Lung function rehabilitation training system and method for pneumology department patient
By using CNN and LSTM algorithms to update the knowledge base and strategy model in real time in the lung function rehabilitation training system of respiratory patients, the problems of low efficiency of traditional methods and lag in strategy are solved, and more efficient and stable lung function rehabilitation training results are achieved.
Patent Information
- Application Number
- CN202411188490.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-27
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional respiratory patients have low efficiency in pulmonary function rehabilitation training methods, and relying on clinical experience, causing the strategy to lag behind the latest research results, affecting stability and predictability.
A system including data acquisition module and rehabilitation analysis module was designed to update the knowledge base and strategy model of rehabilitation training in lung dysfunction through CNN and LSTM algorithms in real time, optimize training strategies, and use databases to collect and analyze theoretical, patient and policy data in real time. Generate real-time updated lung function rehabilitation training strategies.
Real-time optimization of training strategies is achieved, rehabilitation efficiency and effect are improved, the stability and predictability of the strategy are ensured, and the training effect is continuously improved.
Smart Images

Figure CN119092045A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of lung function rehabilitation training, and in particular to a lung function rehabilitation training system and method for respiratory patients. Background Art
[0002] Pulmonary function rehabilitation training is a treatment method that helps respiratory patients improve their lung function and quality of life through a series of exercises and breathing exercises. The purpose of the training is to improve the patient's breathing condition and relieve the symptoms of dyspnea by increasing lung ventilation, improving gas exchange efficiency, and strengthening muscle strength. Pulmonary function rehabilitation training is suitable for various diseases that lead to decreased lung function, including chronic obstructive pulmonary disease (COPD), bronchial asthma, pulmonary fibrosis, etc. For patients with these diseases, pulmonary function rehabilitation training can help them improve their breathing condition and quality of life. Pulmonary function rehabilitation training mainly includes four aspects: aerobic exercise, respiratory muscle training, relaxation training and nutritional guidance. Aerobic exercise such as brisk walking, jogging, swimming, etc. can improve cardiopulmonary function and increase lung ventilation. Respiratory muscle training such as deep breathing and abdominal breathing can enhance the strength of respiratory muscles and improve gas exchange efficiency. Relaxation training such as yoga and meditation can relieve anxiety and tension and reduce the occurrence of dyspnea. Nutritional guidance improves the patient's physical condition and immunity through a reasonable diet structure and nutrient intake.
[0003] At present, traditional respiratory patients' pulmonary function rehabilitation training methods usually take a long time to achieve the expected results, with slow rehabilitation efficiency and long cycles. In the actual training process, pulmonary function rehabilitation training strategies usually rely on medical personnel to formulate them based on clinical experience. Over-reliance on clinical experience can easily cause training strategies to lag behind the latest research results, affecting the stability and predictability of training effects. Summary of the invention
[0004] 1. Technical issues to be resolved
[0005] In view of the shortcomings of the prior art, the present invention provides a lung function rehabilitation training system and method for respiratory patients, which has the advantages of real-time updating of lung dysfunction rehabilitation training data, continuous optimization of strategy models for better rehabilitation training effects, etc., and solves the problems of slow rehabilitation efficiency and long rehabilitation cycle of traditional lung function rehabilitation training methods for respiratory patients, and poor rehabilitation training effects due to reliance on clinical experience.
[0006] (II) Technical solution
[0007] To achieve the above object, the present invention provides the following technical solutions: a pulmonary function rehabilitation training system for respiratory patients, comprising a database for storing pulmonary function rehabilitation training data of respiratory patients, the pulmonary function rehabilitation training system for respiratory patients comprising a data acquisition module and a rehabilitation analysis module;
[0008] The data acquisition module includes a theoretical data unit, a patient data unit and a strategy data unit. The theoretical data unit collects a rehabilitation data set in real time by connecting to a database through a network. The rehabilitation data set includes all theoretical knowledge related to rehabilitation training for pulmonary dysfunction. The theoretical knowledge includes theoretical evaluation data of pulmonary function, theoretical evaluation data of motor ability, theoretical evaluation data of daily living ability and theoretical evaluation data of rehabilitation psychology. The patient data unit collects a patient data set by connecting to a database through a network. The patient data set includes the time point and medical record data of all respiratory patients starting pulmonary function rehabilitation training. The strategy data unit collects a strategy data set in real time by connecting to a database through a network. The strategy data set includes respiratory exercise strategy, nutritional supplement strategy and psychological support strategy. The data acquisition module transmits the rehabilitation data set, the patient data set and the strategy data set to the rehabilitation analysis module through the network.
