Respiratory rehabilitation training system

Abstract

The invention relates to the technical field of respiratory rehabilitation training, in particular to a respiratory rehabilitation training system which comprises rehabilitation training equipment and a control subsystem. The control subsystem comprises a human-computer interaction interface, an encryption transmission unit, a multi-mode signal acquisition unit, a preprocessing unit, a feature extraction module, a feature fusion module, an analysis processing module, a control module, a feedback module and an optimization module; the control subsystem is implanted in rehabilitation training equipment, the human-computer interaction interface is connected with the analysis processing module through the encryption transmission unit, the preprocessing unit is connected with the multi-mode signal acquisition unit, the feature extraction module is connected with the preprocessing unit, and the feature fusion module is connected with the feature extraction module. In this way, the physiological features of the user can be monitored in real time in the training process, the training state and physiological reaction of the patient can be comprehensively and timely mastered, the training scheme is dynamically adjusted according to the individual difference and the real-time physical condition, and the training effect is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of respiratory rehabilitation training, and in particular to a respiratory rehabilitation training system. BACKGROUND

[0002] As an important medical means, respiratory rehabilitation training plays a key role in improving the lung function of patients with respiratory system diseases, enhancing exercise tolerance, and improving the quality of life. With the continuous progress of medical technology and the increasing emphasis on health, respiratory rehabilitation training has gradually received widespread attention, and its application scope has also been expanding, covering the treatment and rehabilitation of various diseases such as chronic obstructive pulmonary disease (COPD), asthma, and postoperative respiratory function recovery.

[0003] However, in the existing mode, the training process lacks real-time monitoring and accurate feedback, and medical staff cannot fully and timely grasp the training state and physiological response of patients, and cannot dynamically adjust the training scheme according to the individual differences and real-time physical condition of patients, resulting in uneven training effect, and some patients may not obtain the best rehabilitation effect. SUMMARY

[0004] The purpose of the present application is to provide a respiratory rehabilitation training system, which aims to solve the technical problems in the prior art that the training process lacks real-time monitoring and accurate feedback, medical staff cannot fully and timely grasp the training state and physiological response of patients, and cannot dynamically adjust the training scheme according to the individual differences and real-time physical condition of patients, resulting in uneven training effect, and some patients may not obtain the best rehabilitation effect.

[0005] To achieve the above-mentioned purpose, a respiratory rehabilitation training system is adopted, which comprises a rehabilitation training device and a control subsystem, the control subsystem comprises a human-computer interaction interface, an encryption transmission unit, a multi-modal signal acquisition unit, a preprocessing unit, a feature extraction module, a feature fusion module, an analysis processing module, a control module, a feedback module and an optimization module;

[0006] The control subsystem is implanted in the rehabilitation training device, the human-computer interaction interface is connected with the analysis processing module through the encryption transmission unit, the preprocessing unit is connected with the multi-modal signal acquisition unit, the feature extraction module is connected with the preprocessing unit, the feature fusion module is connected with the feature extraction module, the feature fusion module is also connected with the analysis processing module, the control module is connected with the analysis processing module, the optimization module is connected with the analysis processing module, and the feedback module is connected with the optimization module and the human-computer interaction interface;

[0007] The rehabilitation training device is used to perform respiratory rehabilitation training tasks;

[0008] The human-computer interaction interface interacts with the patient, facilitating the patient to input personal information and medical records, which are then transmitted to the processing and analysis module via the encrypted transmission unit;

[0009] The multimodal signal acquisition unit is used to acquire multimodal physiological signals of the patient's body;

[0010] The preprocessing unit is used to preprocess the collected multimodal signals and transmit them to the processor module;

[0011] The feature extraction module is used to extract key features from the preprocessed signal, and then the feature fusion module fuses multimodal features to eliminate redundant information and generate a comprehensive physiological state representation;

[0012] The analysis and processing module analyzes the patient's real-time respiratory status based on the fusion features and generates a real-time control strategy according to the real-time respiratory status and the patient's personal information;

[0013] The control module dynamically adjusts the parameters of the rehabilitation training equipment through a real-time control strategy.

