Supervision method and device for respiratory rehabilitation training of old people

Abstract

The invention discloses a monitoring method and device for respiratory rehabilitation training of old people. The method comprises the steps that a first key gait image with a first key gait is acquired; acquiring a first balance state, a first rotation angle and a first chest contour size of the elderly patient in the first key gait, and comparing the information with a reference balance state, a reference rotation angle and a reference chest contour size to determine whether the elderly patient has an abnormal condition; and under the condition that the elderly patient is not abnormal, taking the first key gait image, the first balance state, the first rotation angle and the first chest contour size of the elderly patient and the information marked as normal as training data to train an abnormality monitoring model, and obtaining the trained abnormality monitoring model. By the adoption of the method, key gait recognition, physiological parameter monitoring and machine learning technologies are combined, and an intelligent and efficient supervision means is provided for old people respiratory rehabilitation training.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical diagnosis, and in particular to a supervision method for respiratory rehabilitation training for the elderly. Background Art

[0002] With the acceleration of the aging process of the population, the number of elderly patients with respiratory diseases is increasing, and the importance of elderly respiratory rehabilitation training has become increasingly prominent, which aims to help elderly patients restore their respiratory function and improve their ability to take care of themselves. However, the existing elderly respiratory rehabilitation training technology is facing many difficulties, especially in terms of safety protection / supervision.

[0003] At present, respiratory rehabilitation training for the elderly mainly relies on a combination of manual monitoring and simple monitoring equipment. In terms of monitoring accuracy, traditional monitoring equipment has limited functions and can usually only measure basic physiological parameters such as respiratory rate, blood pressure, and blood oxygen. The monitoring accuracy is seriously insufficient and real-time process monitoring cannot be achieved, making it difficult for medical staff to accurately know the subtle changes in the training of elderly patients and adjust the training strategy in time. If manual monitoring is used, there will be even more problems. Once an emergency situation occurs, such as sudden breathing difficulties or sports accidents, the reaction speed of people is restricted by physiological factors, making it difficult to detect and respond quickly, which undoubtedly poses a huge safety hazard to rehabilitation training. Moreover, each link is isolated and helpless, lacks an automatic linkage mechanism, and cannot quickly work together to ensure the safety of the lives of elderly patients.

[0004] In summary, the existing elderly respiratory rehabilitation training technology has many drawbacks in supervision methods, and it is urgent to introduce innovative means and intelligent technologies to fill these shortcomings and meet the growing rehabilitation needs of elderly patients. Summary of the invention

[0005] The embodiments of the present application provide a supervision method and device for respiratory rehabilitation training for the elderly, which are used to at least solve the technical problems mentioned above.

[0006] An embodiment of the present application provides a supervision method for respiratory rehabilitation training for the elderly, comprising: processing training images of an elderly patient during motor rehabilitation training through a key gait recognition algorithm to obtain a first key gait image with a first key gait; based on time information corresponding to the first key gait image, obtaining a first balance state, a first rotation angle, and a first chest contour size of the elderly patient in the first key gait, and using the first balance state, the first rotation angle, and the first chest contour size to compare with a reference balance state, a reference rotation angle, and a reference chest contour size, respectively, to determine whether the elderly patient has an abnormal condition; if the elderly patient does not have an abnormal condition, using the first key gait image, the first balance state, the first rotation angle, and the first chest contour size of the elderly patient, as well as information marked as "normal" as training data to train an abnormal monitoring model to obtain a trained abnormal monitoring model.

[0007] Optionally, the method further includes: acquiring a second key gait image having a second key gait; inputting the second key gait image into a trained abnormality monitoring model to determine whether the elderly patient has an abnormal condition.

[0008] Optionally, the first key gait image includes a stance phase start image, a stance phase mid-image, a stance phase end image, a swing phase start image, a swing phase mid-image or a swing phase end image.

