Intelligent cerebral apoplexy rehabilitation effect management system based on digital platform

Abstract

The invention relates to the technical field of informatization medical care systems, in particular to a cerebral apoplexy rehabilitation effect intelligent management system based on a digital platform, and the system comprises a database, a modeling module, a data interface module, a calculation processing module, a query module, a rehabilitation management module and a data output module. According to the system, a normal distribution fitting rehabilitation evaluation model is constructed, physiological and motion index data of a patient are accessed, rehabilitation index scores and rehabilitation degree scores are calculated, and individualized and quantifiable rehabilitation effect evaluation is achieved. And the rehabilitation management module judges rehabilitation bias through one-way and multi-way mean deviation degrees, outputs overall good, bias rehabilitation and rehabilitation difference results in a classified manner, and supports real-time feedback and rehabilitation intervention suggestions. The method comprises the steps of model construction, data fitting inspection, abnormal data processing and distribution characteristic analysis, adapts to various rehabilitation evaluation indexes, and realizes dynamic tracking and intelligent management of the whole stroke rehabilitation process.

Description

Technical Field

[0001] The present invention relates to the technical field of information-based healthcare systems, and particularly to an intelligent management system for stroke rehabilitation effects based on a digital platform. Background Art

[0002] Stroke is an acute cerebrovascular disease caused by a disruption in the blood supply to the brain, resulting in brain tissue damage, and is characterized by high incidence, high disability rate, and high recurrence rate. Among stroke sequela patients, problems such as movement disorders, language disorders, cognitive disorders, and decline in daily living ability often occur, and there is an urgent need to improve their functional recovery effects and quality of life through scientific and systematic rehabilitation intervention measures.

[0003] Currently, the rehabilitation assessment and management of stroke in clinical practice mainly rely on the subjective judgment of rehabilitation physicians, paper records, and periodic physical examination data. Although various functional assessment scales (such as the Fugl-Meyer scale, Barthel index, modified Rankin score, etc.) have been widely used, their assessment frequency is limited, data records are scattered, and manual statistics are time-consuming, making it difficult to achieve full-cycle and full-dimensional dynamic monitoring and precise analysis of the patient's rehabilitation process. In addition, there is a lack of unified assessment standards and information sharing mechanisms among different rehabilitation institutions, resulting in the problem of "information islands" in rehabilitation data, further affecting the optimization of rehabilitation programs and the quantitative management of treatment effects.

[0004] In recent years, the rapid development of digital medicine and artificial intelligence technology has provided a new solution path for stroke rehabilitation. By means of intelligent algorithms, model-based assessment of the patient's rehabilitation progress is carried out to identify recovery bottlenecks, optimize the intervention rhythm, and improve the allocation efficiency of rehabilitation resources. However, most existing digital platforms still have problems such as low integration, non-intelligent assessment algorithms, and lagging data updates, making it difficult to meet the high requirements for real-time, accuracy, and continuity in the stroke rehabilitation process. Therefore, there is an urgent need to design an intelligent management system for stroke rehabilitation effects based on a digital platform to improve the visualization, intelligence, and personalization levels of the rehabilitation process, so as to provide more scientific, efficient, and refined rehabilitation services for stroke patients. Summary of the Invention