[0014] The multimodal signal acquisition unit includes a respiratory sensor, a heart rate sensor, a blood oxygen saturation sensor and an electromyographic sensor, and the respiratory sensor, the heart rate sensor, the blood oxygen saturation sensor and the electromyographic sensor are all connected to the preprocessing unit.

[0015] Among them, the preprocessing unit includes an edge computing module, a time alignment module and a space alignment module. The edge computing modules are connected to the multimodal signal acquisition unit and the feature extraction module, and the time alignment module and the space alignment module are both connected to the edge computing module.

[0016] Among them, the encrypted transmission unit includes a transmission module, an encryption module, a decryption module, a key management module and a key generation module. The transmission modules are connected to the human-computer interaction interface and the analysis and processing module. The encryption module and the decryption module are connected to the human-computer interaction interface and the analysis and processing module respectively. The key management modules are connected to the encryption module, the decryption module and the key generation module.

[0017] Among them, the control subsystem also includes a dynamic weighting module, which is connected to the feature fusion module. The dynamic weighting module dynamically adjusts the weights of different modal data in the fusion based on signal quality evaluation, and gives priority to retaining high-confidence signals.

[0018] The control subsystem further comprises a dynamic noise reduction module connected with the preprocessing unit, which suppresses noise in the multi-modal signal in real time through an adaptive filtering algorithm and adjusts filtering parameters according to signal types.

[0019] The control subsystem further comprises a training mode switching module connected with the analysis processing module and a voice control module connected with the training mode switching module.

[0020] The control subsystem further comprises a training log recording module connected with the analysis processing module, which records training logs using a multi-node redundant storage architecture.

[0021] In specific use, the patient inputs personal information and case through the human-computer interaction interface, and key data is securely transmitted to the analysis processing module through the encryption transmission unit. The multi-modal signal acquisition unit captures the patient's multi-modal physiological signals in all directions, covering multi-dimensional information such as respiratory airflow, chest and abdominal movement amplitude, blood oxygen saturation, heart rate variability, etc., to realize comprehensive monitoring of the patient's respiratory physiological state. The preprocessing unit performs a series of preprocessing operations such as filtering, noise reduction, and normalization on the collected multi-modal signals. The feature extraction module uses advanced signal processing algorithms to accurately extract key features closely related to the respiratory state from the preprocessed signals. The feature fusion module deeply fuses features of different modalities, eliminates redundant information through intelligent algorithms, and generates a comprehensive physiological state representation that can fully and accurately represent the patient's real-time physiological state. The analysis processing module, based on the fused features, combines the patient's pre-input personal information and case data, uses deep learning and machine learning algorithms to deeply analyze the patient's real-time respiratory state, accurately judges the patient's current respiratory function status, potential respiratory risks, and rehabilitation progress, and generates real-time control strategies accordingly. After receiving the control strategy, the control module dynamically adjusts the parameters of the rehabilitation training equipment, such as adjusting the respiratory resistance, training intensity, training duration, etc., so that the training process always matches the patient's actual needs and physical condition. In this way, the technical problem of the prior art that the training process lacks real-time monitoring and accurate feedback, making it difficult for medical staff to fully and timely grasp the patient's training state and physiological response, and unable to dynamically adjust the training scheme according to the patient's individual differences and real-time physical condition, resulting in uneven training effectiveness and some patients may not achieve the best rehabilitation effect is solved. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments or prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative effort based on these drawings.

[0023] Figure 1 is a principle block diagram of the first embodiment of the present application.

[0024] Figure 2 is a principle block diagram of the second embodiment of the present application.

[0025] Figure 3 is a principle block diagram of the third embodiment of the present application.

[0026] 101-rehabilitation training device, 102-control subsystem, 103-human-computer interaction interface, 104-encryption transmission unit, 105-multimodal signal acquisition unit, 106-preprocessing unit, 107-feature extraction module, 108-feature fusion module, 109-analysis processing module, 110-control module, 111-feedback module, 112-optimization module, 113-breathing sensor, 114-heart rate sensor, 115-oxygen saturation sensor, 116-electromyography sensor, 117-edge computing module, 118-time alignment module, 119-space alignment module, 120-transmission module, 121-encryption module, 122-decryption module, 123-key management module, 124-key generation module, 201-dynamic weighting module, 202-dynamic noise reduction module, 203-training mode switching module, 204-voice control module, 205-training log recording module, 301-rehabilitation period prediction module, 302-report pushing module, 303-maintenance module. DETAILED DESCRIPTION

[0027] The embodiments of the present application will be described in detail below, examples of which are shown in the accompanying drawings. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application.

