment.

[0074] When the correlation index is larger, it reflects that the degree of simultaneous change in the physiological state during deformation is higher. At this time, it is more likely that the physical state of the patient during training is in a load situation. Therefore, the load segment in a load state is first determined by the magnitude of the correlation index. In the embodiments of the present invention, the method for obtaining the load segment includes:

[0075] For each deformation segment, when there is a correlation index between the deformation segment and the physiological index data greater than the preset correlation threshold, that is, when there is a correlation index between any type of physiological index data and the deformation segment greater than the preset correlation threshold, the deformation segment is used as the load segment.

[0076] The load segment reflects that the patient may deform due to changes in physical condition, which may be caused by fatigue or pain. On the contrary, when the deformation segment is a non-load segment, it reflects that the patient's compliance is poor, the movement does not meet the standard, or the movement does not meet the standard due to distraction, etc. In the embodiment of the present invention, the preset correlation threshold is set to 0.7, and the specific value can be adjusted by the implementer himself / herself and is not limited here.

[0077] For the load segment, there are different factors caused by pain or fatigue, and the corresponding adjustment situations are different. For the pain situation, the training intensity needs to be adjusted to avoid possible secondary injuries. For the fatigue situation, the training frequency can be controlled to adjust the patient's rest rhythm and ensure the training quality.

[0078] Considering that the deformation caused by fatigue factors is more generated by cumulative intensity, it is comprehensively evaluated through the cumulative number of times, the deformation interval, and the overall trend of physiological index data. The more the cumulative number of movement actions, the more it indicates that the current patient has exercised enough, and the movement deformation may be caused by fatigue. The shorter the deformation interval between two adjacent times, the more it indicates that the current patient cannot complete the movement normally due to fatigue. At the same time, the change trend of physiological indicators is also considered. The faster the trend changes, the more likely it is caused by pain. The physiological index changes caused by exercise fatigue are a relatively stable cumulative upward process.

[0079] Preferably, in the embodiment of the present invention, the method for obtaining the cumulative intensity includes:

[0080] For any load segment, count the total number of all previous movement segments before this load segment as the number cumulative degree of this load segment. The more the number of movement segments before the load segment, the more it indicates that the number of action training times is more at this time, and it is very likely that fatigue occurs.

[0081] Furthermore, perform a negative correlation mapping on the time interval between this load segment and the previous load segment to obtain the deformation cumulative degree of this load segment. The smaller the time interval, the more it reflects the situation that it is likely to be unable to maintain the correct posture due to fatigue. It can be understood that for the first load segment without a previous load segment, the frequency approximation situation can be reflected by the time interval with the next load segment. In other embodiments of the present invention, the cumulative intensity analysis of the first load segment can also be excluded and not participate in the subsequent fatigue situation analysis.

[0082] Furthermore, combine the correlation index between this load segment and each physiological index data and the change trend of the physiological index data to obtain the physical sign cumulative degree of this load segment. From the perspective of physiological indexes, the flatter the trend of the physiological index data in the load segment, the more likely it is caused by fatigue, and the analysis credibility is adjusted by combining the correlation index. In the embodiment of the present invention, the method for obtaining the physical sign cumulative degree includes:

[0083] First, for any kind of physiological index data, after curve fitting the physiological index data in this load segment, the time series decomposition algorithm is used to obtain the trend term of the physiological index data. Through time series decomposition, the curve can be decomposed into a trend term, a seasonal term, and a residual. The trend term reflects the overall change trend of the data. Therefore, the mean value of all trend values in the trend term is negatively correlated to obtain the trend smoothness of the physiological index data. When the overall trend value is smaller, it indicates that the trend is smoother and the severity of physical sign changes is not high.

[0084] Furthermore, the product of the physiological index data and the correlation index of this load segment and the trend smoothness is used as the trend probability index of the physiological index data and this load segment. Through the influence of the correlation index, for physiological index data with a higher degree of correlation, more trend smoothness is considered.

[0085] Finally, the cumulative value of the trend probability indexes of this load segment and all physiological index data is normalized to obtain the physical sign cumulative degree of this load segment. Combining with the trend analysis of all physiological index data, when the overall trend level is smoother, the possibility of fatigue accumulation is greater.