[0005] In view of the above technical problems existing in the prior art, the present invention provides an intelligent management system for stroke rehabilitation effects based on a digital platform, including a database, a modeling module, a data interface module, a calculation and processing module, a query module, a rehabilitation management module, and a data output module; The database is used to store rehabilitation data, health data reference indicators, patient data, parameter distribution functions, and rehabilitation management data; The modeling module is used to establish a rehabilitation assessment model based on physiological indicators or motion indicators that conforms to the normal distribution based on rehabilitation data, and fit the rehabilitation data into a normal distribution function as the rehabilitation distribution function. The data interface module is used to receive data of stroke patients from external devices, and receive rehabilitation index scores and rehabilitation degree scores from the calculation and processing module, and store the stroke patient data, rehabilitation index scores and rehabilitation degree scores in a database. The calculation and processing module is used to calculate rehabilitation index scores and overall rehabilitation degree scores based on patient data and the rehabilitation assessment model. Specifically, it includes: C1. Calculate the rehabilitation index score; according to the rehabilitation assessment model, calculate the standard deviation of the rehabilitation distribution function, obtain the bilateral distribution probability of the patient's physiological indicators or motion indicators, obtain the standard deviation interval corresponding to the contralateral distribution probability for the preset distribution ratio interval, and calculate the rehabilitation index score. C2. Calculate the overall rehabilitation degree score; including: C21. Divide the patients into rehabilitation groups based on the rehabilitation degree score. C22. Conduct K-S test, fitting test and normal distribution test on the rehabilitation degree scores of each rehabilitation group, and substitute them into the normal distribution function to obtain the rehabilitation distribution function. C23. Calculate the overall rehabilitation distribution for all rehabilitation groups. C24. Obtain the unilateral distribution probability of the rehabilitation degree score, and the standard deviation corresponding to the unilateral distribution probability of the preset distribution ratio interval of the overall rehabilitation degree score, and calculate the overall rehabilitation degree score. The query module is used to receive a query instruction from a terminal to retrieve any one or a combination of rehabilitation data, patient data, and the rehabilitation assessment model in the database. The rehabilitation management module is used to evaluate and manage the rehabilitation index scores and overall rehabilitation degree scores; the data output module is used to output the rehabilitation assessment model to an external device, or output the query result to a terminal, and output the management result of the rehabilitation management module.

[0006] Further, the specific steps of the rehabilitation management module include: D1. Calculate the one-way mean deviation; for each patient, calculate the rehabilitation index score according to the physiological indicators or motion indicators, calculate the mean of all rehabilitation index scores, and for each rehabilitation index score, obtain the one-way mean deviation. D2. Calculate the multi-way mean deviation; for each patient, calculate the mean of all rehabilitation index scores to obtain the multi-way mean score, and further obtain the multi-way mean deviation. D3. Determine the bias; based on the one-way mean deviation and the multi-way mean deviation, evaluate the bias of rehabilitation.

[0007] Further, the determination of the bias includes the following three forms: E1. Both the one-way mean deviation and the multi-way mean deviation are non-zero numbers, and the variance of the one-way mean deviation is less than that of the multi-way mean deviation, then it is determined that the overall rehabilitation is good; E2. Both the one-way mean deviation and the multi-way mean deviation are non-zero numbers, and the variance of the one-way mean deviation is greater than that of the multi-way mean deviation, then it is determined that there is a bias in rehabilitation; E3. The multi-way mean deviation is zero and the one-way mean deviation is non-zero, then it is determined that the rehabilitation effect is poor.

[0008] Further, the physiological index or the motion index includes at least one of electroencephalogram (EEG), electromyogram (EMG), heart rate variability (HRV), and walking gait.

[0009] The present invention also provides an intelligent management method for the rehabilitation effect of stroke based on a digital platform, including the following steps: S1. System initialization; construct and initialize an intelligent management system for the rehabilitation effect of stroke based on a digital platform; S2. Access to external data; receive the physiological indexes, motion indexes, and patient basic parameters of stroke patients from external devices through a data interface module; S3. Data storage; store the physiological indexes, motion indexes, and patient basic parameters of stroke patients in a database; S4. Construction of a rehabilitation evaluation model; establish a rehabilitation evaluation model based on historical data; S5. Calculate the physiological parameters and the evaluation scores, feedback the physiological parameters and the evaluation scores to the data interface module, and store them in the database; S6. Rehabilitation effect management; evaluate and manage the rehabilitation effect of patients based on the physiological parameters and the evaluation scores.