[0028] The first embodiment of the present application is:

[0029] Please refer to Figure 1 , Figure 1 is a principle block diagram of the first embodiment of the present application.

[0030] The present invention provides a respiratory rehabilitation training system, comprising a rehabilitation training device 101 and a control subsystem 102, wherein the control subsystem 102 comprises a human-computer interaction interface 103, an encryption transmission unit 104, a multimodal signal acquisition unit 105, a preprocessing unit 106, a feature extraction module 107, a feature fusion module 108, an analysis and processing module 109, a control module 110, a feedback module 111 and an optimization module 112; the multimodal signal acquisition unit 105 comprises a respiratory sensor 113, a heart rate sensor 114, a blood oxygen saturation sensor 115 and an electromyographic sensor 116, the preprocessing unit 10 6 includes an edge computing module 117, a time alignment module 118 and a space alignment module 119, and the encrypted transmission unit 104 includes a transmission module 120, an encryption module 121, a decryption module 122, a key management module 123 and a key generation module 124. The above solution solves the technical problem that the training process in the existing technology lacks real-time monitoring and accurate feedback, and it is difficult for medical staff to fully and timely grasp the patient's training status and physiological reactions. It is impossible to dynamically adjust the training plan according to the patient's individual differences and real-time physical condition, resulting in uneven training effects. Some patients may not be able to obtain the best rehabilitation effect.

[0031] For this specific embodiment, the rehabilitation training device 101 is used to perform respiratory rehabilitation training tasks;

[0032] The human-computer interaction interface 103 interacts with the patient, facilitating the patient to input personal information and medical records, which are then transmitted to the processing and analysis module via the encrypted transmission unit 104;

[0033] The multimodal signal acquisition unit 105 is used to acquire multimodal physiological signals of the patient's body;

[0034] The pre-processing unit 106 is used to pre-process the collected multi-modal signals and transmit them to the processor module;

[0035] The feature extraction module 107 is used to extract key features from the preprocessed signal, and then the feature fusion module 108 fuses multimodal features to eliminate redundant information and generate a comprehensive physiological state representation;

[0036] The analysis and processing module 109 analyzes the patient's real-time respiratory status based on the fusion features and generates a real-time control strategy according to the real-time respiratory status and the patient's personal information;

[0037] The control module 110 dynamically adjusts the parameters of the rehabilitation training device 101 through a real-time control strategy.

[0038] The control subsystem 102 is implanted in the rehabilitation training device 101, the human-computer interaction interface 103 is connected with the analysis processing module 109 through the encryption transmission unit 104, the preprocessing unit 106 is connected with the multi-modal signal acquisition unit 105, the feature extraction module 107 is connected with the preprocessing unit 106, the feature fusion module 108 is connected with the feature extraction module 107, the feature fusion module 108 is also connected with the analysis processing module 109, the control module 110 is connected with the analysis processing module 109, the optimization module 112 is connected with the analysis processing module 109, and the feedback module 111 is connected with the optimization module 112 and the human-computer interaction interface 103. In specific use, the patient inputs personal information and cases through the human-computer interaction interface 103, and the key data is safely transmitted to the analysis processing module 109 through the encryption transmission unit 104, the multi-modal signal acquisition unit 105 captures the human multi-modal physiological signals of the patient in all directions, covering respiratory airflow, chest and abdominal movement amplitude, blood oxygen saturation, heart rate variability and other multi-dimensional information, realizing comprehensive monitoring of the respiratory physiological state of the patient, the preprocessing unit 106 performs a series of preprocessing operations such as filtering, noise reduction and normalization on the collected multi-modal signals, the feature extraction module 107 uses advanced signal processing algorithms to accurately extract key features closely related to the respiratory state from the preprocessed signals, the feature fusion module 108 deeply fuses the features of different modalities, eliminates redundant information through intelligent algorithms, and generates a comprehensive physiological state representation that can fully and accurately represent the real-time physiological state of the patient, the analysis processing module 109 is based on the fused features, combined with the pre-input personal information and case data of the patient, uses deep learning and machine learning algorithms to deeply analyze the real-time respiratory state of the patient, accurately judges the current respiratory function status, potential respiratory risk and rehabilitation progress of the patient, and generates a real-time control strategy accordingly. After the control module 110 receives the control strategy, the parameters of the rehabilitation training device 101 are dynamically adjusted, such as adjusting the respiratory resistance, training intensity, training duration, etc., so that the training process always matches the actual needs and physical condition of the patient, thereby solving the technical problems in the prior art that the training process lacks real-time monitoring and accurate feedback, it is difficult for medical staff to fully and timely grasp the training state and physiological response of the patient, and it is impossible to dynamically adjust the training scheme according to the individual differences and real-time physical condition of the patient, resulting in uneven training effect, and some patients may not obtain the best rehabilitation effect.