[0009] Optionally, the key gait recognition algorithm includes a gait recognition model that has been trained using labeled gait images of various stages.

[0010] Optionally, the first equilibrium state and the first rotation angle are data acquired by using a gyroscope and an accelerometer worn on the elderly patient, and the first chest contour size is data acquired by a belt sensor worn on the elderly patient.

[0011] Optionally, the method further includes: when the elderly patient has an abnormality, synchronizing the training status of the elderly patient to a medical terminal corresponding to the elderly patient and receiving an updated exercise rehabilitation training program sent by the medical terminal.

[0012] Optionally, the method further includes: if the elderly patient has an abnormal condition, sending a reminder message and a first key gait image of the elderly patient to an elderly patient terminal of the elderly patient.

[0013] Optionally, the method further includes: continuously tracking the exercise performance of the elderly patient during the exercise rehabilitation training, and evaluating and optimizing the exercise rehabilitation training program of the elderly patient based on whether the elderly patient has any abnormal conditions.

[0014] An embodiment of the present application also provides a monitoring device for respiratory rehabilitation training for the elderly, the device comprising: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to perform the above-mentioned monitoring method.

[0015] An embodiment of the present application also provides a computer-readable storage medium on which computer instructions are stored, and the above-mentioned supervision method is implemented when the instructions are executed.

[0016] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0017] According to the supervision method of elderly respiratory rehabilitation training of the exemplary embodiment of the present application, it is possible to determine whether the elderly patient has abnormalities by comparing the multidimensional data on the key gait, and in this way, the patient's training status can be comprehensively and accurately evaluated. Significant differences in any dimension can trigger abnormal judgment in time to ensure that no possible abnormal situation is missed, and after obtaining the positive sample, the abnormal training model dedicated to the patient is trained by the positive sample, and as the positive samples of the patient increase, the accuracy of the abnormal training model increases, which is convenient for the subsequent use of the abnormal training model dedicated to the patient to directly supervise the patient's training situation. In the case of abnormalities in the elderly patient, the training situation can be quickly synchronized to the medical terminal, and the updated rehabilitation training program can be received to ensure that the patient can get professional guidance and adjustment in time. At the same time, a reminder message is sent to the elderly patient terminal to improve the patient's participation and self-management ability. In addition, the method can continuously track the patient's sports performance, and evaluate and optimize the training program according to whether there is an abnormality, to ensure that the training program always meets the patient's actual needs. This dynamic adjustment method helps patients gradually improve the training effect and promote the rehabilitation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 is a scene diagram showing a supervision system for respiratory rehabilitation training for the elderly according to an embodiment of the present application.

[0020] Figure 2 is a schematic diagram showing a supervision method for respiratory rehabilitation training for the elderly according to an embodiment of the present application.

[0021] Figure 3is a schematic flow chart showing a method for supervising respiratory rehabilitation training for the elderly according to an embodiment of the present application.

[0022] Figure 4 is a block diagram illustrating a monitoring device for respiratory rehabilitation training for the elderly according to an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0024] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0025] As described in the background technology, in order to improve respiratory function and improve the quality of life, respiratory rehabilitation training is essential for the elderly. During the respiratory rehabilitation training for the elderly, supervision is a key link to ensure the effectiveness and safety of the training. However, the elderly may encounter various emergencies during the training process, for example, incorrect breathing methods may lead to breathing difficulties, and elderly patients with more underlying diseases may cause various complications. Based on this, an exemplary embodiment of the present application provides a supervision method for respiratory rehabilitation training for the elderly.

[0026] like Figure 1 As shown, a scenario diagram of a supervision system for respiratory rehabilitation training for the elderly provided by the present application. In implementation, the elderly respiratory rehabilitation training system may also be referred to as an elderly respiratory rehabilitation training platform. The elderly respiratory rehabilitation training system includes at least a user terminal 10, a medical terminal 20 and a remote service terminal 30, which are equipped with network connection functions and can communicate with each other, wherein the user terminal 10 includes a portable terminal capable of collecting and uploading, for example, the user terminal 10 may include the elderly patient's mobile terminal and / or a body movement acquisition device capable of collecting body movements.