[0010] Further, in step S4, the steps for constructing the rehabilitation evaluation model include: S41. Determination of evaluation indexes; determine the physiological indexes or motion indexes used to evaluate the rehabilitation effect as evaluation indexes; S42. Data retrieval; retrieve the relevant data of historical patients in the database, that is, retrieve the data of stroke patients and the rehabilitation index scores in the rehabilitation evaluation model that are the same as the evaluation indexes to form a data set; S43. Data cleaning and data integration; eliminate invalid data in the dataset, reserve positions before adding new data, and enter marker information to facilitate subsequent data integration. The marker information includes the actual data and fitted data of the evaluation indicators, the mean of the actual data and fitted data of the evaluation indicators, scores, and elementary number theory property tests; S44. Data visualization, create a distribution dot plot for the numerical values of patient data; S45. Data fitting and fitting test; perform a normal distribution test on the data. If it conforms to the normal distribution, that is, passes the fitting test, then proceed to step S46; otherwise, perform the following data processing for the data that fails the fitting test; S46. Obtain the normal distribution function as the rehabilitation distribution function and establish a rehabilitation evaluation model.

[0011] Furthermore, the steps for data processing that fails the fitting test include: S451. If the sample size of the sequence is less than the preset rejection value, consider the sequence meaningless and automatically eliminate the sequence; and modify the marker information for data integration; S452. If the sample size of the sequence is greater than the preset rejection value, automatically generate a data warning; S453. Mark the data that cannot be normally fitted, that is, retrieve the data that cannot be normally fitted and enter it into the specified position in the rehabilitation evaluation model, and calculate the mean and score; S454. Feed back the numerical value itself and the mean and score information to the data interface module, prompt relevant data information through the data output module, and wait for external instructions; S455. If an external instruction regarding confirmation of elimination is received from the data interface, automatically eliminate the sequence that cannot be normally fitted and modify the marker information for data integration; S456. If an external instruction regarding confirmation of modification is received from the data interface, wait for the exact modified data on the data interface, write the modified data, and use the modified value as the historical data for this position when performing data integration next time; S457. Only receive the external instruction for this data once and it must include options regarding modification or elimination. If other instructions are received, reject the execution and automatically feedback a prompt indicating an instruction error.

[0012] Furthermore, the specific steps for calculating physiological parameters and evaluation scores in step S5 include: S51. Perform a normal distribution test on the actual numerical values of physiological parameters and motion parameters. If the normal distribution test is passed, then execute step S52; if not passed, then execute step S53; S52. Calculate the mean and score of the above actual numerical values and calculate the variance of the actual numerical values; S53. If the normal distribution value is not passed, a data warning is automatically generated, the above actual value is fed back to the data interface module, and the external device that outputs the value is prompted with the above actual value via the data output module.

[0013] Further, in step S6, the specific steps of rehabilitation effect management include: S61. Calculation of one-way mean deviation; including: S611. For each patient, calculate the rehabilitation index score according to any one of the physiological index and the movement index. S612. Calculate the mean of all rehabilitation index scores, and for each rehabilitation index score, calculate the one-way mean deviation. S62. Calculation of multi-directional mean deviation; for each patient, calculate the mean of all rehabilitation index scores to obtain the multi-directional mean score, and further calculate the multi-directional mean deviation. S63. Judging the bias; based on the one-way mean deviation and the multi-directional mean deviation, evaluate the bias of the rehabilitation.

[0014] Further, the said judging the bias includes the following three forms: S631. Both the one-way mean deviation and the multi-directional mean deviation are non-zero numbers, and the variance of the one-way mean deviation is less than that of the multi-directional mean deviation, then it is determined that the overall rehabilitation is good. S632. Both the one-way mean deviation and the multi-directional mean deviation are non-zero numbers, and the variance of the one-way mean deviation is greater than that of the multi-directional mean deviation, then it is determined that the rehabilitation is biased. S633. The multi-directional mean deviation is zero, and the one-way mean deviation is non-zero, then it is determined that the rehabilitation effect is poor.

[0015] Further, when it is determined that the rehabilitation is biased, the evaluation information is fed back to the terminal device that outputs the value through the data output module. The evaluation information includes the bias ratio of the physiological parameters and the movement parameters in the whole, and a prompt for increasing the training of the weak items is given.

[0016] Beneficial effects The present invention has the following beneficial effects: 1. Realize the structured storage and intelligent modeling of multi-source data: The present invention centrally stores rehabilitation data, patient basic data, health index reference values, parameter distribution functions and management data through the database module, providing complete data support for rehabilitation evaluation modeling and statistical analysis; through the modeling module, a rehabilitation evaluation function based on normal distribution fitting is established, making the rehabilitation index computable and statistically significant, and laying a mathematical foundation for subsequent scoring and risk analysis.

