Postoperative rehabilitation intervention system for neurological disease patient based on evidence of evidence-based medicine

Abstract

The invention relates to the technical field of postoperative rehabilitation, in particular to a neurological disease patient postoperative rehabilitation intervention system based on evidence-based medicine evidence. The method comprises the steps that feature data of a to-be-analyzed neurological disease patient and a reference patient, rehabilitation schemes of the reference patient and a postoperative Barthel index are obtained, and a reference factor of the reference patient corresponding to each rehabilitation scheme in each preoperative state is calculated; obtaining a comprehensive Barthel index of each rehabilitation scheme in each preoperative state by using the reference factor and the postoperative Barthel index, determining a correction factor in each preoperative state according to the comprehensive Barthel index, and correcting the comprehensive Barthel index to obtain a target Barthel index; and determining a reference rehabilitation scheme by combining the similar condition of the feature data of the to-be-analyzed neurological disease patient and the feature data of different reference patients and the target Barthel index. The postoperative rehabilitation intervention effect of the neurological disease patient can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of postoperative rehabilitation, and particularly to a postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medical evidence. Background Art

[0002] As a kind of neurological disease, brain tumors usually compress the surrounding brain tissues, blood vessels and nerves, affecting the normal functions of the nervous system. Generally, surgical treatment is required to relieve the compression of the brain tumor on the surrounding brain tissues and help the patient recover various functions.

[0003] The existing KNN algorithm realizes the postoperative rehabilitation intervention of the motor function of the brain tumor patient to be analyzed by analyzing the rehabilitation programs of a large number of historical brain tumor patients. Specifically, a large number of historical brain tumor patients are collected. First, the motor function recovery situation under each preoperative condition and rehabilitation program is determined. Then, a KNN model is established through physiological state indicators. Then, the data of the brain tumor patient to be analyzed is substituted into the model to determine the postoperative rehabilitation intervention method of the current brain tumor patient. When determining the motor function recovery situation under each preoperative condition and rehabilitation program, because there are many historical brain tumor patients selected, there may be multiple historical brain tumor patients under one preoperative condition and rehabilitation program. Generally, the average value of the Barthel index of these patients is used to evaluate the motor function rehabilitation situation under this preoperative condition and rehabilitation program. However, some patients may have a large deviation between the postoperative Barthel index and the normal index due to some special conditions such as other diseases, which cannot reflect some general situations, thus affecting the postoperative rehabilitation intervention effect of the brain tumor patient to be analyzed. Summary of the Invention

[0004] In order to solve the problem of poor intervention effect in the postoperative rehabilitation of the brain tumor patient to be analyzed by the existing method, the purpose of the present invention is to provide a postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medical evidence. The specific technical solutions adopted are as follows: The present invention provides a postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medical evidence, including a memory and a processor. The processor executes the computer program stored in the memory to implement the following steps: Obtain the characteristic data of different dimensions of the neurological disease patient to be analyzed and several reference patients, as well as the rehabilitation program and postoperative Barthel index of each reference patient. The reference patients are neurological disease patients; Based on the discrete distribution characteristics of the postoperative Barthel index of all reference patients, obtain the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state; use the proportion of the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state and the postoperative Barthel index to obtain the comprehensive Barthel index of each rehabilitation plan under each preoperative state, where the preoperative state is determined according to the characteristic data; Determine the correction factor for each preoperative state according to the comprehensive Barthel index; use the correction factor to correct the comprehensive Barthel index to obtain the target Barthel index of each rehabilitation plan under each preoperative state; Combine the similarity between the characteristic data of the patient with the neurological disease to be analyzed and that of different reference patients, the proportion of the number of reference patients corresponding to different rehabilitation plans, and the target Barthel index to determine the reference rehabilitation plan for the patient with the neurological disease to be analyzed.

[0005] Preferably, the obtaining of the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state based on the discrete distribution characteristics of the postoperative Barthel index of all reference patients includes: For the state to be evaluated: For all reference patients in the state to be evaluated and who adopt the candidate plan postoperatively, draw a two-dimensional rectangular coordinate system according to the size of the postoperative Barthel index to obtain the data points corresponding to each reference patient in the state to be evaluated, where both the abscissa and ordinate of the two-dimensional rectangular coordinate system are the postoperative Barthel index; Obtain the local outlier factor of each data point in the two-dimensional rectangular coordinate system; Respectively take the difference between the constant 1 and the normalized value of the local outlier factor of the data point corresponding to each reference patient in the state to be evaluated as the reference factor of each reference patient corresponding to the candidate plan in the state to be evaluated; The state to be evaluated is any preoperative state, and the candidate plan is any one rehabilitation plan.