[0027] As an example, the body motion acquisition device shown may include various wearable sensors, pressure sensors, motion capture cameras, etc. The wearable sensors may include accelerometers and gyroscopes, which can monitor the movement direction, speed and angle changes of the elderly's limbs in real time. Figure 1These sensors can be fixed on key parts such as wrists, ankles, knees and waists, so that the elderly can wear them naturally when performing daily activities or rehabilitation training, without causing too much restriction on their actions. According to an exemplary embodiment of the present application, the method can process the data of the accelerometer to obtain linear acceleration, and then calculate the components of the acceleration of the elderly patient in different directions to determine the balance state of the elderly patient. The data of the gyroscope is then integrated to obtain the rotation angle. Optionally, the method can combine the data of the accelerometer and the gyroscope, and use a fusion algorithm (such as complementary filtering, Kalman filtering, etc.) to improve the accuracy and stability of the measurement.

[0028] In addition, the body motion acquisition device may also include a pressure sensor. The pressure sensor is usually installed at the contact parts of the rehabilitation training equipment, such as seats, armrests and foot pedals. For elderly rehabilitation training, the pressure sensor can sense the force distribution and pressure changes of the elderly when sitting, grasping and standing, and help understand the muscle force and body support stability. For example, when using a walker for standing training, the pressure sensor can feedback whether the force on the elderly's feet is uniform, thereby judging whether their standing posture is correct.

[0029] As an example, the motion capture camera of the exemplary embodiment of the present application can use advanced image recognition technology to capture the body movements of the elderly in all directions. By analyzing and processing the video captured by the camera, the system can accurately identify various complex motion patterns, such as the trajectory and amplitude of movements such as raising hands, turning around, squatting, etc. At the same time, the combination of multiple cameras can realize all-round motion monitoring of the elderly and ensure the integrity of data collection.

[0030] The medical terminal 20 may include a terminal of a medical staff member matched with the elderly patient or a hospital server corresponding to the hospital matched with the elderly patient. The hospital server may allocate a corresponding doctor terminal according to the patient's illness level and / or rehabilitation training status.

[0031] The remote service end 30 may include a processor and a data storage device. The processor may receive patient information and rehabilitation training data uploaded by a user terminal, and the results after data processing are fed back to the user terminal and / or the medical terminal. The data storage device stores data of each user terminal and various types of data for the convenience of processing by the processor.

[0032] After describing the monitoring system of the elderly respiratory rehabilitation training, the following will refer to Figure 2 To describe methods for supervising respiratory rehabilitation training in the elderly. Figure 2 A schematic diagram of a supervision method for respiratory rehabilitation training for the elderly according to an embodiment of the present application is shown.

[0033] Reference Figure 2 , the elderly respiratory rehabilitation training system can obtain baseline data through the body movement acquisition device, and then upload the baseline data to the remote server. After the remote server uses the baseline data to find abnormal conditions, it will issue an early warning to the user terminal, adjust the respiratory rehabilitation training program, and then send the adjusted program to the user terminal. This process will be synchronized to the medical terminal. The medical terminal can optimize these exercise rehabilitation programs by tracking the execution of the rehabilitation training programs of each elderly patient, and provide the optimized exercise rehabilitation training program (also called exercise prescription) to the user terminal. During this period, the remote server and / or medical terminal conducts irregular professional evaluations under continuous tracking to form a closed-loop management process to achieve safety protection and supervision during the elderly respiratory rehabilitation training process.

[0034] Specifically, according to the supervision method of respiratory rehabilitation training for the elderly according to the exemplary embodiment of the present application, the baseline data of the patient may be first determined, and the baseline data may include various motion parameters / indicators of the patient under low intensity or normal state.