[0017] 2. Improve the scientificity and accuracy of rehabilitation assessment: Through the calculation and processing module, the system introduces statistical tools such as standard deviation probability intervals, bilateral distribution probabilities, and K-S tests to achieve the quantitative output of rehabilitation index scores and overall rehabilitation scores. It can not only reflect the current rehabilitation status but also support the comparison of stage treatment effects, effectively overcoming the deficiencies of traditional methods that rely on subjective scoring and qualitative judgments.

[0018] 3. Support hierarchical and classified rehabilitation intervention management: The rehabilitation management module calculates the one-way mean deviation and multi-way mean deviation to establish a rehabilitation effect bias determination mechanism, distinguishing three types of rehabilitation statuses: "overall good rehabilitation", "biased rehabilitation", and "poor rehabilitation effect", which helps physicians accurately identify the key points of intervention; when it is identified as biased rehabilitation, the system can automatically feedback training suggestions, prompting to strengthen the functional training of weak items and realizing personalized adjustment of the rehabilitation path.

[0019] 4. Enhance the system's data exception handling ability and security: For data that fails to fit the normal distribution, the present invention designs a perfect exception handling process, including operations such as automatic elimination, data warning, and external instruction confirmation, to ensure that the rehabilitation model only adopts data with statistical significance; at the same time, suspicious data is marked and tracked to improve the transparency and controllability of data management and ensure the credibility of the analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic structural diagram of the system of the present invention; Figure 2 is a schematic flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] To deepen the understanding of the present invention, the following will further elaborate on the present invention in combination with embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation to the protection scope of the present invention.

[0022] Embodiment 1 According to Figure 1 shown, this embodiment provides an intelligent management system for stroke rehabilitation effects based on a digital platform, including a database, a modeling module, a data interface module, a calculation and processing module, a query module, a rehabilitation management module, and a data output module; The database is used to store rehabilitation data, health data reference indicators, patient data, parameter distribution functions, and rehabilitation management data; The modeling module is used to establish a rehabilitation assessment model based on physiological indicators or motion indicators that conforms to the normal distribution based on rehabilitation data, and fit the rehabilitation data into a normal distribution function as the rehabilitation distribution function; The data interface module is used to receive stroke patient data from external devices, as well as receive rehabilitation index scores and rehabilitation degree scores from the calculation and processing module, and store the stroke patient data, rehabilitation index scores, and rehabilitation degree scores in a database.

[0023] The calculation and processing module is used to calculate rehabilitation index scores and overall rehabilitation degree scores based on patient data and a rehabilitation evaluation model; The query module is used to receive a query instruction from a terminal to retrieve any one or a combination of multiple items of rehabilitation data, patient data, and the rehabilitation evaluation model in the database; The rehabilitation management module is used to evaluate and manage rehabilitation index scores and overall rehabilitation degree scores; the data output module is used to output the rehabilitation evaluation model to an external device, or output a query result to a terminal, and output the management result of the rehabilitation management module.

[0024] Embodiment 2 This embodiment further limits based on Embodiment 1: In the calculation and processing module, rehabilitation index scores and overall rehabilitation degree scores are calculated based on patient data and a rehabilitation evaluation model; specifically including: C1. Calculate the rehabilitation index score; according to the rehabilitation evaluation model, calculate the standard deviation of the rehabilitation distribution function, obtain the bilateral distribution probability of the patient's physiological index or motor index, obtain the standard deviation interval corresponding to the contralateral distribution probability for a preset distribution ratio interval, and calculate the rehabilitation index score; C2. Calculate the overall rehabilitation degree score; including: C21. Divide patients into rehabilitation groups based on the rehabilitation degree score , where represents the th rehabilitation group, represents the th patient in the th rehabilitation group; The rehabilitation group satisfies:

[0025] where represents the patient's rehabilitation degree score, and respectively represent the minimum and maximum values of the patient's rehabilitation degree; represents the grouping cut-off value; C22. Perform K-S test, goodness-of-fit test, and normal distribution test on the rehabilitation degree scores of each rehabilitation group , and substitute them into the normal distribution function to obtain the rehabilitation distribution function; C23. Calculate the overall rehabilitation distribution for all rehabilitation groups, satisfying:

[0026]

[0027] Among them, represents the overall rehabilitation distribution function; represents the th rehabilitation distribution function of the rehabilitation group; represents the weight number; C24. Obtain the one-sided distribution probability of the rehabilitation degree score and the corresponding standard deviation of the one-sided distribution probability of the preset distribution proportion interval of the overall rehabilitation degree score, and calculate the overall rehabilitation degree score; Example 3 This example further limits based on Example 1: The specific steps of the rehabilitation management module include: D1. Calculation of the one-way mean deviation; For each patient, calculate the rehabilitation index score according to the physiological index or the movement index, average all the rehabilitation index scores, and for each rehabilitation index score, obtain the one-way mean deviation; D2. Calculation of the multi-directional mean deviation; For each patient, average all the rehabilitation index scores to obtain the multi-directional mean score, and further obtain the multi-directional mean deviation; D3. Judge the bias; Based on the one-way mean deviation and the multi-directional mean deviation, evaluate the bias of the rehabilitation.

[0028] In step D1, the one-way mean deviation satisfies:

[0029] Among them, represents the one-way mean deviation for the th rehabilitation index score; represents the th actual value of the rehabilitation index score; represents the th mean value of the rehabilitation index score; The formula after averaging is:

[0030] Among them, represents the total amount of the rehabilitation index score.

[0031] In step D2, the one-way mean deviation satisfies:

[0032] Among them, represents the multi-directional mean deviation; The multi-directional mean score representing the scores of all rehabilitation indicators.

[0033] The judgment bias includes the following three forms: E1. When both the unidirectional mean deviation and the multi-directional mean deviation are non-zero numbers, and the variance of the unidirectional mean deviation is less than the multi-directional mean deviation, it is determined that the overall rehabilitation is good; E2. When both the unidirectional mean deviation and the multi-directional mean deviation are non-zero numbers, and the variance of the unidirectional mean deviation is greater than the multi-directional mean deviation, it is determined that the rehabilitation is biased; E3. When the multi-directional mean deviation is zero and the unidirectional mean deviation is non-zero, it is determined that the rehabilitation effect is poor.

[0034] The physiological index or the motion index includes at least one of electroencephalogram (EEG), electromyogram (EMG), heart rate variability (HRV), and walking gait.

[0035] Example 4 As Figure 2 shown: A smart management method for stroke rehabilitation effect based on a digital platform in this example includes the following steps: S1. System initialization; constructing and initializing a smart management system for stroke rehabilitation effect based on a digital platform; S2. Access of external data; receiving the physiological and motion indexes of stroke patients and the patient's basic parameters from external devices through the data interface module; S3. Data storage; storing the physiological and motion indexes of stroke patients and the patient's basic parameters in the database; S4. Construction of a rehabilitation evaluation model; establishing a rehabilitation evaluation model based on historical data; S5. Calculating physiological parameters and evaluation scores, feeding back the physiological parameters and evaluation scores to the data interface module, and storing them in the database; S6. Rehabilitation effect management; evaluating and managing the rehabilitation effect of patients based on physiological parameters and evaluation scores.

[0036] Example 5 This example further limits based on Example 4: In step S4, the steps of constructing the rehabilitation evaluation model include: S41. Determination of evaluation indicators; determining the physiological or motion indicators used to evaluate the rehabilitation effect as evaluation indicators; S42. Data retrieval; retrieving the relevant data of historical patients in the database, that is, retrieving the stroke patient data and the rehabilitation index scores in the rehabilitation evaluation model that are the same as the evaluation indicators to form a data set; S43. Data cleaning and data integration; eliminate invalid data in the dataset, reserve positions before adding new data, and enter marker information to facilitate subsequent data integration. The marker information includes the actual data and fitted data of the evaluation indicators, the mean of the actual data and fitted data of the evaluation indicators, scores, and basic number theory property tests. S44. Data visualization, create a distribution dot plot for the numerical values of patient data. S45. Data fitting and fitting test; perform a normal distribution test on the data. If it conforms to the normal distribution, that is, passes the fitting test, then proceed to step S46; otherwise, perform the following data processing for the data that fails the fitting test. S46. Obtain the normal distribution function as the rehabilitation distribution function and establish a rehabilitation evaluation model.

