Accurate rehabilitation effect evaluation and personalized rehabilitation scheme generation method

Abstract

The invention discloses a precise rehabilitation effect evaluation and personalized rehabilitation scheme generation method, which is applied to the technical field of rehabilitation treatment. Comprising the following steps: acquiring historical and real-time rehabilitation evaluation data of a patient in a preset time period; according to the influence of all rehabilitation evaluation indexes in the rehabilitation evaluation data on the health degrees of the patient in different evaluation fields, determining the influence weights of all rehabilitation evaluation indexes in different evaluation fields; evaluating rehabilitation effects of the patient in different historical periods, inputting the rehabilitation effects into the rehabilitation effect prediction model, and outputting a predicted rehabilitation effect; setting a rehabilitation scheme generation rule of the patient; and based on the predicted rehabilitation effect and the real-time rehabilitation effect of the patient, generating a personalized rehabilitation scheme based on a rehabilitation scheme generation rule. According to the method, the future rehabilitation state is predicted by evaluating the historical rehabilitation condition of the patient, so that the rehabilitation scheme is formulated in a targeted manner, and a more accurate and personalized rehabilitation training scheme can be provided.

Description

Technical Field

[0001] The present invention relates to the field of rehabilitation therapy technology, and more specifically to a method for accurately evaluating rehabilitation effects and generating personalized rehabilitation programs. Background Art

[0002] Rehabilitation assessment is an indispensable part of the rehabilitation treatment process. Accurate rehabilitation assessment results can help formulate rehabilitation plans. According to the patient's specific situation and needs, a personalized rehabilitation plan can help patients restore physical function, improve quality of life, and prevent complications and secondary injuries, so as to achieve the best rehabilitation effect. However, most of the existing rehabilitation plans are formulated by doctors based on the patient's current gender, age, and severity of the disease. In the subsequent rehabilitation process, there is a lack of a comprehensive assessment of the patient's rehabilitation status, and the initial rehabilitation plan may not be suitable for subsequent rehabilitation treatment. Therefore, how to provide a method for accurate rehabilitation effect evaluation and personalized rehabilitation plan generation is an urgent problem that technicians in this field need to solve. Summary of the invention

[0003] In view of this, the present invention provides a method for accurately evaluating rehabilitation effects and generating personalized rehabilitation plans, which generates personalized rehabilitation plans based on the accurate evaluation results of rehabilitation effects.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A method for accurately evaluating rehabilitation effects and generating personalized rehabilitation plans comprises the following steps:

[0006] S1, obtain the patient's historical and real-time rehabilitation assessment data within a preset time period;

[0007] S2. Determine the influence weights of all rehabilitation assessment indicators on different assessment areas according to the influence of all rehabilitation assessment indicators in the rehabilitation assessment data on the health of patients in different assessment areas;

[0008] S3. Evaluate the rehabilitation effects of patients in different historical periods based on all rehabilitation evaluation indicators and the corresponding influence weights. The rehabilitation effects are the health scores of patients in different evaluation areas.

[0009] S4, inputting the rehabilitation effects of the patient in different historical periods into the rehabilitation effect prediction model, and outputting the predicted rehabilitation effects of the patient in multiple future periods;

[0010] S5. Setting the rules for generating a rehabilitation plan for the patient;

[0011] S6. Based on the patient's predicted rehabilitation effect and real-time rehabilitation effect, a personalized rehabilitation plan is generated based on rehabilitation plan generation rules.