[0039] Secondly, the respiratory sensor 113, the heart rate sensor 114, the blood oxygen saturation sensor 115 and the electromyography sensor 116 are connected with the preprocessing unit 106, and in specific use, the respiratory sensor 113 is used for monitoring respiratory frequency, respiratory depth and respiratory rhythm; the heart rate sensor 114 is used for monitoring heart rate variation; the blood oxygen saturation sensor 115 is used for monitoring blood oxygen saturation; and the electromyography sensor 116 is used for monitoring electromyography signal of respiratory muscle.

[0040] Meanwhile, the edge computing module 117 is connected with the multi-modal signal acquisition unit 105 and the feature extraction module 107, the time alignment module 118 and the space alignment module 119 are connected with the edge computing module 117, the edge computing module 117 is used for preliminarily processing the collected multi-modal signal and reducing data transmission amount; the time alignment module 118 is used for time synchronization processing of different modal signals and ensuring alignment of the signals in time dimension; and the space alignment module 119 is used for space calibration of the multi-modal signal and ensuring alignment of the signals in space dimension.

[0041] In addition, the transmission module 120 is connected with the man-machine interaction interface 103 and the analysis processing module 109, the encryption module 121 and the decryption module 122 are connected with the man-machine interaction interface 103 and the analysis processing module 109 respectively, and the key management module 123 is connected with the encryption module 121, the decryption module 122 and the key generation module 124. During operation of the respiratory rehabilitation training system, when the patient inputs personal information, case data or training feedback and other sensitive information through the man-machine interaction interface 103, the key generation module 124 immediately responds and generates a unique encryption key according to a random algorithm. The key management module 123 safely stores the key and synchronously transmits it to the encryption module 121. The encryption module 121 uses the key and adopts an encryption algorithm (such as AES high-level encryption standard) to encrypt data to be sent to the analysis processing module 109 through the transmission module 120 in real time, and converts the original data into ciphertext. The data is transmitted in the form of ciphertext in the transmission module 120, and after reaching the analysis processing module 109, the decryption module 122 obtains the corresponding key from the key management module 123, uses a decryption algorithm matched with the encryption algorithm to restore the ciphertext to the original data, for subsequent analysis and processing.