[0006] Preferably, the obtaining of the comprehensive Barthel index of each rehabilitation plan under each preoperative state by using the proportion of the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state and the postoperative Barthel index includes: Calculate the first sum value of the reference factors of all reference patients corresponding to the candidate plan in the state to be evaluated; Calculate the first ratio between the reference factor of each reference patient corresponding to the candidate plan in the state to be evaluated and the first sum value; According to the first ratio and the postoperative Barthel index of each reference patient corresponding to the candidate solution in the state to be evaluated, the comprehensive Barthel index of the candidate solution in the state to be evaluated is obtained, and both the first ratio and the postoperative Barthel index are positively correlated with the comprehensive Barthel index.

[0007] Preferably, obtaining the comprehensive Barthel index of the candidate solution in the state to be evaluated according to the first ratio and the postoperative Barthel index of each reference patient corresponding to the candidate solution in the state to be evaluated includes: Taking the product of the first ratio and the postoperative Barthel index of the corresponding reference patient as the first eigenvalue of the corresponding reference patient; Determining the sum of the first eigenvalues of all reference patients corresponding to the candidate solution in the state to be evaluated as the comprehensive Barthel index of the candidate solution in the state to be evaluated.

[0008] Preferably, determining the correction factor for each preoperative state according to the comprehensive Barthel index includes: Denoting the maximum value of the comprehensive Barthel indices of all rehabilitation solutions in all preoperative states as the first maximum value; Denoting the maximum value of the comprehensive Barthel indices of all rehabilitation solutions in the state to be evaluated as the second maximum value; Determining the ratio between the first maximum value and the second maximum value as the correction factor in the state to be evaluated.

[0009] Preferably, correcting the comprehensive Barthel index by using the correction factor to obtain the target Barthel index of each rehabilitation solution in each preoperative state includes: Taking the product of the correction factor in the state to be evaluated and the comprehensive Barthel index of the candidate solution in the state to be evaluated as the target Barthel index of the candidate solution in the state to be evaluated.

[0010] Preferably, determining the reference rehabilitation solution for the neurological disease patient to be analyzed by combining the similarity between the characteristic data of the neurological disease patient to be analyzed and that of different reference patients, the proportion of the number of reference patients corresponding to different rehabilitation solutions, and the target Barthel index includes: Performing dimensionality reduction processing on the feature matrix by using the PCA algorithm to obtain the data after dimensionality reduction; the feature matrix is composed of the characteristic data of all reference patients; Constructing a KNN model based on the data after dimensionality reduction, and screening a preset first number of reference preoperative states based on the KNN model; All rehabilitation programs referring to the preoperative state are denoted as the first reference programs; according to the target Barthel index of each first reference program in all preoperative states and the number of reference patients corresponding to each first reference program, the recommendation index of each first reference program is obtained. Based on the recommendation indices of all the first reference programs, the reference rehabilitation programs for the patients with neurological diseases to be analyzed are screened.

[0011] Preferably, the step of obtaining the recommendation index of each first reference program according to the target Barthel index of each first reference program in all preoperative states and the number of reference patients corresponding to each first reference program includes: For any one of the first reference programs: The ratio between the number of reference patients corresponding to this first reference program in each preoperative state and the total number of reference patients corresponding to this first reference program in all preoperative states is taken as the proportion of the number of patients of this first reference program in each preoperative state; the product of the proportion of the number of patients and the target Barthel index of this first reference program in each preoperative state is denoted as the second eigenvalue of this first reference program in each preoperative state. According to the overall distribution of the second eigenvalues of this first reference program in all preoperative states, the recommendation index of this first reference program is obtained.

[0012] Preferably, the step of screening the reference rehabilitation programs for the patients with neurological diseases to be analyzed based on the recommendation indices of all the first reference programs includes: sorting all the first reference programs in descending order of the recommendation index to obtain a rehabilitation program sequence, and determining the first reference programs of the first preset second number in the rehabilitation program sequence as the reference rehabilitation programs for the patients with neurological diseases to be analyzed, where the preset second number is an integer greater than or equal to 1.