[0012] Optionally, S2 is specifically:

[0013] S21. Divide the evaluation areas into physiological parameters A1, motor ability A2, psychological state A3 and cognitive ability A4; set the sub-category of the evaluation area to impact degree B i1 , Rehabilitation Difficulty B i2 and health hazards B i3 , where i∈[1,4]; will affect the degree B i1 The subclass is set to C ijm , Rehabilitation Difficulty B i2 The subclass is set to C ijn , Health Hazard B i3 The subclass is set to C ijk , where j∈[1,3], m represents the level of impact, n represents the level of rehabilitation difficulty, and k represents the health hazard;

[0014] S22. Determine the influence coefficient of all subclasses on their parent class and the subclasses of all rehabilitation assessment indicators through expert evaluation, and calculate their influence coefficient F i :

[0015]

[0016] In the formula, f ij represents the impact coefficient of subclass j of evaluation domain i, f ijm represents the impact coefficient of the impact level m of subcategory j in the evaluation field i, f ijn represents the impact coefficient of rehabilitation difficulty level n of subcategory j in assessment domain i, f ijk represents the impact coefficient of health hazard k of subcategory j in assessment domain i;

[0017] S23. Impact coefficient F of different assessment areas based on rehabilitation assessment indicators i Calculate the weight W i .

[0018] Optionally, the weight of the rehabilitation assessment indicator is calculated based on its influence coefficient as follows:

[0019]

[0020] Where W i represents the weight of rehabilitation assessment index for assessment area i, F ix 、F iy 、F iz Both represent the impact coefficient of rehabilitation assessment indicators on assessment field i, and X is the number of rehabilitation assessment indicators.

[0021] Optionally, S3 specifically includes: setting scoring standards for all rehabilitation assessment indicators, determining the scores of the rehabilitation assessment indicators of patients in different historical periods; calculating the health scores of each assessment field in different historical periods according to the influence weights of the rehabilitation assessment indicators on different assessment fields and the rehabilitation indicator score values.

[0022] Optionally, the rehabilitation effect prediction model in S4 is an LSTM network, and the rehabilitation effects in different historical periods are used as input sequences of the LSTM network for prediction.

[0023] Optionally, S5 is specifically:

[0024] S51. Setting a risk assessment rule for the patient, and determining the patient's rehabilitation training risk level based on the patient's medical record data;

[0025] S52, setting a classification rule for patients, and judging the rehabilitation training intensity of the patients based on the predicted rehabilitation effect and the real-time rehabilitation effect of the patients;

[0026] S53. Set the patient's rehabilitation training program generation rules, and determine the training programs in the rehabilitation program based on the predicted rehabilitation effects and real-time rehabilitation effects in different assessment fields.

[0027] Optionally, S6 specifically includes: based on the predicted rehabilitation effects of the patient in multiple future periods, for each period, a rehabilitation plan is generated through rehabilitation plan generation rules based on the predicted rehabilitation effects, and the rehabilitation plans for multiple future periods constitute the patient's personalized rehabilitation plan.

[0028] Optionally, S6 and later versions also include:

[0029] S7. After executing the personalized rehabilitation plan, the predicted rehabilitation effects of the patient in multiple future periods are re-predicted based on the rehabilitation effect prediction model and a personalized rehabilitation plan is generated.

[0030] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation, which has the following beneficial effects: the present invention can comprehensively and accurately evaluate the patient's current rehabilitation status, making it easier for doctors to understand the patient's condition; by evaluating the patient's historical rehabilitation status and predicting the future rehabilitation status, and then formulating a targeted rehabilitation program, it can provide a more accurate and personalized rehabilitation training program to provide a reference for doctors; during the rehabilitation process, the rehabilitation program is adjusted in real time to ensure that the rehabilitation program is in line with the patient's rehabilitation condition. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0032] Figure 1 This is a flow chart of the method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] The embodiment of the present invention discloses a method for accurately evaluating rehabilitation effects and generating personalized rehabilitation plans, such as Figure 1 As shown, the following steps are included:

[0035] S1, obtain the patient's historical and real-time rehabilitation assessment data within a preset time period;

[0036] S2. Determine the influence weights of all rehabilitation assessment indicators on different assessment areas according to the influence of all rehabilitation assessment indicators in the rehabilitation assessment data on the health of patients in different assessment areas;

[0037] S3. Evaluate the rehabilitation effects of patients in different historical periods based on all rehabilitation evaluation indicators and the corresponding influence weights. The rehabilitation effects are the health scores of patients in different evaluation areas.