[0037] The steps for data processing that fails the fitting test include: S451. If the sample size of the sequence is less than the preset rejection value, the sequence is considered meaningless and is automatically eliminated; and the marker information for data integration is modified. S452. If the sample size of the sequence is greater than the preset rejection value, a data warning is automatically generated. S453. Mark the data that cannot be normally fitted, that is, retrieve the data that cannot be normally fitted and enter it into the specified position in the rehabilitation evaluation model, and calculate the mean and scores. S454. Feed back the numerical value itself, as well as the mean and score information, to the data interface module, prompt relevant data information through the data output module, and wait for external instructions. S455. If an external instruction for confirmation of elimination is received from the data interface, the sequence that cannot be normally fitted is automatically eliminated, and the marker information for data integration is modified. S456. If an external instruction for confirmation of modification is received from the data interface, wait for the exact modified data on the data interface, write the modified data, and use the modified value as the historical data at this position when performing data integration next time. S457. For external instructions regarding this data, only one instruction is received and it must include options for modification or elimination. If other instructions are received, they are rejected and an error prompt instruction is automatically fed back.

[0038] Example 6 This example further limits Example 4: The specific steps for calculating the physiological parameters and evaluation scores in step S5 include: S51. Perform a normal distribution test on the actual values of the physiological parameters and exercise parameters. If the normal distribution test is passed, then proceed to step S52; if not, then proceed to step S53. S52. Calculate the mean and score of the above actual values, and calculate the variance of the actual values; S53. If the normal distribution value is not passed, automatically generate a data warning, feedback the above actual values to the data interface module, and prompt the above actual values to the external device that outputs the values via the data output module.

[0039] Example 7 This example further limits based on Example 4: In step S6, the specific steps of rehabilitation effect management include: S61. Calculation of one-way mean deviation; including: S611. For each patient, calculate the rehabilitation index score according to any one of the physiological index and the movement index; S612. Calculate the mean of all rehabilitation index scores, and calculate the one-way mean deviation for each rehabilitation index score; S62. Calculation of multi-directional mean deviation; for each patient, calculate the mean of all rehabilitation index scores to obtain the multi-directional mean score, and further calculate the multi-directional mean deviation; S63. Judge the bias; based on the one-way mean deviation and the multi-directional mean deviation, evaluate the bias of the rehabilitation.

[0040] In step S61, the one-way mean deviation satisfies:

[0041] Among them, represents the one-way mean deviation for the th rehabilitation index score; represents the actual value of the th rehabilitation index score; represents the mean of the th rehabilitation index score; The formula for calculating the mean after is:

[0042] Among them, represents the total amount of rehabilitation index scores.

[0043] In step S62, the one-way mean deviation satisfies:

[0044] Among them, represents the multi-directional mean deviation; represents the multi-directional mean score of all rehabilitation index scores.

[0045] The judgment bias includes the following three forms: S631. When both the unidirectional mean deviation degree and the multi-directional mean deviation degree are non-zero numbers, and the variance of the unidirectional mean deviation degree is less than that of the multi-directional mean deviation degree, it is determined that the overall rehabilitation is good; S632. When both the unidirectional mean deviation degree and the multi-directional mean deviation degree are non-zero numbers, and the variance of the unidirectional mean deviation degree is greater than that of the multi-directional mean deviation degree, it is determined that the rehabilitation is biased; S633. When the multi-directional mean deviation degree is zero and the unidirectional mean deviation degree is non-zero, it is determined that the rehabilitation effect is poor.

[0046] When it is determined that the rehabilitation is biased, the data output module feeds back the evaluation information to the terminal device that outputs the value. The evaluation information includes the physiological parameters and the bias ratio of the motion parameters in the whole, and gives a prompt on increasing the training of the weak items.