[0035] As an example, the method shown may use a body motion acquisition device to collect baseline data. Specifically, the method may use an accelerometer to record the rhythm and stride size of the steps, and combine the data obtained by the gyroscope to calculate the balance state and rotation angle of the elderly patient at each key gait, and may also use a belt sensor worn on the chest to determine the size of the chest contour at each key gait. As an exemplary embodiment, during the patient's motor rehabilitation training, the method may collect reference data at each key gait, and the reference data includes a balance state, a reference rotation angle, and a reference chest contour size. It should be noted that these reference data may be the data obtained from the patient in a low-intensity exercise state or a normal state as described above, or may be data determined by a technician in this field based on the collected training data of various patients.

[0036] In implementation, the method may utilize a motion capture device that captures the patient's movements to capture the video / image of the patient performing motor rehabilitation training, and obtain a key gait image through a key gait recognition algorithm. As an exemplary embodiment of the present application, the key gait refers to a representative moment that can be captured during the walking process of the elderly patient. The key gait images in the embodiments of the present application may include a support phase start image, a support phase mid-stage image, a support phase end-stage image, a swing phase start image, a swing phase mid-stage image, and a swing phase end-stage image.

[0037] The support phase start image refers to the image at the moment the foot first touches the ground. The key gait image at this time records the patient's posture when he begins to bear weight. The mid-support phase image refers to the image when the center of gravity of the body is perpendicular to the supporting foot. The end-support phase image refers to the image when the support phase is about to end and the other foot is ready to step forward. The swing phase start image refers to the image at the moment when the supporting leg ends pushing off the ground and the leg leaves the ground and begins to swing forward. The mid-swing phase image refers to the image of the swing leg in the middle position during the forward swing process. The end-swing phase image refers to the image of the swing leg about to touch the ground and prepare to enter the next support phase.

[0038] According to an embodiment of the present application, the key gait recognition algorithm can be a gait recognition model that has been trained. Specifically, the method can collect a large amount of image data containing various gait stages. These images should cover different individuals and different rehabilitation states to ensure that the model has a wide range of applicability. For example, by installing multiple cameras in a rehabilitation training site, the patient's walking process is photographed from different angles to obtain a rich variety of gait images. Then, those skilled in the art annotate these images to clarify the key gait stage to which each image belongs, such as the start of the support phase, the middle of the support phase, etc.

[0039] The preprocessed image data is divided into training set, validation set and test set. The training set is used for model parameter learning, the validation set is used to adjust the model's hyperparameters, and the test set is used to evaluate the final performance of the model.

[0040] During the training process, the image data in the training set is input into the model batch by batch. The model predicts the image based on the current parameters and calculates the loss value between the predicted result and the true label. Then, the optimizer back-propagates the gradient according to the loss value and updates the parameters of the model so that the loss value gradually decreases. Then, the performance of the model is evaluated using the validation set, and the hyperparameters (such as learning rate, number of network layers, etc.) are adjusted to prevent the model from overfitting. As an example, the model can select a convolutional neural network (CNN) architecture or a model designed specifically for object detection.

[0041] In implementation, the method can input the real-time collected images into the trained model to detect the key gait images. Then, the method uses the time information of the key gait images to determine the first balance state, the first rotation angle and the first chest contour size at the same time. That is, the data collected by the body motion acquisition device can be synchronously transmitted to the remote server together with the time information. The remote server searches the corresponding first balance state, the first rotation angle and the first chest contour size from the acquired data (i.e., the data collected by the body motion acquisition device) according to the time information of the key gait images.