[0013] Preferably, the step of screening the first preset number of reference preoperative states based on the KNN model includes: Multiplying the vector composed of the characteristic data of the patient with neurological diseases to be analyzed by the characteristic matrix to obtain the position of the patient with neurological diseases to be analyzed in the KNN model; where the characteristic matrix is composed of the eigenvectors corresponding to the three principal components with the largest eigenvalues in the PCA dimensionality reduction process. Obtaining the first preset number of preoperative states most similar to the patient with neurological diseases to be analyzed in the KNN model, denoted as the reference preoperative states.

[0014] The present invention has at least the following beneficial effects: First, based on the discrete distribution characteristics of the postoperative Barthel index of all reference patients, the referenceability of the reference patients corresponding to each rehabilitation plan under each preoperative state was evaluated to obtain reference factors. Then, by integrating the proportion of the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state and the postoperative Barthel index, the comprehensive Barthel index of each rehabilitation plan under each preoperative state was obtained. The comprehensive Barthel index can better represent the real postoperative motor function rehabilitation of patients. The correction factor of each preoperative state was determined based on the comprehensive Barthel index to correct the comprehensive Barthel index and obtain the target Barthel index. Finally, by combining the similarity between the characteristic data of the neurological disease patients to be analyzed and different reference patients, the proportion of the number of reference patients corresponding to different rehabilitation plans, and the target Barthel index, the reference rehabilitation plan for the neurological disease patients to be analyzed was screened for doctors' reference, which can improve the postoperative rehabilitation intervention effect of the neurological disease patients to be analyzed and the recovery speed of various functions of the patients after surgery. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] 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 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.

[0016] Figure 1 It is a flowchart of the method executed by a postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medicine evidence provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following will be described in detail with reference to the drawings and preferred embodiments. A postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medicine evidence proposed by the present invention is described as follows.

[0018] 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.

[0019] The following will specifically describe the specific solution of a postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medicine evidence provided by the present invention with reference to the drawings.

[0020] Embodiment of a postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medicine evidence: The specific scenario targeted by this embodiment is as follows: In order to reduce the postoperative rehabilitation duration of patients with neurological diseases to be analyzed and enable the rapid recovery of various functions of postoperative patients to the normal state, rehabilitation intervention treatment is often carried out on postoperative patients. During the process of postoperative rehabilitation intervention for patients with neurological diseases to be analyzed, this embodiment combines the rehabilitation plans of patients after surgery, the indicators of patients in different dimensions, and the postoperative recovery situation within a historical time period, and screens the rehabilitation plans of patients similar to the patients to be analyzed, and provides them to doctors for reference.

[0021] This embodiment proposes a postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medical evidence. This system is to achieve the steps as Figure 1 shown below. The specific steps are as follows: Step S1, obtain the feature data of patients with neurological diseases to be analyzed and several reference patients in different dimensions, as well as the rehabilitation plan and postoperative Barthel index of each reference patient. The reference patients are patients with neurological diseases.

[0022] First, collect the feature data of multiple reference patients in multiple dimensions from the database, and collect the feature data of patients with neurological diseases to be analyzed. In this embodiment, the feature data collected for each patient includes age, height, weight, number of disease types, tumor grade, and operation duration. That is, the feature data of each patient in multiple dimensions is collected. In specific applications, the implementer can select the types of feature data according to the specific situation. Then, obtain the rehabilitation plan of each reference patient. The rehabilitation plan includes drug group 1, drug group 2, drug group 3, etc., as well as physical therapy method 1, drug group 2, drug group 3, etc. It should be noted that: The reference patients are all patients who have undergone brain tumor surgery and postoperative rehabilitation, and the patients with neurological diseases to be analyzed are patients who have undergone brain tumor surgery and need postoperative rehabilitation. In this embodiment, the number of reference patients is 200. In specific applications, the implementer can set it according to the specific situation. One month after each reference patient's surgery, the doctor evaluates the motor function of each reference patient one month after surgery based on multiple items such as eating, dressing, bathing, toileting, gait, and activity transfer, and obtains the postoperative Barthel index of each reference patient. The process of obtaining the Barthel index is the prior art in Xi'an and will not be elaborated here.

[0023] So far, this embodiment has collected the feature data of patients with neurological diseases to be analyzed, the feature data of each reference patient, the rehabilitation plan of each reference patient, and the postoperative Barthel index of each reference patient.

[0024] Step S2: Based on the discrete distribution characteristics of the postoperative Barthel indexes of all reference patients, obtain the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state; use the proportion of the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state and the postoperative Barthel index to obtain the comprehensive Barthel index of each rehabilitation plan under each preoperative state, where the preoperative state is determined according to the characteristic data.