[0047] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent management system for stroke rehabilitation effects based on a digital platform, characterized in that: It includes a database, a modeling module, a data interface module, a calculation processing module, a query module, a rehabilitation management module and a data output module; The database is used to store rehabilitation data, health data reference indicators, patient data, parameter distribution functions and rehabilitation management data; The modeling module is used to establish a rehabilitation assessment model based on physiological indicators or motor indicators and conforming to normal distribution based on rehabilitation data, and fit the rehabilitation data to a normal distribution function as a rehabilitation distribution function; The data interface module is used to receive the stroke patient data from an external device, and receive the rehabilitation index score and the rehabilitation degree score from the calculation processing module, and store the stroke patient data, the rehabilitation index score and the rehabilitation degree score in a database; The calculation processing module is used to calculate the rehabilitation index score and the overall rehabilitation score based on the patient data and the rehabilitation assessment model; The query module is used to receive a query instruction from the terminal to retrieve any one or a combination of multiple items of rehabilitation data, patient data, and rehabilitation assessment models in the database; The rehabilitation management module is used to evaluate and manage the rehabilitation index scores and the overall rehabilitation degree scores; the data output module is used to output the rehabilitation evaluation model to an external device, or output the query results to a terminal, and output the management results of the rehabilitation management module.

2. The intelligent management system for stroke rehabilitation effect based on a digital platform according to claim 1 is characterized in that: In the calculation processing module, the calculation of the rehabilitation index score and the overall rehabilitation score specifically includes: C1. Calculate the rehabilitation index score; according to the rehabilitation assessment model, calculate the standard deviation of the rehabilitation distribution function, obtain the bilateral distribution probability for the patient's physiological index or motor index, obtain the standard deviation interval corresponding to the contralateral distribution probability for the preset distribution ratio interval, and calculate the rehabilitation index score; C2. Calculate the overall recovery score; including: C21. Based on the recovery scores, patients are divided into rehabilitation groups; C22. Perform KS test, fitting test and normal distribution test on the rehabilitation scores of each rehabilitation group, and substitute into the normal distribution function to obtain the rehabilitation distribution function; C23. Calculate the overall recovery distribution for all recovery groups; C24. Calculate the one-sided distribution probability of the rehabilitation score and the corresponding standard deviation of the one-sided distribution probability of the preset distribution ratio interval of the overall rehabilitation score, and calculate the overall rehabilitation score.

3. The intelligent management system for stroke rehabilitation effect based on a digital platform according to claim 2 is characterized in that: In the rehabilitation management module, the specific steps of evaluating and managing the rehabilitation index scores and the overall rehabilitation degree scores include: D1. Calculation of one-way mean deviation: For each patient, the rehabilitation index score is calculated based on the physiological index or the motor index, all rehabilitation index scores are averaged, and for each rehabilitation index score, the one-way mean deviation is calculated; D2. Calculation of multi-directional mean deviation: For each patient, all rehabilitation index scores are averaged to obtain the multi-directional mean score, and the multi-directional mean deviation is further calculated; D3. Determine bias; evaluate the bias of rehabilitation based on unidirectional mean deviation and multidirectional mean deviation.

4. The intelligent management system for stroke rehabilitation effect based on a digital platform according to claim 3 is characterized by: The judgment bias includes the following three forms: E1. If both the unidirectional mean deviation and the multidirectional mean deviation are non-zero, and the variance of the unidirectional mean deviation is less than that of the multidirectional mean deviation, then the overall recovery is judged to be good; E2. If both the unidirectional mean deviation and the multidirectional mean deviation are non-zero, and the variance of the unidirectional mean deviation is greater than that of the multidirectional mean deviation, it is determined to be biased recovery; E3. If the multi-directional mean deviation is zero and the unidirectional mean deviation is non-zero, it is judged that the rehabilitation effect is poor.

5. The intelligent management system for stroke rehabilitation effect based on a digital platform according to claim 1 is characterized in that: The physiological index or motion index includes at least one of electroencephalogram (EEG), electromyogram (EMG), heart rate variability (HRV) and walking gait.