[0042] like Figure 2As shown, the method uses the first balance state, the first rotation angle and the first chest contour size to compare with the reference balance state, the reference rotation angle and the reference chest contour size in the baseline data, respectively, to determine whether the patient has an abnormal condition. Specifically, the first balance state of the patient in the current critical gait can be compared with the reference balance state. The reference balance state is set based on the patient's past normal training performance or the general standard of similar patients. For example, the reference balance state is set so that the patient's body center of gravity deviation range remains stable within ±5 cm during walking. In the first balance state, if the patient's body center of gravity deviates beyond this range, such as reaching ±8 cm, it indicates that the balance state is abnormal, which may indicate that the patient's muscle strength is weakened, the joint stability is decreased, or there is a problem with the nervous system's balance control.

[0043] At the same time, the method can compare the first rotation angle in the current key gait with the reference rotation angle. The reference rotation angle is determined based on the normal rehabilitation training movement specifications, such as the normal rotation angle range of the hip joint in the swing phase is set to 30° to 40°. If the first rotation angle is only 20°, which is significantly lower than the reference range, it may mean that the hip joint movement is limited, which may be caused by muscle contracture, joint inflammation or improper training movements.

[0044] In addition, the method can compare the first chest contour size under the current key gait with the reference chest contour size. The reference chest contour size reflects the patient's breathing amplitude during normal training. For example, the chest contour circumference increases by 5 to 8 cm during normal inhalation. If the first chest contour size only increases by 3 cm during inhalation, it means that the breathing amplitude is abnormal, which may indicate that the patient's respiratory function is affected, such as respiratory muscle weakness, decreased cardiopulmonary function, etc.

[0045] like Figure 2 As shown, when the method confirms that the patient is in an abnormal state, the remote server will push the patient's mobile terminal, such as Figure 1 As shown, the information "It is recommended that you change your exercise training program to **" is pushed on the application interface of the mobile terminal. In addition, the method can also provide the elderly patient with the key gait image and the reference gait image, so that the elderly patient can clearly understand the problems in the training and rehabilitation process. In addition, the method can also suggest corrective actions to the patient based on the difference between the key gait image and the reference gait image, for example, pushing the information "It is recommended that you increase the rotation angle appropriately" on the application interface of the mobile terminal.

[0046] During this process, the remote server can synchronize the training status of the elderly patient to the medical terminal. As an example, the remote server can synchronize the training status of the elderly patient to the professional doctor of the hospital. The professional doctor can adjust the respiratory rehabilitation training program of the elderly patient according to the training status of the elderly patient. During this process, the remote server or the hospital can also conduct professional evaluation of the respiratory rehabilitation training program and the training status of the elderly patient to form a closed-loop management process to achieve safety protection and supervision during the elderly respiratory rehabilitation training.

[0047] As another exemplary embodiment, if any one of the first balance state, the first rotation angle and the first chest contour size is abnormal through comparison, it is determined that the elderly patient has an abnormality. That is to say, through the comparison of the above three dimensions, if any dimension shows a significant difference, it is determined that the patient has an abnormality; if all dimensions meet the reference standard, it is determined that the patient's current training state is normal.

[0048] After the patient's training status is determined to be normal, key gait images, chest contour size, balance state and rotation angle of the patient during the training process, as well as information marked as "normal", are collected. Key gait images may include images of different stages such as the start of the support phase and the middle of the swing phase, which are used to capture the patient's key posture when walking; chest contour size reflects the breathing state, such as the amplitude and frequency of chest contour changes; balance state and rotation angle data record body posture and movement amplitude.

[0049] The collected data are used as training data to input into the abnormal monitoring model. The abnormal monitoring model is usually built based on machine learning algorithms, such as decision trees, support vector machines (SVMs) or neural networks. These data are used as positive samples (normal samples) to help the model learn the characteristic patterns of multi-dimensional data under normal rehabilitation training. For example, in the neural network model, the pixel features of key gait images, the numerical features of chest contour size, the quantitative features of balance state and rotation angle, etc., through the layer-by-layer operation of the network, enable the model to gradually grasp the internal connection and distribution law of these data under normal conditions.