[0009] The rehabilitation analysis module includes a model training unit, a strategy analysis unit and a function optimization unit. The model training unit is provided with a CNN algorithm model. The model training unit substitutes the rehabilitation data set into the CNN algorithm model to construct a real-time updated pulmonary dysfunction rehabilitation training knowledge base ZSK, and transmits it to the strategy analysis unit through the network. The strategy analysis unit is provided with an LSTM algorithm model. The strategy analysis unit substitutes the strategy data set and the pulmonary dysfunction rehabilitation training knowledge base ZSK into the LSTM algorithm model to construct a pulmonary function rehabilitation training strategy model LSTM + , the pulmonary function rehabilitation training strategy model LSTM + According to the patient data set, a training data set Xlsj is calculated and generated, wherein the training data set Xlsj includes the pulmonary dysfunction status of the respiratory patients at the current time point, the corresponding pulmonary function rehabilitation training strategy and the predicted pulmonary function status after training, and the function optimization unit evaluates the pulmonary function rehabilitation training strategy model LSTM according to the patient data set and the training data set Xlsj. + The immediate reward value r is calculated and the optimization index Yh zs is generated by reward and punishment function. The function optimization unit is provided with an optimization threshold. The function optimization unit compares the optimization index Yh zs with the optimization threshold and then manages the pulmonary function rehabilitation training strategy model LSTM accordingly. + .
[0010] Preferably, the rehabilitation data set expression is {K gn , K yd , K sh , K xl}, K gn represents the theoretical evaluation data of lung function for respiratory patients, K yd represents the theoretical evaluation data of the exercise capacity of respiratory patients, K shrepresents the theoretical evaluation data of the daily living ability of respiratory patients, K xl Represents theoretical evaluation data on the rehabilitation psychology of respiratory patients.
[0011] Preferably, the patient data set is expressed as to The data correspond to the time point and medical record data of each respiratory patient starting pulmonary function rehabilitation training. t represents the time point when the respiratory patient starts pulmonary function rehabilitation training, b represents the medical record data of the respiratory patient, and the subscripts 1 to z represent that there are z respiratory patients participating in pulmonary function rehabilitation training.
[0012] Preferably, the strategy data set expression is {C hx , C yy , C xl}, C hx represents the breathing exercise strategy used for pulmonary function rehabilitation training, C yy Represents the nutritional supplement strategy for pulmonary function rehabilitation training, C xl To represent psychological support strategies used in pulmonary rehabilitation training.
[0013] A lung function rehabilitation training method for respiratory patients, comprising the following steps:
[0014] Step 1: The data acquisition module obtains all theoretical knowledge related to pulmonary dysfunction rehabilitation training, the time point and medical record data of all respiratory patients starting pulmonary function rehabilitation training, as well as respiratory exercise strategies, nutritional supplement strategies and psychological support strategies through network connection database in real time, and classifies them into rehabilitation data set, patient data set and strategy data set;
[0015] Step 2: The rehabilitation analysis module is provided with a CNN algorithm model, and the CNN algorithm model constructs and generates a real-time updated pulmonary dysfunction rehabilitation training knowledge base ZSK based on the rehabilitation data set;
[0016] Step 3: The rehabilitation analysis module is equipped with an LSTM algorithm model. The LSTM algorithm model builds a pulmonary function rehabilitation training strategy model LSTM based on the rehabilitation data set, strategy data set and pulmonary dysfunction rehabilitation training knowledge base ZSK. + ;
[0017] Step 4: Lung Function Rehabilitation Training Strategy Model LSTM + According to the patient data set, the training data set Xlsj is calculated and generated, and the pulmonary dysfunction status of respiratory patients at the current time point, the corresponding pulmonary function rehabilitation training strategy and the predicted pulmonary function status after training are analyzed;
[0018] Step 5: The rehabilitation analysis module evaluates the pulmonary function rehabilitation training strategy model LSTM based on the patient data set and the training data set Xlsj + The immediate reward value r is calculated and the optimization index Yh zs is generated through the reward and punishment function;
[0019] Step 6: The rehabilitation analysis module is set with an optimization threshold, and the optimization index Yh zs is compared with the optimization threshold, corresponding to the management of the pulmonary function rehabilitation training strategy model LSTM + .