6. A method for intelligent management of stroke rehabilitation effects based on a digital platform, implemented based on the intelligent management system for stroke rehabilitation effects based on a digital platform as claimed in any one of claims 1 to 5, comprising the following steps: S1. System initialization: Build and initialize the intelligent management system for stroke rehabilitation effect based on digital platform; S2. Access to external data: receiving the physiological and motor indicators of stroke patients and basic parameters of patients from external devices through the data interface module; S3, data storage; The physiological indexes and motor indexes of stroke patients, as well as basic parameters of patients are stored in a database; S4. Construction of rehabilitation assessment model: Based on historical data, a rehabilitation assessment model is established; S5, calculating the physiological parameters and the evaluation scores, feeding back the physiological parameters and the evaluation scores to the data interface module, and storing them in the database; S6. Rehabilitation effect management: Evaluate and manage the patient's rehabilitation effect based on physiological parameters and assessment scores.

7. The method for intelligent management of stroke rehabilitation effects based on a digital platform according to claim 6, characterized in that: In step S4, the steps of constructing the rehabilitation assessment model include: S41. Determine the evaluation index: determine the physiological index or exercise index used to evaluate the rehabilitation effect as the evaluation index; S42, data retrieval: retrieve relevant data of historical patients in the database, that is, retrieve the data of stroke patients and the rehabilitation index scores that are the same as the evaluation index in the rehabilitation evaluation model to form a data set; S43, data cleaning and data integration: remove invalid data from the data set, reserve space before adding new data, enter marking information to facilitate subsequent data integration, marking information includes actual data and fitted data of evaluation indicators, mean and score of actual data and fitted data of evaluation indicators, and basic number theory property test; S44, data visualization, make distribution point graph for the patient data values; S45, data fitting and fitting test; perform normal distribution test on the data, if the data conforms to the normal distribution, i.e. passes the fitting test, proceed to step S46, otherwise perform the following data processing that fails the fitting test; S46. Obtain a normal distribution function as a rehabilitation distribution function and establish a rehabilitation assessment model.

8. The method for intelligent management of stroke rehabilitation effects based on a digital platform according to claim 7, characterized in that: The specific steps of calculating the physiological parameters and the evaluation scores in step S5 include: S51, performing a normal distribution test on the actual values ​​of the physiological parameters and the motion parameters, if the normal distribution test is passed, executing step S52; if not, executing step S53; S52, obtaining the mean and score of the actual values, and obtaining the variance of the actual values; S53. If the value does not pass the normal distribution, a data warning is automatically generated, the actual value is fed back to the data interface module, and the actual value is prompted to the external device that outputs the value via the data output module.

9. The method for intelligent management of stroke rehabilitation effects based on a digital platform according to claim 6, characterized in that: In step S6, the specific steps of rehabilitation effect management include: S61. Calculation of one-way mean deviation; including: S611. For each patient, calculate a rehabilitation index score according to any one of the physiological index and the motor index; S612, calculating the average of all rehabilitation index scores, and calculating the one-way mean deviation for each rehabilitation index score; S62, calculation of multi-directional mean deviation: for each patient, all rehabilitation index scores are averaged to obtain a multi-directional mean score, and the multi-directional mean deviation is further calculated; S63. Determine bias; evaluate the bias of rehabilitation based on unidirectional mean deviation and multidirectional mean deviation.

10. The method for intelligent management of stroke rehabilitation effects based on a digital platform according to claim 9, characterized in that: The judgment bias includes the following three forms: S631. If both the unidirectional mean deviation and the multidirectional mean deviation are non-zero, and the variance of the unidirectional mean deviation is smaller than that of the multidirectional mean deviation, it is determined that the overall recovery is good; S632: If both the unidirectional mean deviation and the multidirectional mean deviation are non-zero, and the variance of the unidirectional mean deviation is greater than that of the multidirectional mean deviation, it is determined to be biased recovery; S633. If the multi-directional mean deviation is zero and the unidirectional mean deviation is non-zero, it is determined that the rehabilitation effect is poor.

Images