[0025] In this embodiment, data of multiple reference patients under different preoperative states and rehabilitation plans are collected in step S1, and each patient has a Barthel index one month after the operation. Since the rehabilitation conditions of some extreme patients may be relatively complex, such as patients with multiple complications, older patients or weak patients, it may cause large deviations in the data, and their postoperative Barthel indexes may deviate greatly from the normal values and cannot reflect the general situation. Given that the intervention plan usually needs to rely on results with high universality, therefore, before that, it is necessary to determine the reference factors of the reference patients corresponding to each rehabilitation plan under each preoperative state to reflect the Barthel availability of each patient under the preoperative state and the rehabilitation plan.

[0026] Since the number of patient examples selected for each preoperative state and rehabilitation plan in this embodiment is sufficient, and a large number of samples can improve the statistical significance of the results, most of these patients are normal patients. For relatively normal patients, the Barthel indexes show an aggregation trend compared with abnormal Barthel indexes. Therefore, the higher the aggregation of the Barthel index, the higher the availability. So next, first calculate the reference factor of each reference patient through aggregation, and then weight the Barthel index of each patient under the preoperative state and the rehabilitation plan by the reference factor to obtain the comprehensive Barthel index of each preoperative state and each rehabilitation plan in the example.

[0027] First, determine the preoperative state of each reference patient according to all the characteristic data of each reference patient collected. When all the characteristic data of two reference patients are the same, these two reference patients belong to the same preoperative state; as long as one characteristic data of two reference patients is different, these two patients do not belong to the same preoperative state. That is, the preoperative states of all reference patients are divided into multiple types according to the characteristic data of all reference patients.

[0028] Next, this embodiment takes one rehabilitation plan and one preoperative state as an example for analysis. The method provided in this embodiment can be used to process other rehabilitation plans and other preoperative states.

[0029] Specifically, any rehabilitation plan is denoted as a candidate plan, and any preoperative state is denoted as a state to be evaluated. For all reference patients in the state to be evaluated who adopt the candidate plan after surgery, a two-dimensional rectangular coordinate system is drawn according to the size of the postoperative Barthel index, and the data points corresponding to each reference patient in the state to be evaluated are obtained. Both the abscissa and ordinate of the two-dimensional rectangular coordinate system are the postoperative Barthel index; the local outlier factor of each data point in the two-dimensional rectangular coordinate system is obtained. The method for obtaining the local outlier factor of a data point is a prior art and will not be elaborated here. The difference between the constant 1 and the normalized value of the local outlier factor of the data point corresponding to each reference patient in the state to be evaluated is respectively used as the reference factor of each reference patient corresponding to the candidate plan in the state to be evaluated.

[0030] In this embodiment, a specific calculation formula for the reference factor is given. The reference factor of the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state can be expressed as: where, represents the reference factor of the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state, represents the local outlier factor of the data point corresponding to the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state, and norm( ) represents the normalization function.

[0031] The local outlier factor is used to reflect the isolation of the data point. The more discrete the distribution of the data points around the data point corresponding to the i-th reference patient, the larger its corresponding local outlier factor; is used to characterize the relative isolation of the data point corresponding to the i-th reference patient. When the local outlier factor of the data point corresponding to the i-th reference patient is larger, it indicates that the data point corresponding to the i-th reference patient is more isolated and has less aggregation, that is, the reference factor is smaller.

[0032] The higher the reference factor of the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state, the higher the contribution to the postoperative Barthel index of the patients under the b-th rehabilitation plan in the a-th preoperative state. Next, the postoperative Barthel index is weighted using the proportion of the reference factors of the reference patients corresponding to each rehabilitation plan in each preoperative state to obtain the comprehensive Barthel index of each rehabilitation plan in each preoperative state.

[0033] Specifically, calculate the first sum value of the reference factors of all reference patients corresponding to the candidate solution in the state to be evaluated; calculate the first ratio between the reference factor of each reference patient corresponding to the candidate solution in the state to be evaluated and the first sum value; take the product of the first ratio and the postoperative Barthel index of the corresponding reference patient as the first eigenvalue after the operation of the corresponding reference patient; determine the sum of the first eigenvalues after the operation of all reference patients corresponding to the candidate solution in the state to be evaluated as the comprehensive Barthel index of the candidate solution in the state to be evaluated.