[0020] Compared with the prior art, the present invention provides a system and method for pulmonary function rehabilitation training of respiratory patients, which has the following beneficial effects:
[0021] 1. The present invention connects to the database through a data acquisition module to obtain in real time all theoretical knowledge related to pulmonary dysfunction rehabilitation training, the time point and medical record data of all respiratory patients starting pulmonary function rehabilitation training, as well as respiratory exercise strategies, nutritional supplement strategies and psychological support strategies, and classifies them into rehabilitation data sets, patient data sets and strategy data sets. The rehabilitation analysis module is provided with a CNN algorithm model. The CNN algorithm model constructs and generates a real-time updated pulmonary dysfunction rehabilitation training knowledge base ZSK based on the rehabilitation data set, and updates the latest research results and technical standards related to pulmonary dysfunction rehabilitation training in real time.
[0022] 2. The present invention is provided with an LSTM algorithm model through the rehabilitation analysis module. The LSTM algorithm model can better process long time series data. The LSTM algorithm model is constructed based on the rehabilitation data set, the strategy data set and the lung dysfunction rehabilitation training knowledge base ZSK to generate the lung function rehabilitation training strategy model LSTM + , analyze and predict the patient's long-term rehabilitation data, and then formulate a more reasonable rehabilitation training strategy. The rehabilitation analysis module evaluates the pulmonary function rehabilitation training strategy model LSTM based on the patient data set and the training data set Xlsj + The immediate reward value r is calculated and the optimization index Yh zs is generated through the reward and punishment function, realizing the digital evaluation of the pulmonary function rehabilitation training strategy model LSTM + The rehabilitation analysis module is set with an optimization threshold, and the optimization index Yh zs is compared with the optimization threshold. When the optimization index Yh zs is lower than the optimization threshold, it means that the pulmonary function rehabilitation training strategy model LSTM + The effectiveness is low. The function optimization unit generates an optimization signal and transmits it to the strategy analysis unit. The strategy analysis unit reconstructs and generates a new pulmonary function rehabilitation training strategy model LSTM based on the optimization signal. + Until the optimization index Yh zs is included in the optimization threshold or exceeds the optimization threshold, the LSTM model of pulmonary function rehabilitation training strategy is effectively guaranteed. +The stability and predictability of training effects and continuous optimization of strategy models will lead to better rehabilitation training effects. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 It is a schematic diagram of the system of the present invention;
[0024] Figure 2 It is a step diagram of the method of the present invention. DETAILED DESCRIPTION
[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0026] Because traditional pulmonary function rehabilitation training methods for respiratory patients usually take a long time to achieve the expected results, the rehabilitation efficiency is slow and the cycle is long, and in the actual training process, the pulmonary function rehabilitation training strategy usually relies on medical personnel to formulate it manually based on clinical experience. Over-reliance on clinical experience can easily cause the training strategy to lag behind the latest research results, affecting the stability and predictability of the training effect. Therefore, a pulmonary function rehabilitation training system and method for respiratory patients is provided. Please refer to Figure 1-Figure 2 , including a database, the database is used to store the lung function rehabilitation training data of respiratory patients, and the lung function rehabilitation training system for respiratory patients includes a data acquisition module and a rehabilitation analysis module;
[0027] The data acquisition module includes a theoretical data unit, a patient data unit, and a strategy data unit. The theoretical data unit collects rehabilitation data sets in real time through a network connection to a database. The rehabilitation data set includes all theoretical knowledge related to rehabilitation training for pulmonary dysfunction. The theoretical knowledge includes theoretical evaluation data of pulmonary function, theoretical evaluation data of motor ability, theoretical evaluation data of daily living ability, and theoretical evaluation data of rehabilitation psychology. The expression of the rehabilitation data set is {K gn , K yd , K sh , K xl}, K gn Represents theoretical evaluation data of lung function of respiratory patients, including obstructive ventilatory dysfunction, restrictive pulmonary dysfunction and diffusion dysfunction, etc. yd Represents theoretical evaluation data on the exercise capacity of respiratory patients, including the detailed steps, indications, contraindications and related evaluation indicators of the six-minute walk test (6MWT), such as walking distance, heart rate, degree of dyspnea, etc. shThe Barthe l index is used to evaluate the patient's ability to carry out activities in daily life, such as eating, dressing, and personal hygiene. xl It represents theoretical evaluation data on the rehabilitation psychology of respiratory patients, including depression mood measurement form, HAD mood test form, etc., which are used to evaluate the mental health status of patients during the rehabilitation process;
[0028] The patient data unit collects patient data sets through a network connection database. The patient data set includes the time points and medical records of all respiratory patients who started pulmonary function rehabilitation training. The patient data set expression is: to The data correspond to the time point and medical record data of each respiratory patient who started pulmonary function rehabilitation training. t represents the time point when the respiratory patient started pulmonary function rehabilitation training, which ensures the traceability of historical rehabilitation training data. b represents the medical record data of the respiratory patient, which directly reflects the patient's pulmonary dysfunction status. The subscripts 1 to z represent that there are z respiratory patients who participated in pulmonary function rehabilitation training.