[0050] The collected normal data are input into the abnormal monitoring model as training data. The abnormal monitoring model is usually built based on machine learning or deep learning algorithms. By continuously learning the characteristics and patterns of these normal data, the model can better understand the patient's performance under normal rehabilitation training. And the trained abnormal monitoring model is a model trained using the data of the elderly patients themselves, that is, a model tailored / trained for each elderly patient. Therefore, it can more accurately identify data that is different from the normal pattern, so as to detect abnormal situations in time. For example, after a large amount of normal data training, the model can more keenly capture the subtle changes in the balance state, rotation angle or chest contour size of the elderly patient, and improve the early warning ability of abnormal situations. By continuously collecting and using the patient's normal training data to train the abnormal monitoring model, the model performance can be continuously optimized, providing more reliable abnormal monitoring guarantees for the patient's rehabilitation training, and ensuring that the rehabilitation training is carried out safely and effectively.

[0051] Figure 3 is a schematic flow chart showing a method for supervising respiratory rehabilitation training for the elderly according to an embodiment of the present application. The method comprises step S210, step S220 and step S230.

[0052] In step S210, a training image of an elderly patient during motor rehabilitation training is processed using a key gait recognition algorithm to obtain a first key gait image having a first key gait.

[0053] In step S220, based on the time information corresponding to the first key gait image, the first balance state, the first rotation angle and the first chest contour size of the elderly patient in the first key gait are obtained, and the first balance state, the first rotation angle and the first chest contour size are compared with the reference balance state, the reference rotation angle and the reference chest contour size, respectively, to determine whether the elderly patient has an abnormal condition.

[0054] In step S230, if there is no abnormality in the elderly patient, the first key gait image, the first balance state, the first rotation angle and the first chest contour size of the elderly patient, as well as the information marked as "normal" are used as training data to train the abnormal monitoring model to obtain a trained abnormal monitoring model.

[0055] Optionally, the method further includes: acquiring a second key gait image having a second key gait; inputting the second key gait image into a trained abnormality monitoring model to determine whether the elderly patient has an abnormal condition.

[0056] That is to say, according to the exemplary embodiment of the present application, after the abnormal monitoring model is trained and updated using the training data of the elderly patient itself, an abnormal monitoring model specific to the elderly patient is obtained. The processing result of such an abnormal monitoring model will be more accurate than the method of comparing with reference data. Therefore, in the subsequent processing, the method can directly input the acquired second key gait image into the abnormal monitoring model to determine whether the patient has an abnormal condition.

[0057] Optionally, the first key gait image includes a stance phase start image, a stance phase mid-image, a stance phase end image, a swing phase start image, a swing phase mid-image or a swing phase end image.

[0058] Optionally, the key gait recognition algorithm includes a gait recognition model that has been trained using labeled gait images of various stages.

[0059] Optionally, the first equilibrium state and the first rotation angle are data acquired by using a gyroscope and an accelerometer worn on the elderly patient, and the first chest contour size is data acquired by a belt sensor worn on the elderly patient.

[0060] Optionally, the method further includes: when the elderly patient has an abnormality, synchronizing the training status of the elderly patient to a medical terminal corresponding to the elderly patient and receiving an updated exercise rehabilitation training program sent by the medical terminal.

[0061] Optionally, the method further includes: if the elderly patient has an abnormal condition, sending a reminder message and a first key gait image of the elderly patient to an elderly patient terminal of the elderly patient.

[0062] Optionally, the method further includes: continuously tracking the exercise performance of the elderly patient during the exercise rehabilitation training, and evaluating and optimizing the exercise rehabilitation training program of the elderly patient based on whether the elderly patient has any abnormal conditions.