[0034] In this embodiment, a specific calculation formula for the comprehensive Barthel index is given. The comprehensive Barthel index of the b-th rehabilitation plan in the a-th preoperative state can be expressed as: Among them, represents the comprehensive Barthel index of the b-th rehabilitation plan in the a-th preoperative state, represents the number of reference patients corresponding to the b-th rehabilitation plan in the a-th preoperative state, represents the reference factor of the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state, represents the postoperative Barthel index of the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state.

[0035] represents the first sum value, represents the first ratio corresponding to the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state, represents the first eigenvalue of the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state, which is used to reflect the contribution degree of the postoperative Barthel index of the i-th reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state. When the first eigenvalue of the reference patient corresponding to the b-th rehabilitation plan in the a-th preoperative state is larger and the postoperative Barthel index is also larger, the comprehensive Barthel index of the b-th rehabilitation plan in the a-th preoperative state is larger.

[0036] By using the above method, the comprehensive Barthel index of each rehabilitation plan in each preoperative state can be obtained.

[0037] Step S3, determine the correction factor for each preoperative state according to the comprehensive Barthel index; use the correction factor to correct the comprehensive Barthel index to obtain the target Barthel index of each rehabilitation plan in each preoperative state.

[0038] The rehabilitation process is related not only to the rehabilitation plan, but also to various preoperative conditions such as the patient's basic health status and age. For example, for young patients, even with a simple rehabilitation plan, due to their better physical condition, they may recover very quickly, and the improvement of the Barthel index is relatively significant; while for older patients, although they may receive the same rehabilitation measures, due to physiological limitations, their recovery speed is slower, and the increase in the Barthel index is also smaller. However, this does not mean that the recovery of the rehabilitation plan under this preoperative condition is worse. It is also necessary to correct the comprehensive Barthel index of each rehabilitation plan under each preoperative state, so that patients in different preoperative states can be at the same starting point when evaluating the rehabilitation effect, and can reflect the true situation of the patient's recovery, rather than just the deviation caused by the differences in preoperative conditions.

[0039] For each preoperative state, the maximum Barthel index obtained under all rehabilitation plans represents the maximum recovery ability under this preoperative state. To enable patients in different preoperative states to be at the same starting point when evaluating the rehabilitation effect, the maximum recovery ability can be used as a correction factor. If a patient has a relatively weak maximum recovery ability, then even if his / her Barthel index has increased to a certain extent, he / she still cannot be directly compared with patients with better preoperative health conditions. Therefore, a relatively increased factor needs to be given to reflect his / her maximum effort and recovery potential in the current situation. For patients with a relatively strong maximum recovery ability, since their maximum recovery ability is strong and their recovery potential is relatively large, the Barthel index of such patients should be given a relatively decreased factor because they are already at a relatively high recovery starting point and cannot achieve the same effect through the same improvement amplitude.

[0040] Based on the above characteristics, the maximum value of the comprehensive Barthel index of all rehabilitation plans under all preoperative states is denoted as the first maximum value; the maximum value of the comprehensive Barthel index of all rehabilitation plans under the state to be evaluated is denoted as the second maximum value; the ratio between the first maximum value and the second maximum value is determined as the correction factor under the state to be evaluated. The product of the correction factor under the state to be evaluated and the comprehensive Barthel index of the candidate plan under the state to be evaluated is determined as the target Barthel index of the candidate plan under the state to be evaluated.

[0041] So far, by using the above method, the target Barthel index of each rehabilitation plan under each preoperative state can be obtained.

[0042] Step S4: Determine the reference rehabilitation plan for the patient with neurological disease to be analyzed by combining the similarity between the characteristic data of the patient with neurological disease to be analyzed and that of different reference patients, the proportion of the number of reference patients corresponding to different rehabilitation plans, and the target Barthel index.

[0043] Since there are many factors affecting the recovery of the patient's motor function, and the scales of these factors are difficult to calculate and unify, and KNN performs poorly in high-dimensional spaces, in this step, the method of Principal Component Analysis (PCA) is used to unify the characteristic scales of the preoperative states, reduce the dimension of the data, and alleviate the problem of the curse of dimensionality, making the established model more accurate.

[0044] Specifically, first, construct a feature matrix using the characteristic data of all reference patients. Each row element in the feature matrix is composed of the characteristic data of all dimensions of a reference patient, and each column element in the feature matrix is composed of the characteristic data of the same dimension of all reference patients. That is, one row represents a reference patient, and one column represents a dimension.

[0045] Use the PCA algorithm to perform dimensionality reduction on the feature matrix, obtain the eigenvectors corresponding to the three principal components with the largest eigenvalues, and obtain the data after dimensionality reduction; the PCA algorithm is a prior art and will not be elaborated here.