[0029] The strategy data unit collects the strategy data set in real time through the network connection database. The strategy data set includes respiratory exercise strategy, nutritional supplement strategy and psychological support strategy. The strategy data set expression is {C hx , C yy , C xl}, C hx Represents the breathing exercise strategy used for lung function rehabilitation training, including endurance training, strength training, abdominal breathing, pursed lip breathing and other training methods. yy Represents the nutritional supplementation strategy used for pulmonary function rehabilitation training, for example, the types of food and eating habits that enhance the patient's immunity and resistance. xl It indicates the psychological support strategy used for pulmonary function rehabilitation training, providing professional psychological support and encouragement to help patients establish a positive attitude and confidence in rehabilitation;
[0030] The data acquisition module transmits the rehabilitation data set, the patient data set and the strategy data set to the rehabilitation analysis module through the network;
[0031] The rehabilitation analysis module includes a model training unit, a strategy analysis unit and a function optimization unit. The model training unit is provided with a CNN algorithm model. The model training unit substitutes the rehabilitation data set into the CNN algorithm model to construct and generate a real-time updated pulmonary dysfunction rehabilitation training knowledge base ZSK, which is then transmitted to the strategy analysis unit through the network. The calculation formula is as follows:
[0032] ZSK=∑ CNN (K gn +K yd +Ksh +K xl )
[0033] In the formula, ZSK represents the knowledge base of pulmonary dysfunction rehabilitation training, ∑ CNN (K gn +K yd +K sh +K xl ) indicates that the CNN algorithm model is trained based on the real-time updated pulmonary function theoretical evaluation data, motor ability theoretical evaluation data, daily living ability theoretical evaluation data and rehabilitation psychology theoretical evaluation data in the rehabilitation data set, and builds a real-time updated knowledge base for pulmonary dysfunction rehabilitation training, and updates the latest research results and technical standards related to pulmonary dysfunction rehabilitation training in real time;
[0034] The strategy analysis unit is equipped with an LSTM algorithm model. The strategy analysis unit substitutes the strategy data set and the lung dysfunction rehabilitation training knowledge base ZSK into the LSTM algorithm model to construct a pulmonary function rehabilitation training strategy model LSTM. + , and its calculation formula is as follows:
[0035] LSTM + =∑ LSTM ∑ ZSK [(K gn +K yd +C hx )+(K sh +C yy )+(K xl +C xl )]
[0036] In the formula, LSTM + represents the pulmonary function rehabilitation training strategy model, K gn +K yd +C hx K represents the theoretical evaluation data of lung function, theoretical evaluation data of exercise capacity and respiratory exercise strategies for lung function rehabilitation training in the comprehensive lung dysfunction rehabilitation training knowledge base ZSK for respiratory patients, and then substitutes them into the LSTM algorithm model for training. sh +C yy K represents the theoretical evaluation data of daily living ability of respiratory patients in the comprehensive pulmonary dysfunction rehabilitation training knowledge base ZSK and the nutritional supplement strategy for pulmonary function rehabilitation training in the strategy dataset, which is then substituted into the LSTM algorithm model for training. xl +C xl represents the psychological support strategy for pulmonary function rehabilitation training in the rehabilitation psychology theory evaluation data and strategy dataset for respiratory patients in the comprehensive pulmonary dysfunction rehabilitation training knowledge base ZSK, and then substitutes it into the LSTM algorithm model for training, ∑ LSTM ∑ZSK [(K gn +K yd +C hx )+(K sh +C yy )+(K xl +C xl )] indicates that the LSTM algorithm model is trained three times based on the strategy data set and the lung dysfunction rehabilitation training knowledge base ZSK to build a lung function rehabilitation training strategy model, analyze and predict the patient's long-term rehabilitation data, and subsequently formulate a more reasonable rehabilitation training strategy;