[0038] S4, inputting the rehabilitation effects of the patient in different historical periods into the rehabilitation effect prediction model, and outputting the predicted rehabilitation effects of the patient in multiple future periods;

[0039] S5. Setting the rules for generating a rehabilitation plan for the patient;

[0040] S6. Based on the patient's predicted rehabilitation effect and real-time rehabilitation effect, a personalized rehabilitation plan is generated based on rehabilitation plan generation rules.

[0041] In an embodiment of the present invention, rehabilitation assessment indicators include: cardiopulmonary function, muscle strength, balance ability, joint range of motion, movement speed, endurance, attention, memory, tactile perception ability, emotional disorder diagnosis, self-care ability, etc.

[0042] Furthermore, S2 is specifically:

[0043] S21. Divide the evaluation areas into physiological parameters A1, motor ability A2, psychological state A3 and cognitive ability A4; set the sub-category of the evaluation area to impact degree B i1 , Rehabilitation Difficulty B i2 and health hazards B i3 , where i∈[1,4]; will affect the degree B i1 The subclass is set to C ijm , Rehabilitation Difficulty B i2 The subclass is set to C ijn , Health Hazard B i3 The subclass is set to C ijk , where j∈[1,3], m represents the level of impact, n represents the level of rehabilitation difficulty, and k represents the health hazard;

[0044] S22. Determine the influence coefficient of all subclasses on their parent class and the subclasses of all rehabilitation assessment indicators through expert evaluation, and calculate their influence coefficient F i :

[0045]

[0046] In the formula, f ij represents the impact coefficient of subclass j of evaluation domain i, f ijm represents the impact coefficient of the impact level m of subcategory j in the evaluation field i, f ijn represents the impact coefficient of rehabilitation difficulty level n of subcategory j in assessment domain i, f ijk represents the impact coefficient of health hazard k of subcategory j in assessment domain i;

[0047] In the embodiment of the present invention, experts classify all rehabilitation assessment indicators and determine the categories to which they belong. For example, muscle strength has a certain impact on different fields, such as physiological parameters A1, motor ability A2, psychological state A3, and cognitive ability A4. i1 , Rehabilitation Difficulty B i2 and health hazards B i3 The specific classification is determined by expert evaluation: Physiological parameter influence degree C 11m , Physiological parameters Rehabilitation difficulty C 12n 、Physiological parameters Health hazards C 13k , the degree of influence on sports ability C 21m , Difficulty of sports ability rehabilitation C 22n , Sports ability health hazards C 23k , the degree of influence on sports ability C 31m , Difficulty of sports rehabilitation C 32n , Sports ability health hazards C33k , the degree of impact on cognitive ability C 41m , Cognitive ability number Rehabilitation difficulty C 42n 、Cognitive health hazards 43k , where C 11m It means that the influence of muscle strength on physiological parameters is m level, C 12n Indicates that the current muscle strength is at level n in terms of the difficulty of recovery of physiological parameters, C 13k It indicates that the health hazard that the current muscle strength can cause to the physiological parameters is k.

[0048] Assign values ​​to the influence coefficients of subclasses on their parent classes in all categories, and the sum of the influence coefficients of each subclass is 1.

[0049] Then, in this embodiment, the influence coefficient of muscle strength on physiological parameters is:

[0050] F1=f 11 (f 11m +f 11n +f 11k )+f 12 (f 12m +f 12n +f 12k )+f 13 (f 13m +f 13n +f 13k )

[0051] By analogy, we can obtain the influence coefficients of muscle strength on athletic ability, psychological state and cognitive ability, F2, F3 and F4 respectively.

[0052] S23. Impact coefficient F of different assessment areas based on rehabilitation assessment indicators i Calculate the weight W i .

[0053] Furthermore, the weights of the rehabilitation evaluation indicators are calculated based on their influence coefficients as follows:

[0054]

[0055] Where W i represents the weight of rehabilitation assessment index for assessment area i, F ix 、F iy 、F iz Both represent the impact coefficient of rehabilitation assessment indicators on assessment field i, and X is the number of rehabilitation assessment indicators.