[0046] Plot the data after dimensionality reduction in a three-dimensional image, that is, construct a KNN model. Take the three obtained eigenvectors as the three column elements in the matrix respectively. That is, the elements in an eigenvector form a column in the matrix, and denote the matrix obtained at this time as the feature matrix. Then, multiply the vector composed of the characteristic data of the patient with neurological disease to be analyzed by the feature matrix to obtain a three-dimensional vector, which is a point in the three-dimensional vector KNN model. Find the point corresponding to this three-dimensional vector in the KNN model, and this point is the position of the patient with neurological disease to be analyzed in the KNN model.

[0047] Obtain the preoperative states corresponding to the preset first number of data points that are most similar to the patient with neurological disease to be analyzed in the KNN model, denoted as the reference preoperative states. That is, obtain the preset first number of nearest neighbor nodes corresponding to the patient with neurological disease to be analyzed in the KNN model. The method for obtaining the nearest neighbors is a prior art and will not be elaborated here. Denote all the rehabilitation plans in the reference preoperative states as the first reference plans, that is, obtain multiple first reference plans. In this embodiment, the preset first number is 10. In specific applications, the implementer can set it according to specific circumstances.

[0048] For any rehabilitation plan, patients with multiple different preoperative states may have adopted this rehabilitation plan. For each preoperative state, there is a corresponding comprehensive Barthel index for this rehabilitation plan. Generally, the recommendability of a certain rehabilitation plan is selected by the Barthel average value of this rehabilitation plan under all reference preoperative states. However, in order to select a more universal result, in this embodiment, the recommendability of each first reference plan will be evaluated according to the target Barthel index of each first reference plan and the number of reference patients corresponding to each first reference plan under all reference preoperative states, and a recommendation index for each first reference plan will be obtained.

[0049] For any first reference plan: The ratio between the number of reference patients corresponding to this first reference plan in each reference preoperative state and the total number of reference patients corresponding to this first reference plan under all reference preoperative states is used as the proportion of the number of patients of this first reference plan in each reference preoperative state; the product of the proportion of the number of patients and the target Barthel index of this first reference plan in each reference preoperative state is denoted as the second eigenvalue of this first reference plan in each reference preoperative state; according to the overall distribution of the second eigenvalues of this first reference plan under all reference preoperative states, the recommendation index of this first reference plan is obtained.

[0050] In this embodiment, a specific calculation formula for the recommendation index is given. The recommendation index of the u-th first reference plan can be expressed as: Where, represents the recommendation index of the u-th first reference plan, K represents the number of reference preoperative states, represents the number of reference patients corresponding to the u-th first reference plan in the k-th reference preoperative state, represents the target Barthel index of the u-th first reference plan in the k-th reference preoperative state.

[0051] represents the proportion of the number of patients of the u-th first reference plan in the k-th reference preoperative state, which is used to reflect the contribution rate of the u-th first reference plan to the Barthel index of the k-th reference preoperative state. The larger this proportion, the greater the contribution rate of the u-th first reference plan to the Barthel index of the k-th reference preoperative state. is the second eigenvalue. When the proportion of the number of patients of the first reference plan in each reference preoperative state is larger and the target Barthel index is also larger, the recommendation index of the u-th first reference plan is larger.

[0052] By using the above method, the recommendation index of each first reference solution can be obtained. The rehabilitation solution with a larger recommendation index is more suitable for recommendation to the patient with the neurological disease to be analyzed. Therefore, all the first reference solutions are sorted in descending order of the recommendation index to obtain a rehabilitation solution sequence, and the first preset second number of first reference solutions in the rehabilitation solution sequence are determined as the reference rehabilitation solutions for the patient with the neurological disease to be analyzed, where the preset second number is an integer greater than or equal to 1. In this embodiment, the preset second number is 3, that is, the three first reference solutions with the largest recommendation indexes are used as the reference rehabilitation solutions for the patient with the neurological disease to be analyzed. In specific applications, the implementer can set it according to the specific situation.

[0053] Subsequently, when performing rehabilitation intervention on the patient with the neurological disease to be analyzed, the doctor can refer to the reference rehabilitation solution of the patient with the neurological disease to be analyzed to perform corresponding rehabilitation treatment on the patient with the neurological disease to be analyzed.