[0037] Lung function rehabilitation training strategy model LSTM + According to the patient data set, the training data set Xlsj is calculated and generated. The training data set Xlsj includes the pulmonary dysfunction status of respiratory patients at the current time point, the corresponding pulmonary function rehabilitation training strategy and the predicted pulmonary function status after training. The calculation formula is as follows:
[0038]
[0039] In the formula, Xlsj represents the training data set, represents the time point and medical record data of the i-th patient starting pulmonary function rehabilitation training in the patient data set, Representing the pulmonary function rehabilitation training strategy model LSTM + At the same time, the pulmonary dysfunction status of the z patients in the respiratory department at the current time point in the patient data set is calculated, and the pulmonary function rehabilitation training strategies of the z patients are generated accordingly. gt represents the observation period used to test the effectiveness of the pulmonary function rehabilitation training strategy. Indicates that within a single observation period gt, all z patients in the patient data set are trained according to their respective pulmonary function rehabilitation training strategies. Representing the pulmonary function rehabilitation training strategy model LSTM + At the same time, predict the lung function status of z patients after training;
[0040] The function optimization unit evaluates the pulmonary function rehabilitation training strategy model LSTM based on the patient data set and the training data set Xlsj + The immediate reward value r is calculated and the optimization index Yh zs is generated through the reward and punishment function. The calculation formula is as follows:
[0041] According to the patient data set and the training data set Xlsj, after a single observation period gt, the medical records of z patients in the patient data set who have completed training according to their respective pulmonary function rehabilitation training strategies are counted and marked as H t+g ,in,
[0042] The medical record data H of z patients after trainingt+g Compare the predicted lung function status after training with the training data set Xlsj, and evaluate the lung function rehabilitation training strategy model LSTM + The immediate reward value r is: if the medical record data of a single patient after training is consistent with the predicted lung function status after training in the training data set Xlsj, the immediate reward value r increases by 1; if the medical record data of a single patient after training is inconsistent with the predicted lung function status after training in the training data set Xlsj, the immediate reward value r decreases by 1;
[0043] Yh zs=∑ z [r+kV(t)]
[0044] In the formula, Yhzs represents the optimization index, k represents the discount factor for the immediate reward value r in the optimization index calculation formula, and V(t) represents the value function for the time point when respiratory patients start pulmonary function rehabilitation training, which is the pulmonary function rehabilitation training strategy model LSTM + The sum of the immediate reward values r obtained after the respiratory patients started pulmonary function rehabilitation training, realizing the digital evaluation of the pulmonary function rehabilitation training strategy model LSTM + the degree of effectiveness;
[0045] The function optimization unit is set with an optimization threshold. The function optimization unit compares the optimization index Yhzs with the optimization threshold and then manages the pulmonary function rehabilitation training strategy model LSTM accordingly. + When the optimization index Yhzs is lower than the optimization threshold, it means that the LSTM model of pulmonary function rehabilitation training strategy + The effectiveness is low. The function optimization unit generates an optimization signal and transmits it to the strategy analysis unit. The strategy analysis unit reconstructs and generates a new pulmonary function rehabilitation training strategy model LSTM based on the optimization signal. + Until the optimization index Yhzs is included in the optimization threshold or exceeds the optimization threshold, when the optimization index Yhzs is included in the optimization threshold or exceeds the optimization threshold, it indicates that the pulmonary function rehabilitation training strategy model LSTM + The effectiveness is high, and the function optimization unit does not generate signals, which effectively guarantees the LSTM model of lung function rehabilitation training strategy + The stability and predictability of training effects and continuous optimization of strategy models will lead to better rehabilitation training effects.