[0056] In the embodiment of the present invention, the influence coefficients of all rehabilitation assessment indicators on different assessment fields are calculated. For all rehabilitation assessment indicators, the weight of one rehabilitation assessment indicator is calculated by the above formula, where: It represents the sum of the ratios of the influence coefficients of all current rehabilitation assessment indicators and all other rehabilitation assessment indicators on domain i.

[0057] Furthermore, S3 is specifically as follows: setting scoring standards for all rehabilitation assessment indicators, determining the scores of rehabilitation assessment indicators for patients in different historical periods; calculating the health scores of each assessment field in different historical periods according to the influence weights of rehabilitation assessment indicators on different assessment fields and the scores of rehabilitation indicators.

[0058] Furthermore, the rehabilitation effect prediction model in S4 is an LSTM network, and the rehabilitation effects of different historical periods are used as the input sequence of the LSTM network for prediction.

[0059] In an embodiment of the present invention, the LSTM network is a special recurrent neural network architecture used to process and predict important events with very long intervals and delays in time series data.

[0060] LSTM has three layers, namely the forget gate, input gate and output gate. The forget gate uses the Sigmoid activation function to achieve information forgetting and retention:

[0061] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0062] Where W f is the weight matrix of the forget gate, b f is the bias vector of the forget gate, σ(·) represents the Sigmoid activation function; h t-1 is the hidden state at the previous moment, x t For the current input, the forget gate outputs a value between 0 and 1 for each state value to indicate the degree of retention of each state;

[0063] The input gate determines the update value through the Sigmoid activation function and creates a new candidate value vector through the tanh function layer:

[0064] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0065]

[0066] Where W i , W C are the weight matrices of the input gate and candidate values, respectively, and b i , b C are the bias vectors of the input gate and candidate value, respectively, i t is the updated value, is a candidate vector;

[0067] The output gate generates an output value through the Sigmoid function, and multiplies the output value after processing with the tanh activation function to produce the final hidden state output:

[0068] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0069] h t =o t *tanh(C t )

[0070] Where W o is the weight matrix of the output gate, b o is the bias vector of the output gate; o t is the output value, h t is the hidden state output, C t is the cell state.

[0071] Cell State C t By updating the candidate vector get:

[0072]

[0073] Old cell state C t-1 And the forget gate output f t Multiply to determine what to forget, and add the new candidate value Get cell state C t .

[0074] Furthermore, S5 is specifically:

[0075] S51. Setting a risk assessment rule for the patient, and determining the patient's rehabilitation training risk level based on the patient's medical record data;

[0076] S52, setting a classification rule for patients, and judging the rehabilitation training intensity of the patients based on the predicted rehabilitation effect and the real-time rehabilitation effect of the patients;

[0077] S53. Set the patient's rehabilitation training program generation rules, and determine the training programs in the rehabilitation program based on the predicted rehabilitation effects and real-time rehabilitation effects in different assessment fields.

[0078] In the embodiment of the present invention, rule matching is implemented based on the Rete algorithm to generate a rehabilitation plan.

[0079] Furthermore, S6 is specifically as follows: based on the predicted rehabilitation effects of the patient in multiple future periods, for each period, a rehabilitation plan is generated through rehabilitation plan generation rules based on the predicted rehabilitation effects, and the rehabilitation plans for multiple future periods constitute the patient's personalized rehabilitation plan.

[0080] Furthermore, after S6, it also includes:

[0081] S7. After executing the personalized rehabilitation plan, the predicted rehabilitation effects of the patient in multiple future periods are re-predicted based on the rehabilitation effect prediction model and a personalized rehabilitation plan is generated.