[0054] In this embodiment, first, based on the discrete distribution characteristics of the postoperative Barthel index of all reference patients, the referenceability of the reference patients corresponding to each rehabilitation solution in each preoperative state is evaluated to obtain reference factors. Then, by combining the proportion of the reference factors of the reference patients corresponding to each rehabilitation solution in each preoperative state and the postoperative Barthel index, the comprehensive Barthel index of each rehabilitation solution in each preoperative state is obtained. The comprehensive Barthel index can better represent the real postoperative motor function rehabilitation situation of the patients, and the correction factor in each preoperative state is determined according to the comprehensive Barthel index to correct the comprehensive Barthel index to obtain the target Barthel index. Finally, by combining the similarity between the characteristics data of the patient with the neurological disease to be analyzed and different reference patients, the proportion of the number of reference patients corresponding to different rehabilitation solutions, and the target Barthel index, the reference rehabilitation solutions for the patient with the neurological disease to be analyzed are screened for the doctor's reference, which can improve the postoperative rehabilitation intervention effect of the patient with the neurological disease to be analyzed and improve the recovery speed of various functions of the patient after surgery.

[0055] In other embodiments, a device is further provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the method performed by the above-mentioned postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medicine evidence. The device may specifically be a chip, a component or a module. The chip may include a processor and a memory connected to each other. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the method performed by the above-mentioned postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medicine evidence provided in the above embodiment.

[0056] In other embodiments, a computer program product is further provided. When the computer program product runs on a computer, it causes the computer to execute the above-related steps to implement the method performed by the postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medical evidence provided in the above embodiments.

[0057] In other embodiments, a computer-readable storage medium is further provided. The computer-readable storage medium stores computer program code. When the computer program code runs on a computer, it causes the computer to execute the above-related method steps to implement the method performed by the postoperative rehabilitation intervention system for neurological disease patients based on evidence-based medical evidence provided in the above embodiments.

[0058] Among them, the provided system, device, computer program product, and computer-readable storage medium are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be elaborated here.

[0059] It should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine, comprising a memory and a processor, characterized in that: The processor executes the computer program stored in the memory to implement the following steps: Obtaining characteristic data of different dimensions of the patient with a neurological disease to be analyzed and several reference patients, as well as the rehabilitation program and postoperative Barthel index of each reference patient, where the reference patient is a patient with a neurological disease; Based on the discrete distribution characteristics of the postoperative Barthel index of all reference patients, the reference factors of the reference patients corresponding to each rehabilitation program under each preoperative state are obtained; using the proportion of the reference factors of the reference patients corresponding to each rehabilitation program under each preoperative state and the postoperative Barthel index, the comprehensive Barthel index of each rehabilitation program under each preoperative state is obtained, where the preoperative state is determined based on the characteristic data; Determine a correction factor for each preoperative state according to the comprehensive Barthel index; use the correction factor to correct the comprehensive Barthel index to obtain a target Barthel index for each rehabilitation program under each preoperative state; The reference rehabilitation program for the patient with the neurological disease to be analyzed is determined by combining the similarities between the characteristic data of the patient with the neurological disease to be analyzed and different reference patients, the proportion of the number of reference patients corresponding to different rehabilitation programs and the target Barthel index.

2. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 1 is characterized in that: The discrete distribution characteristics of the Barthel index of all reference patients after surgery are used to obtain the reference factors of the reference patients corresponding to each rehabilitation program under each preoperative state, including: For the pending evaluation status: For all reference patients in the state to be evaluated and who adopted the candidate regimen after surgery, a two-dimensional rectangular coordinate system is drawn according to the size of the postoperative Barthel index to obtain the data points corresponding to each reference patient in the state to be evaluated, where the abscissa and ordinate of the two-dimensional rectangular coordinate system are both the postoperative Barthel index; Obtaining a local anomaly factor for each data point in the two-dimensional rectangular coordinate system; The difference between the constant 1 and the normalized value of the local abnormal factor of the data point corresponding to each reference patient in the state to be evaluated is respectively used as the reference factor of each reference patient corresponding to the candidate scheme in the state to be evaluated; The state to be evaluated is any preoperative state, and the candidate plan is any rehabilitation plan.

3. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 2 is characterized in that: The comprehensive Barthel index of each rehabilitation program under each preoperative state is obtained by using the proportion of the reference factor of the reference patient corresponding to each rehabilitation program under each preoperative state and the postoperative Barthel index, including: Calculate the first sum of the reference factors of all reference patients corresponding to the candidate regimens in the state to be evaluated; Calculate a first ratio between the reference factor of each reference patient corresponding to the candidate regimen in the state to be evaluated and the first sum value; According to the first ratio and the postoperative Barthel index of each reference patient corresponding to the candidate regimen in the state to be evaluated, the comprehensive Barthel index of the candidate regimen in the state to be evaluated is obtained, and the first ratio and the postoperative Barthel index are both positively correlated with the comprehensive Barthel index.

4. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 3 is characterized in that: According to the first ratio and the postoperative Barthel index of each reference patient corresponding to the candidate regimen in the state to be evaluated, a comprehensive Barthel index of the candidate regimen in the state to be evaluated is obtained, including: taking the product of the first ratio and the corresponding reference patient's postoperative Barthel index as the first characteristic value of the corresponding reference patient; The sum of the first eigenvalues ​​of all reference patients corresponding to the candidate regimens under the status to be evaluated is determined as the comprehensive Barthel index of the candidate regimens under the status to be evaluated.

5. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 2, characterized in that: Determining the correction factor for each preoperative state according to the comprehensive Barthel index includes: The maximum value of the combined Barthel index of all rehabilitation programs in all preoperative conditions was recorded as the first maximum value; The maximum value of the combined Barthel index of all rehabilitation programs in the state to be evaluated is recorded as the second maximum value; The ratio between the first maximum value and the second maximum value is determined as a correction factor under the state to be evaluated.

6. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 2, characterized in that: The comprehensive Barthel index is corrected by using the correction factor to obtain the target Barthel index of each rehabilitation program under each preoperative state, including: The product of the correction factor under the state to be evaluated and the comprehensive Barthel index of the candidate solution under the state to be evaluated is determined as the target Barthel index of the candidate solution under the state to be evaluated.

7. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 1, characterized in that: The step of combining the similarity between the characteristic data of the patient with a neurological disease to be analyzed and different reference patients, the proportion of the number of reference patients corresponding to different rehabilitation programs, and the target Barthel index to determine the reference rehabilitation program for the patient with a neurological disease to be analyzed includes: The PCA algorithm is used to perform dimension reduction processing on the feature matrix to obtain dimension-reduced data; the feature matrix is ​​composed of the feature data of all reference patients; Building a KNN model based on the dimensionally reduced data, and screening a preset first number of reference preoperative states based on the KNN model; Record all the rehabilitation plans under the reference preoperative state as the first reference plan; obtain the recommendation index of each first reference plan according to the target Barthel index of each first reference plan under all the reference preoperative states and the number of reference patients corresponding to each first reference plan; Based on the recommendation indexes of all first reference regimens, a reference rehabilitation regimen for the neurological disease patient to be analyzed is screened.

8. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 7, characterized in that: The method of obtaining a recommendation index for each first reference scheme according to the target Barthel index of each first reference scheme under all reference preoperative states and the number of reference patients corresponding to each first reference scheme comprises: For any first reference solution: The ratio between the number of reference patients corresponding to the first reference scheme under each reference preoperative state and the total number of reference patients corresponding to the first reference scheme under all reference preoperative states is taken as the proportion of the number of patients of the first reference scheme under each reference preoperative state; the product of the proportion of the number of patients and the target Barthel index of the first reference scheme under each reference preoperative state is recorded as the second eigenvalue of the first reference scheme under each reference preoperative state; A recommendation index of the first reference scheme is obtained according to the overall distribution of the second characteristic value of the first reference scheme under all reference pre-operative states.

9. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 7, characterized in that: The method of screening reference rehabilitation plans for patients with neurological diseases to be analyzed based on the recommendation index of all first reference plans includes: sorting all first reference plans in descending order of the recommendation index to obtain a rehabilitation plan sequence, and determining the first preset second number of first reference plans in the rehabilitation plan sequence as reference rehabilitation plans for patients with neurological diseases to be analyzed, wherein the preset second number is an integer greater than or equal to 1.

10. The postoperative rehabilitation intervention system for patients with neurological diseases based on evidence-based medicine according to claim 7, characterized in that: The screening of a preset first number of reference preoperative states based on the KNN model includes: Multiplying the vector formed by the characteristic data of the patient with a neurological disease to be analyzed by the characteristic matrix to obtain the position of the patient with a neurological disease to be analyzed in the KNN model; wherein the characteristic matrix is ​​formed by the characteristic vectors corresponding to the three principal components with the largest eigenvalues ​​in the PCA dimensionality reduction process; A preset first number of preoperative states that are most similar to the neurological disease patient to be analyzed in the KNN model are obtained and recorded as reference preoperative states.

Images