[0046] The above-mentioned pulmonary function rehabilitation training system for respiratory patients is implemented by a pulmonary function rehabilitation training method for respiratory patients, and the pulmonary function rehabilitation training method for respiratory patients comprises the following steps:
[0047] Step 1: The data acquisition module obtains all theoretical knowledge related to pulmonary dysfunction rehabilitation training, the time point and medical record data of all respiratory patients starting pulmonary function rehabilitation training, as well as respiratory exercise strategies, nutritional supplement strategies and psychological support strategies through network connection database in real time, and classifies them into rehabilitation data set, patient data set and strategy data set;
[0048] Step 2: The rehabilitation analysis module is provided with a CNN algorithm model. The CNN algorithm model constructs and generates a real-time updated lung dysfunction rehabilitation training knowledge base ZSK based on the rehabilitation data set, and updates the latest research results and technical standards related to lung dysfunction rehabilitation training in real time;
[0049] Step 3: The rehabilitation analysis module is equipped with an LSTM algorithm model, which can better process long time series data. The LSTM algorithm model is constructed based on the rehabilitation data set, strategy data set and the lung dysfunction rehabilitation training knowledge base ZSK to generate the lung function rehabilitation training strategy model LSTM. + , analyze and predict the patient's long-term rehabilitation data, and formulate more reasonable rehabilitation training strategies in the future;
[0050] Step 4: Lung Function Rehabilitation Training Strategy Model LSTM + According to the patient data set, the training data set Xlsj is calculated and generated, and the pulmonary dysfunction status of respiratory patients at the current time point, the corresponding pulmonary function rehabilitation training strategy and the predicted pulmonary function status after training are analyzed;
[0051] Step 5: The rehabilitation analysis module evaluates the pulmonary function rehabilitation training strategy model LSTM based on the patient data set and the training data set Xlsj + The immediate reward value r is calculated and the optimization index Yh zs is generated through the reward and punishment function, realizing the digital evaluation of the pulmonary function rehabilitation training strategy model LSTM + the degree of effectiveness;
[0052] Step 6: The rehabilitation analysis module is set with an optimization threshold, and the optimization index Yh zs is compared with the optimization threshold, corresponding to the management of the pulmonary function rehabilitation training strategy model LSTM + , effectively ensuring the LSTM model of lung function rehabilitation training strategy + The stability and predictability of training effects and continuous optimization of strategy models will lead to better rehabilitation training effects.
[0053] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A pulmonary function rehabilitation training system for respiratory patients, comprising a database for storing pulmonary function rehabilitation training data of respiratory patients, characterized in that: The lung function rehabilitation training system for respiratory patients comprises a data acquisition module and a rehabilitation analysis module.
2. A pulmonary function rehabilitation training system for respiratory patients according to claim 1, characterized in that: The data acquisition module includes a theoretical data unit, a patient data unit and a strategy data unit. The theoretical data unit acquires a rehabilitation data set in real time through a network connection to a database. The rehabilitation data set includes all theoretical knowledge related to rehabilitation training for pulmonary dysfunction.
3. A pulmonary function rehabilitation training system for respiratory patients according to claim 2, characterized in that: Theoretical knowledge includes theoretical evaluation data of lung function, theoretical evaluation data of motor ability, theoretical evaluation data of daily living ability and theoretical evaluation data of rehabilitation psychology. The patient data unit collects patient data sets through a network connection to a database. The patient data sets include the time points and medical records of all respiratory patients starting lung function rehabilitation training. The strategy data unit collects strategy data sets in real time through a network connection to a database. The strategy data sets include respiratory exercise strategies, nutritional supplement strategies and psychological support strategies. The data acquisition module transmits the rehabilitation data sets, patient data sets and strategy data sets to the rehabilitation analysis module through the network.
4. A pulmonary function rehabilitation training system for respiratory patients according to claim 3, characterized in that: The rehabilitation analysis module includes a model training unit, a strategy analysis unit and a function optimization unit. The model training unit is provided with a CNN algorithm model. The model training unit substitutes the rehabilitation data set into the CNN algorithm model to construct and generate a real-time updated pulmonary dysfunction rehabilitation training knowledge base ZSK, and transmits it to the strategy analysis unit through the network.