[0063] According to the supervision method of elderly respiratory rehabilitation training of the exemplary embodiment of the present application, it is possible to determine whether the elderly patient has abnormalities by comparing the multidimensional data on the key gait, and in this way, the patient's training status can be comprehensively and accurately evaluated. Significant differences in any dimension can trigger abnormal judgment in time to ensure that no possible abnormal situation is missed, and after obtaining the positive sample, the abnormal training model dedicated to the patient is trained by the positive sample, and as the positive samples of the patient increase, the accuracy of the abnormal training model increases, which is convenient for the subsequent use of the abnormal training model dedicated to the patient to directly supervise the patient's training situation. In the case of abnormalities in the elderly patient, the training situation can be quickly synchronized to the medical terminal, and the updated rehabilitation training program can be received to ensure that the patient can get professional guidance and adjustment in time. At the same time, the system can stimulate the patient's training enthusiasm and improve the patient's self-management ability by sending reminder information and displaying key gait images to the elderly patient terminal. In addition, the method can continuously track the patient's sports performance, and evaluate and optimize the training program according to whether there is an abnormality, to ensure that the training program always meets the patient's actual needs. This dynamic adjustment method helps patients gradually improve the training effect and promote the rehabilitation process.

[0064] Figure 4 A block diagram of a monitoring device for elderly respiratory rehabilitation training according to an exemplary embodiment of the present application is shown. Figure 4 At the hardware level, the device includes a processor, an internal bus and a computer-readable storage medium, wherein the computer-readable storage medium includes a volatile memory and a non-volatile memory. The processor reads the corresponding computer program from the non-volatile memory and then runs it. Of course, in addition to the software implementation, this application does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0065] Specifically, the processor performs the following operations: processing the training images of the elderly patient during the motor rehabilitation training through the key gait recognition algorithm to obtain a first key gait image with a first key gait; based on the time information corresponding to the first key gait image, obtaining the first balance state, first rotation angle and first chest contour size of the elderly patient in the first key gait, and using the first balance state, first rotation angle and first chest contour size to compare with the reference balance state, reference rotation angle and reference chest contour size respectively to determine whether the elderly patient has an abnormality; if the elderly patient does not have an abnormality, using the first key gait image, first balance state, first rotation angle and first chest contour size of the elderly patient, and the information marked as "normal" as training data to train the abnormal monitoring model, and obtain the trained abnormal monitoring model.

[0066] Optionally, the processor may also perform the following operations: acquiring a second key gait image having a second key gait; inputting the second key gait image into a trained abnormal monitoring model to determine whether the elderly patient has an abnormal condition.

[0067] Optionally, the first key gait image includes a stance phase start image, a stance phase mid-image, a stance phase end image, a swing phase start image, a swing phase mid-image or a swing phase end image.

[0068] Optionally, the key gait recognition algorithm includes a gait recognition model that has been trained using labeled gait images of various stages.

[0069] Optionally, the first equilibrium state and the first rotation angle are data acquired by using a gyroscope and an accelerometer worn on the elderly patient, and the first chest contour size is data acquired by a belt sensor worn on the elderly patient.

[0070] Optionally, the processor may further perform the following operations: when the elderly patient has an abnormality, synchronizing the training status of the elderly patient to a medical terminal corresponding to the elderly patient and receiving an updated exercise rehabilitation training program sent by the medical terminal.

[0071] The processor may further perform the following operations: if the elderly patient has an abnormal condition, sending a reminder message and a first key gait image of the elderly patient to the elderly patient terminal of the elderly patient.

[0072] Optionally, the processor may also perform the following operations: continuously tracking the exercise performance of the elderly patient during the exercise rehabilitation training, and evaluating and optimizing the exercise rehabilitation training program of the elderly patient according to whether the elderly patient has any abnormal conditions.