[0042] The respiratory rehabilitation training system of the embodiment is used in specific use. A patient inputs personal information and a case through the man-machine interaction interface 103. Key data is safely transmitted to the analysis processing module 109 through the encrypted transmission unit 104. The multi-modal signal acquisition unit 105 comprehensively captures multi-modal physiological signals of the patient, including respiratory airflow, chest and abdominal movement amplitude, blood oxygen saturation, heart rate variability and other multi-dimensional information, so as to comprehensively monitor the respiratory physiological state of the patient. The preprocessing unit 106 performs filtering, noise reduction, normalization and other series of preprocessing operations on the collected multi-modal signals. The feature extraction module 107 uses advanced signal processing algorithms to accurately extract key features closely related to the respiratory state from the preprocessed signals. The feature fusion module 108 deeply fuses features of different modalities, eliminates redundant information through intelligent algorithms, and generates a comprehensive physiological state representation that can comprehensively and accurately represent the real-time physiological state of the patient. The analysis processing module 109 analyzes the real-time respiratory state of the patient based on the fused features, combines the pre-input personal information and case data of the patient, uses deep learning and machine learning algorithms to accurately judge the current respiratory function status, potential respiratory risks and rehabilitation progress of the patient, and generates real-time control strategies accordingly. After receiving the control strategies, the control module 110 dynamically adjusts the parameters of the rehabilitation training device 101, such as adjusting the respiratory resistance, training intensity, training duration and the like, so that the training process always matches the actual needs and physical condition of the patient. In this way, the technical problem that the training process lacks real-time monitoring and accurate feedback in the prior art, it is difficult for medical staff to comprehensively and timely grasp the training state and physiological response of the patient, and it is impossible to dynamically adjust the training scheme according to the individual differences and real-time physical condition of the patient, resulting in uneven training effect and the possibility that some patients cannot obtain the best rehabilitation effect is solved.

[0043] The second embodiment of the present application is:

[0044] Based on the first embodiment, please refer to Figure 2 , Figure 2 is the principle block diagram of the second embodiment of the present application.

[0045] The present application provides a respiratory rehabilitation training system, and the control subsystem 102 further comprises a dynamic weighting module 201, a dynamic noise reduction module 202, a training mode switching module 203, a voice control module 204 and a training log recording module 205.

[0046] For this specific embodiment, the dynamic weighting module 201 is connected with the feature fusion module 108. The dynamic weighting module 201 dynamically adjusts the weight of different modal data in fusion based on signal quality, and preferentially retains high-confidence signals.

[0047] The dynamic noise reduction module 202 is connected with the preprocessing unit 106, and the noise in the multi-modal signal is inhibited in real time through an adaptive filtering algorithm, and the filtering parameter is adjusted according to the signal type.

[0048] Secondly, the training mode switching module 203 is connected with the analysis processing module 109, and the voice control module 204 is connected with the training mode switching module 203, so that the user can switch the mode through the voice control module 204, and can select the training in different environments, such as walking or climbing.

[0049] Secondly, the training log recording module 205 is connected with the analysis processing module 109, and the training log recording module 205 adopts a multi-node redundant storage architecture to record the training log, and the training log recording module 205 can realize high reliability, low cost, and strong expansion of the log management capability, and provides data support for the precision and individualization of respiratory rehabilitation training.

[0050] The respiratory rehabilitation training system of the embodiment is used in specific use, the dynamic weighting module 201 adjusts the weight of different modal data in fusion based on signal quality evaluation, and high confidence signals are preferentially retained. The user can switch the mode through the voice control module 204, and can select the training in different environments, such as walking or climbing. The training log recording module 205 adopts a multi-node redundant storage architecture to record the training log, and the training log recording module 205 can realize high reliability, low cost, and strong expansion of the log management capability, and provides data support for the precision and individualization of respiratory rehabilitation training.

[0051] The third embodiment of the present application is:

[0052] On the basis of the second embodiment, please refer to Figure 3 , Figure 3 is the principle block diagram of the third embodiment of the present application.

[0053] The present application provides a respiratory rehabilitation training system, and the control subsystem 102 further comprises a rehabilitation period prediction module 301, a report pushing module 302 and a maintenance module 303.

[0054] For the specific embodiment, the rehabilitation period prediction module 301 is connected with the analysis processing module 109, the report pushing module 302 is connected with the rehabilitation period prediction module 301, the rehabilitation period prediction module 301 can predict the rehabilitation course based on the rehabilitation training data, and push to the user end through the pushing module.

[0055] The maintenance module 303 is connected with the analysis processing module 109, and is configured to maintain the analysis processing module 109 when maintenance is needed.

[0056] In use, the rehabilitation period prediction module 301 is connected with the analysis processing module 109, the report pushing module 302 is connected with the rehabilitation period prediction module 301, the rehabilitation period prediction module 301 can predict a rehabilitation course based on rehabilitation training data, and push the rehabilitation course to a user terminal through the pushing module, and the maintenance module 303 is configured to maintain the analysis processing module 109 when maintenance is needed.