[0086] Finally, combining the frequency cumulative degree, deformation cumulative degree, and physical sign cumulative degree of this load segment, the cumulative intensity of this load segment is obtained. In the embodiment of the present invention, the product of the frequency cumulative degree, deformation cumulative degree, and physical sign cumulative degree of this load segment is normalized to obtain the cumulative intensity of this load segment. When the cumulative intensity is greater, it indicates that the possibility of the load segment being in a fatigue situation is higher.

[0087] S5: Adjust the rehabilitation training of the patient based on the cumulative intensity of the load segment in the deformation segmentation and the number of non-load segments.

[0088] Through the number of non-load segments of the patient in the period after deformation, the training compliance of the current patient is reflected, so as to adjust the movement, and through the cumulative intensity of the load segment, the fatigue and pain load situations are further divided, so as to make different adjustments to the subsequent training plan, making the adjustment suggestions for the rehabilitation training more reliable and accurate.

[0089] In the embodiment of the present invention, the proportion of the number of non-load segments in the deformation segmentation is used as the compliance judgment index. When there are more non-load segments in the deformation, the non-standard situation of the patient needs to be considered to avoid affecting the rehabilitation effect. Therefore, when the compliance judgment index is greater than the preset compliance adjustment threshold, the rehabilitation training of the current patient is recorded as movement adjustment, and the movement standard situation of the patient's training needs to be paid attention to and adjusted in time. In the embodiment of the present invention, the compliance adjustment threshold is set to 0.7, and the specific value can be adjusted by the implementer himself.

[0090] Furthermore, when the cumulative intensity magnitude of all load segments does not exceed a preset fatigue threshold, it indicates that the existing load segments are highly likely to be caused by fatigue, and it is very likely that the load caused by the patient's pain is relatively high. In this case, the current patient's rehabilitation training is recorded as intensity adjustment. At this time, it is necessary to pay attention to whether the patient may have secondary injuries and adjust the training intensity accordingly. Otherwise, the current patient's rehabilitation training is recorded as frequency adjustment, indicating that the patient may already be in a fatigued state, and it is necessary to promptly pay attention to whether to rest to ensure the training quality. In the embodiment of the present invention, the preset fatigue threshold is set to 0.65, and the specific value can be adjusted by the implementer himself.

[0091] In summary, by analyzing the morphological changes of different motion segments divided from the motion performance data, the deformed segments with non-standard movements are determined. For the non-standard deformed segments, the relationship between the changes in the physiological index data of the patient's body during rehabilitation training and the morphological changes of the motion segments is analyzed. Considering the different stress responses caused by fatigue pain load physiologically and the poor compliance when the movement is non-standard, different situations after the movement deformation are quantified through the analysis of the relationship, and the load segments are determined to classify different influencing factors. Furthermore, from the previous deformed motion situation of the load segment and the recovery trend of the physiological index data, the cumulative situation of the patient's body movement intensity is reflected, and fatigue and pain are further classified. Finally, from the non-load situation and the cumulative intensity of the load segment, the rehabilitation training of the patient is adjusted according to different classification results, so as to improve the accuracy of the proposed auxiliary rehabilitation training plan. The present invention combines the analysis of the relationship between the motion deformation situation and the changes in the body's physiological index during the patient's rehabilitation training, classifies and analyzes the patient's motion state, and makes the adjustment of the rehabilitation training more accurate and reliable.

[0092] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. A health information management method for rehabilitation training, characterized in that: The method comprises: Acquire different physiological index data and sports performance data in time series during the patient's rehabilitation training; The patient's motion performance data is segmented to obtain different motion segments; the morphological matching between the data in each motion segment and the standard motion template is analyzed to obtain the morphological fluctuation index of each motion segment; the deformation segment is selected based on the size of the morphological fluctuation index of the motion segment; According to the numerical distribution and change rate of each physiological indicator data at each moment, the fluctuation characteristic index of each physiological indicator data at each moment is obtained; in each deformation segment, the correlation between the morphological fluctuation index and the fluctuation characteristic index of each physiological indicator data is analyzed to obtain the correlation index between each deformation segment and each physiological indicator data; Based on the size of the correlation index between each deformation segment and all physiological index data, the load segment is determined; according to the time difference between the load segment and the previous deformation segment and the number of previous movement segments, as well as the correlation index between the load segment and each physiological index data combined with the time series change trend, the cumulative intensity of the load segment is obtained; Adjust the patient's rehabilitation training based on the cumulative intensity of the loaded segments and the number of unloaded segments in the deformation segmentation; The step of adjusting the patient's rehabilitation training based on the cumulative strength of the load segment and the number of the non-load segments in the deformation segmentation includes: The proportion of the number of non-load segments in the deformation segmentation is used as a compliance judgment index; when the compliance judgment index is greater than the preset compliance adjustment threshold, the current patient rehabilitation training is recorded as action adjustment; When the cumulative intensity of all load segments is not greater than the preset fatigue threshold, the current patient rehabilitation training is recorded as intensity adjustment; otherwise, the current patient rehabilitation training is recorded as frequency adjustment.