[0073] In summary, according to the exemplary embodiment of the present application, the monitoring device for elderly respiratory rehabilitation training can determine whether the elderly patient has abnormalities by comparing the multidimensional data on the key gait, and in this way, the patient's training status can be comprehensively and accurately evaluated. Significant differences in any dimension can trigger abnormal judgment in time to ensure that no possible abnormal situation is missed, and after obtaining the positive sample, the abnormal training model dedicated to the patient is trained by the positive sample, and as the positive samples of the patient increase, the accuracy of the abnormal training model increases, which is convenient for the subsequent use of the abnormal training model dedicated to the patient to directly supervise the patient's training situation. In the case of abnormalities in the elderly patient, the training situation can be quickly synchronized to the medical terminal, and the updated rehabilitation training program can be received to ensure that the patient can get professional guidance and adjustment in time. At the same time, a reminder message is sent to the elderly patient terminal to improve the patient's participation and self-management ability. In addition, the method can continuously track the patient's sports performance, and evaluate and optimize the training program according to whether there is an abnormality, to ensure that the training program always meets the patient's actual needs. This dynamic adjustment method helps patients gradually improve the training effect and promote the rehabilitation process.

[0074] It should be noted that the execution subject of each step of the method provided in Example 1 can be the same device, or the method can be executed by different devices. For example, the execution subject of step 21 and step 22 can be device 1, and the execution subject of step 23 can be device 2; for another example, the execution subject of step 21 can be device 1, and the execution subject of step 22 and step 23 can be device 2; and so on.

[0075] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0076] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0077] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0079] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0080] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0081] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0082] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0083] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0084] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for supervising respiratory rehabilitation training for the elderly, characterized in that: include: Using a key gait recognition algorithm to process training images of elderly patients during motor rehabilitation training, to obtain a first key gait image having a first key gait; Based on the time information corresponding to the first key gait image, a first balance state, a first rotation angle, and a first chest contour size of the elderly patient in the first key gait are obtained, and the first balance state, the first rotation angle, and the first chest contour size are compared with a reference balance state, a reference rotation angle, and a reference chest contour size, respectively, to determine whether the elderly patient has an abnormal condition; If the elderly patient does not have any abnormal situation, the first key gait image, the first balance state, the first rotation angle and the first chest contour size of the elderly patient, as well as the information marked as "normal" are used as training data to train the abnormal monitoring model to obtain a trained abnormal monitoring model.

2. The method according to claim 1, characterized in that Also includes: acquiring a second key gait image having a second key gait; The second key gait image is input into the trained abnormal monitoring model to determine whether the elderly patient has an abnormal condition.

3. The method according to claim 2, characterized in that The first key gait image includes a stance phase start image, a stance phase mid-image, a stance phase end image, a swing phase start image, a swing phase mid-image or a swing phase end image.

4. The method according to claim 3, characterized in that The key gait recognition algorithm includes a gait recognition model trained using labeled gait images of various stages.

5. The method according to claim 1, characterized in that The first equilibrium state and the first rotation angle are data acquired by using a gyroscope and an accelerometer worn on the elderly patient, and the first chest contour size is data acquired by a belt sensor worn on the elderly patient.

6. The method according to claim 1, characterized in that Also includes: If the elderly patient has an abnormal condition, the training condition of the elderly patient is synchronized to the medical terminal corresponding to the elderly patient and an updated exercise rehabilitation training program sent by the medical terminal is received.

7. The method according to claim 1, characterized in that Also includes: If the elderly patient has an abnormal condition, a reminder message and a first key gait image of the elderly patient are sent to the elderly patient terminal of the elderly patient.

8. The method according to claim 1, characterized in that Also includes: The exercise performance of the elderly patient during the exercise rehabilitation training is continuously tracked, and the exercise rehabilitation training program of the elderly patient is evaluated and optimized according to whether the elderly patient has any abnormal conditions.

9. A monitoring device for elderly respiratory rehabilitation training, characterized in that: include: processor; as well as A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the method for supervising respiratory rehabilitation training for the elderly as claimed in any one of claims 1 to 8.

10. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instructions are executed, the supervision method for respiratory rehabilitation training for the elderly as described in any one of claims 1 to 8 is implemented.

Images