[0057] The above only discloses one preferred embodiment of the present application, and of course cannot limit the scope of the present application, and those skilled in the art can understand that all or part of the above-mentioned embodiments can be implemented, and equivalent changes made according to the claims of the present application still belong to the scope of the present application.

Claims

1. A respiratory rehabilitation training system, characterized in that: It includes rehabilitation training equipment and a control subsystem, wherein the control subsystem includes a human-computer interaction interface, an encryption transmission unit, a multimodal signal acquisition unit, a preprocessing unit, a feature extraction module, a feature fusion module, an analysis and processing module, a control module, a feedback module and an optimization module; The control subsystem is implanted in the rehabilitation training device, the human-computer interaction interface is connected to the analysis and processing module through the encryption transmission unit, the preprocessing unit is connected to the multimodal signal acquisition unit, the feature extraction module is connected to the preprocessing unit, the feature fusion module is connected to the feature extraction module, the feature fusion module is also connected to the analysis and processing module, the control module is connected to the analysis and processing module, the optimization module is connected to the analysis and processing module, and the feedback module is connected to the optimization module and the human-computer interaction interface; The rehabilitation training device is used to perform respiratory rehabilitation training tasks; The human-computer interaction interface interacts with the patient, facilitating the patient to input personal information and medical records, which are then transmitted to the processing and analysis module via the encrypted transmission unit; The multimodal signal acquisition unit is used to acquire multimodal physiological signals of the patient's body; The preprocessing unit is used to preprocess the collected multimodal signals and transmit them to the processor module; The feature extraction module is used to extract key features from the preprocessed signal, and then the feature fusion module fuses multimodal features to eliminate redundant information and generate a comprehensive physiological state representation; The analysis and processing module analyzes the patient's real-time respiratory status based on the fusion features and generates a real-time control strategy according to the real-time respiratory status and the patient's personal information; The control module dynamically adjusts the parameters of the rehabilitation training device through a real-time control strategy.

2. The respiratory rehabilitation training system according to claim 1, characterized in that: The multimodal signal acquisition unit includes a respiratory sensor, a heart rate sensor, a blood oxygen saturation sensor and an electromyographic sensor, and the respiratory sensor, the heart rate sensor, the blood oxygen saturation sensor and the electromyographic sensor are all connected to the preprocessing unit.

3. The respiratory rehabilitation training system according to claim 2, wherein: The preprocessing unit includes an edge computing module, a time alignment module and a space alignment module. The edge computing modules are connected to the multimodal signal acquisition unit and the feature extraction module, and the time alignment module and the space alignment module are both connected to the edge computing module.

4. The respiratory rehabilitation training system according to claim 3, wherein: The encrypted transmission unit includes a transmission module, an encryption module, a decryption module, a key management module and a key generation module. The transmission modules are connected to the human-computer interaction interface and the analysis and processing module. The encryption module and the decryption module are connected to the human-computer interaction interface and the analysis and processing module respectively. The key management modules are connected to the encryption module, the decryption module and the key generation module.

5. The respiratory rehabilitation training system according to claim 4, characterized in that: The control subsystem also includes a dynamic weighting module, which is connected to the feature fusion module. The dynamic weighting module dynamically adjusts the weights of different modal data in the fusion based on signal quality evaluation, and gives priority to retaining high-confidence signals.

6. The respiratory rehabilitation training system according to claim 5, characterized in that: The control subsystem further includes a dynamic noise reduction module, which is connected to the preprocessing unit and suppresses noise in the multimodal signal in real time through an adaptive filtering algorithm and adjusts filtering parameters according to the signal type.

7. The respiratory rehabilitation training system according to claim 6, characterized in that: The control subsystem further includes a training mode switching module and a voice control module. The training mode switching module is connected to the analysis and processing module, and the voice control module is connected to the training mode switching module.

8. The respiratory rehabilitation training system according to claim 7, characterized in that: The control subsystem further includes a training log recording module, which is connected to the analysis and processing module. The training log recording module adopts a multi-node redundant storage architecture to record the training log.

Images