[0082] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0083] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for accurately evaluating rehabilitation effects and generating personalized rehabilitation plans, characterized in that: The following steps are involved: S1, obtain the patient's historical and real-time rehabilitation assessment data within a preset time period; S2. Determine the influence weights of all rehabilitation assessment indicators on different assessment areas according to the influence of all rehabilitation assessment indicators in the rehabilitation assessment data on the health of patients in different assessment areas; S3. Evaluate the rehabilitation effects of patients in different historical periods based on all rehabilitation evaluation indicators and the corresponding influence weights. The rehabilitation effects are the health scores of patients in different evaluation areas. S4, inputting the rehabilitation effects of the patient in different historical periods into the rehabilitation effect prediction model, and outputting the predicted rehabilitation effects of the patient in multiple future periods; S5. Setting the rules for generating a rehabilitation plan for the patient; S6. Based on the patient's predicted rehabilitation effect and real-time rehabilitation effect, a personalized rehabilitation plan is generated based on rehabilitation plan generation rules.

2. The method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation according to claim 1, characterized in that: S2 is specifically: S21. Divide the evaluation areas into physiological parameters A1, motor ability A2, psychological state A3 and cognitive ability A4; set the sub-category of the evaluation area to impact degree B i1 , Rehabilitation Difficulty B i2 and health hazards B i3 , where i∈[1,4]; will affect the degree B i1 The subclass is set to C ijm , Rehabilitation Difficulty B i2 The subclass is set to C ijn , Health Hazard B i3 The subclass is set to C ijk , where j∈[1,3], m represents the level of impact, n represents the level of rehabilitation difficulty, and k represents the health hazard; S22. Determine the influence coefficient of all subclasses on their parent class and the subclasses of all rehabilitation assessment indicators through expert evaluation, and calculate their influence coefficient F i : In the formula, f ij represents the impact coefficient of subclass j of evaluation domain i, f ijm represents the impact coefficient of the impact level m of subcategory j in the evaluation field i, f ijn represents the impact coefficient of rehabilitation difficulty level n of subcategory j in assessment domain i, f ijk represents the impact coefficient of health hazard k of subcategory j in assessment domain i; S23. Impact coefficient F of different assessment areas based on rehabilitation assessment indicators i Calculate the weight W i .

3. The method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation according to claim 2, characterized in that: The weights calculated based on the impact coefficient of the rehabilitation assessment index are as follows: Where W i represents the weight of rehabilitation assessment index for assessment area i, F ix 、F iy 、F iz Both represent the impact coefficient of rehabilitation assessment indicators on assessment field i, and X is the number of rehabilitation assessment indicators.

4. The method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation according to claim 1, characterized in that: S3 is specifically as follows: setting scoring standards for all rehabilitation assessment indicators, determining the scores of rehabilitation assessment indicators for patients in different historical periods; calculating the health scores of each assessment field in different historical periods based on the influence weights of rehabilitation assessment indicators on different assessment fields and the scores of rehabilitation indicators.

5. The method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation according to claim 1, characterized in that: The rehabilitation effect prediction model in S4 is an LSTM network, which uses the rehabilitation effects of different historical periods as the input sequence of the LSTM network for prediction.

6. The method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation according to claim 1, characterized in that: S5 is as follows: S51. Setting a risk assessment rule for the patient, and determining the patient's rehabilitation training risk level based on the patient's medical record data; S52, setting a classification rule for patients, and judging the rehabilitation training intensity of the patients based on the predicted rehabilitation effect and the real-time rehabilitation effect of the patients; S53. Set the patient's rehabilitation training program generation rules, and determine the training programs in the rehabilitation program based on the predicted rehabilitation effects and real-time rehabilitation effects in different assessment fields.

7. The method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation according to claim 1, characterized in that: S6 is specifically: based on the predicted rehabilitation effects of the patient in multiple future periods, for each period, a rehabilitation plan is generated through rehabilitation plan generation rules based on the predicted rehabilitation effects, and the rehabilitation plans for multiple future periods constitute the patient's personalized rehabilitation plan.

8. The method for accurate rehabilitation effect evaluation and personalized rehabilitation program generation according to claim 1, characterized in that: S6 and later also include: S7. After executing the personalized rehabilitation plan, the predicted rehabilitation effects of the patient in multiple future periods are re-predicted based on the rehabilitation effect prediction model and a personalized rehabilitation plan is generated.

Images