5. A pulmonary function rehabilitation training system for respiratory patients according to claim 4, characterized in that: The strategy analysis unit is provided with an LSTM algorithm model. The strategy analysis unit substitutes the strategy data set and the pulmonary dysfunction rehabilitation training knowledge base ZSK into the LSTM algorithm model to construct a pulmonary function rehabilitation training strategy model LSTM. + , the pulmonary function rehabilitation training strategy model LSTM + According to the patient data set, a training data set Xlsj is calculated and generated, wherein the training data set Xlsj includes the pulmonary dysfunction status of the respiratory patients at the current time point, the corresponding pulmonary function rehabilitation training strategy and the predicted pulmonary function status after training, and the function optimization unit evaluates the pulmonary function rehabilitation training strategy model LSTM according to the patient data set and the training data set Xlsj. + The immediate reward value r is calculated and the optimization index Yh zs is generated by reward and punishment function. The function optimization unit is provided with an optimization threshold. The function optimization unit compares the optimization index Yh zs with the optimization threshold and then manages the pulmonary function rehabilitation training strategy model LSTM accordingly. + .
6. A pulmonary function rehabilitation training system for respiratory patients according to claim 5, characterized in that: The rehabilitation data set expression is {K gn , K yd , K sh , K xl }, K gn represents the theoretical evaluation data of lung function for respiratory patients, K yd represents the theoretical evaluation data of the exercise capacity of respiratory patients, K sh represents the theoretical evaluation data of the daily living ability of respiratory patients, K xl Represents theoretical evaluation data on the rehabilitation psychology of respiratory patients.
7. A pulmonary function rehabilitation training system for respiratory patients according to claim 6, characterized in that: The patient dataset expression is: to The data correspond to the time point and medical record data of each respiratory patient starting pulmonary function rehabilitation training. t represents the time point when the respiratory patient starts pulmonary function rehabilitation training, b represents the medical record data of the respiratory patient, and the subscripts 1 to z represent that there are z respiratory patients participating in pulmonary function rehabilitation training.
8. A pulmonary function rehabilitation training system for respiratory patients according to claim 7, characterized in that: The strategy data set expression is {C hx , C yy , C xl }, C hx represents the breathing exercise strategy used for pulmonary function rehabilitation training, C yy Represents the nutritional supplement strategy for pulmonary function rehabilitation training, C xl To represent psychological support strategies used in pulmonary rehabilitation training.
9. A method for pulmonary function rehabilitation training of respiratory patients, applied to the pulmonary function rehabilitation training system for respiratory patients according to claims 1-8, characterized in that: The following steps are involved: Step 1: The data acquisition module obtains all theoretical knowledge related to pulmonary dysfunction rehabilitation training, the time point and medical record data of all respiratory patients starting pulmonary function rehabilitation training, as well as respiratory exercise strategies, nutritional supplement strategies and psychological support strategies through network connection database in real time, and classifies them into rehabilitation data set, patient data set and strategy data set; Step 2: The rehabilitation analysis module is provided with a CNN algorithm model, and the CNN algorithm model constructs and generates a real-time updated pulmonary dysfunction rehabilitation training knowledge base ZSK based on the rehabilitation data set; Step 3: The rehabilitation analysis module is equipped with an LSTM algorithm model. The LSTM algorithm model builds a pulmonary function rehabilitation training strategy model LSTM based on the rehabilitation data set, strategy data set and pulmonary dysfunction rehabilitation training knowledge base ZSK. + ; Step 4: Lung Function Rehabilitation Training Strategy Model LSTM + According to the patient data set, the training data set Xlsj is calculated and generated, and the pulmonary dysfunction status of respiratory patients at the current time point, the corresponding pulmonary function rehabilitation training strategy and the predicted pulmonary function status after training are analyzed; Step 5: The rehabilitation analysis module evaluates the pulmonary function rehabilitation training strategy model LSTM based on the patient data set and the training data set Xlsj + The immediate reward value r is calculated and the optimization index Yh zs is generated through the reward and punishment function; Step 6: The rehabilitation analysis module is set with an optimization threshold, and the optimization index Yh zs is compared with the optimization threshold, corresponding to the management of the pulmonary function rehabilitation training strategy model LSTM + .
Citation Information
Cited By
Rehabilitation medical training management system
CN119517296A