2. A health information management method for rehabilitation training according to claim 1, characterized in that: The method for obtaining the morphological fluctuation index includes: For any motion segment, the motion performance data on the motion segment is matched with the standard motion template using the DTW algorithm to obtain several matching pairs; In each matching pair, the data value difference between the mean value of the motion performance data on the motion segment and the mean value of the standard motion template is used as the deviation degree of each matching pair; The sum of the deviations of all matching pairs on the motion segment is normalized to obtain the morphological fluctuation index of the motion segment.

3. A health information management method for rehabilitation training according to claim 1, characterized in that: The method for obtaining the deformation segment comprises: When the normalized value of the morphological fluctuation index is greater than the preset deformation threshold, the corresponding motion segment is used as the deformation segment.

4. A health information management method for rehabilitation training according to claim 1, characterized in that: The method for obtaining the fluctuation characteristic index includes: For any physiological index data, curve fitting is performed on the physiological index data in time series to obtain an index curve; Obtain the slope of the indicator curve at each moment; take the difference in slope between each moment and the previous moment in the time series as the degree of change difference; The product of the value of the physiological indicator data at each moment and the change difference is used as the fluctuation characteristic index of the physiological indicator data at each moment.

5. A health information management method for rehabilitation training according to claim 1, characterized in that: The method for obtaining the associated index includes: For any deformation segment, each physiological index data is used as analysis data in turn; the mean value of the fluctuation characteristic index of the analysis data at all moments on the deformation segment is normalized to obtain the physiological fluctuation degree of the analysis data on the deformation segment; The difference between the physiological fluctuation degree and the morphological fluctuation index of the analysis data on the deformation segment is negatively correlated to obtain the correlation index between the deformation segment and the analysis data.

6. A health information management method for rehabilitation training according to claim 1, characterized in that: The method for determining the load segment includes: For each deformation segment, when there is a correlation index between the deformation segment and the physiological index data that is greater than a preset correlation threshold, the deformation segment is used as a load segment.

7. A health information management method for rehabilitation training according to claim 1, characterized in that: The method for obtaining the cumulative intensity includes: For any load segment, the total number of all exercise segments before the load segment is counted as the cumulative number of times for the load segment; Negative correlation mapping is performed on the time interval between the load segment and the previous load segment to obtain the deformation accumulation degree of the load segment; Combining the correlation index between the load segment and each physiological index data and the change trend of the physiological index data, the cumulative degree of the physical signs of the load segment is obtained; The cumulative intensity of the load segment is obtained by combining the cumulative number of times, cumulative deformation and cumulative signs of the load segment.

8. A health information management method for rehabilitation training according to claim 7, characterized in that: The method for obtaining the physical sign accumulation degree includes: For any physiological index data, after curve fitting is performed on the physiological index data on the load segment, a time series decomposition algorithm is used to obtain the trend item of the physiological index data; the mean of all trend values ​​in the trend item is negatively correlated to obtain the trend flatness of the physiological index data; The product of the correlation index between the physiological index data and the load segment and the trend flatness is used as a possible trend index between the physiological index data and the load segment; The cumulative value of the trend possible index of the load segment and all physiological index data is normalized to obtain the cumulative degree of the physical signs of the load segment.

9. A health information management method for rehabilitation training according to claim 1, characterized in that: The method for acquiring the motion segments includes: dividing the motion performance data in time series into different motion segments by trajectory segmentation.

Images