Medical rehabilitation training system based on deep learning

Abstract

The invention discloses a medical rehabilitation training system based on deep learning, which comprises a data collection module, a training management module and an equipment management module, and is characterized in that the data collection module is used for collecting fracture diagnosis data of a patient, the healing condition of a fracture part of the patient and a visual image of rehabilitation training of the patient; the training management module is used for formulating a personalized rehabilitation training scheme for the patient according to the illness state and the treatment condition of the patient, evaluating whether the rehabilitation training scheme of the patient conforms to the physical state of the patient or not, and performing adjustment according to the physical state of the patient; the equipment management module is used for adjusting the operation mode of the auxiliary equipment according to a rehabilitation training scheme, the data collection module, the training management module and the equipment management module are electrically connected with one another, and the data collection module comprises a patient condition comprehensive data input module, a data acquisition module and a sensor module. The method has the characteristics of improving accuracy and rehabilitation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of rehabilitation training, and particularly to a medical rehabilitation training system based on deep learning. Background Technique

[0002] Rehabilitation training refers to physical activities that are beneficial to the recovery or improvement of functions after an injury. Except for severe injuries that require rest and treatment, general injuries do not necessarily require complete cessation of physical exercises. Appropriate and scientific physical exercises play a positive role in the rapid healing of injuries and the promotion of functional recovery.

[0003] With the continuous development of society, people pay more attention to physical health and are more inclined to follow medical advice for treatment. Currently, the rehabilitation training of orthopedic patients is mainly carried out under the guidance of doctors. Most doctors communicate with patients to understand their physical conditions and formulate rehabilitation training plans in combination with the patients' test reports. However, most doctors mainly rely on their medical experience to formulate rehabilitation training plans, which are easily affected by subjective factors, resulting in poor accuracy of the rehabilitation training plans. At the same time, when communicating with some older patients, there may be certain deviations in medical advice, leading to injuries during rehabilitation training. Moreover, the existing technology requires patients to constantly visit doctors for guidance on rehabilitation training. Due to the shortage of medical resources, patients cannot obtain new rehabilitation training and rehabilitation guidance in a timely manner, resulting in an extended rehabilitation period for patients. Therefore, it is necessary to design a medical rehabilitation training system based on deep learning to improve accuracy and rehabilitation efficiency. Summary of the Invention

[0004] The purpose of the present invention is to provide a medical rehabilitation training system based on deep learning to solve the problems raised in the above background technique.

[0005] To solve the above technical problems, the present invention provides the following technical solution: A medical rehabilitation training system based on deep learning, including a data collection module, a training management module, and an equipment management module, characterized in that: the data collection module is used to collect the fracture diagnosis data and treatment data of patients, collect the healing condition of the fracture site of patients and the visual images of patients' rehabilitation training; the training management module is used to formulate a personalized rehabilitation training plan for patients according to their conditions and treatment situations, evaluate whether the patients' rehabilitation training plans conform to their physical states, and adjust according to the patients' physical states; the equipment management module is used to adjust the operation mode of the auxiliary equipment according to the rehabilitation training plan when adjusting the rehabilitation training plan; the data collection module, the training management module, and the equipment management module are electrically connected to each other;

[0006] The rehabilitation training plan formulation module includes a disease condition analysis sub-module, a model training sub-module, and a plan formulation sub-module. The disease condition analysis sub-module is used to analyze the degree of damage to the muscle tissue at the patient's injury site and the relationship between the damaged muscle tissues. The model training sub-module is used to train a plan analysis model, formulate a rehabilitation training plan according to the plan analysis model and optimize it. The plan formulation sub-module is used to preliminarily formulate a rehabilitation training plan for the patient according to the analysis results;

[0007] The training evaluation module includes a training quality detection sub-module, a training warning sub-module, and a training plan adjustment sub-module. The training quality detection sub-module is used to monitor in real time whether the various indicators of the patient during the rehabilitation training meet the requirements. The training warning sub-module is used to analyze the physical state characteristics of the patient during the rehabilitation training, and issue an alarm to stop the patient's rehabilitation training according to the analysis results. The training plan adjustment sub-module is used to analyze the patient's injury healing situation and physical characteristics, and adjust the rehabilitation training plan according to the analysis results.

[0008] According to the above technical solution, the data collection module includes a patient disease condition comprehensive data entry module, a data acquisition module, and a sensor module. The patient disease condition comprehensive data entry module is used to enter the patient's disease condition comprehensive information into the system. The data acquisition module is used to collect in real time the healing situation of the patient's disease condition. The sensor module is used to collect visual images of the patient during the rehabilitation training.

[0009] According to the above technical solution, the training management module includes a rehabilitation training plan formulation module. The rehabilitation training plan formulation module is used to analyze the patient's disease condition data and treatment data, and formulate a personalized rehabilitation training plan according to the analysis results.

[0010] According to the above technical solution, the training management module further includes a training evaluation module. The training evaluation module is used to analyze whether the training meets the standards when the patient is doing rehabilitation training, and adjust the patient's rehabilitation training plan according to the patient's rehabilitation situation.

[0011] According to the above technical solution, the equipment management module includes an intelligent adjustment module and a data storage module. The intelligent adjustment module is used to adjust the operation mode of the rehabilitation training auxiliary equipment. The data storage module is used to store the patient's historical rehabilitation training data and rehabilitation training plan.

[0012] According to the above technical solution, the operation method of the medical rehabilitation training system includes the following steps:

[0013] Step S1: Enter the patient's fracture diagnosis data and physical characteristic data into the system through the patient's condition comprehensive data entry module, collect the patient's fracture healing status and the duration of each exercise in real time during the rehabilitation training process through the data acquisition module, and collect the patient's physical characteristic parameters and visual images of the patient's training process in real time through the sensor module;

[0014] Step S2: After the patient completes the surgery, the system starts the rehabilitation training program formulation module, starts analyzing the patient's detailed data on the condition, and formulates the first program for the patient based on the analysis results;

[0015] Step S3: When the patient is undergoing rehabilitation training, the system starts the training evaluation module and begins to analyze whether the various indicators of the patient's rehabilitation training meet the standards. According to the analysis results, an alarm is issued to remind the patient to stop the rehabilitation training, and the training plan is further adjusted according to the patient's healing condition.

[0016] Step S4: When the patient is undergoing rehabilitation training, the system adjusts the working mode of the auxiliary training equipment in real time according to the training plan, and records the patient's rehabilitation training data.

[0017] According to the above technical solution, step S2 further includes the following steps:

[0018] Step S21: Obtain historical patient fracture diagnosis data and corresponding doctor-guided rehabilitation training records in the hospital database, pre-process the data, remove duplicate, incomplete and erroneous data in the data, use the pre-processed data as sample data, use the sample data to train the solution analysis model, call the test sample and the corresponding doctor-guided solution, use the solution analysis model to analyze the model recommendation solution corresponding to the test sample, and compare the doctor-guided solution. If the similarity between the model recommendation solution and the doctor-guided solution is greater than the system-set threshold, select the model recommendation solution as the first solution, otherwise optimize the solution analysis model;

[0019] Step S22: Obtain the patient's fracture diagnosis data, identify the patient's fracture location, retrieve the corresponding human body structure model in the database, anchor the muscle groups around the patient's fracture location according to the patient's fracture location, identify the number of damaged muscle lines in the muscle group, and retrieve the corresponding muscle contraction training cycle and the corresponding number of times in the database as the second plan according to the location of the damaged muscle group and the number of damaged muscle lines, obtain the first plan, calculate the difference between the muscle contraction training cycle and the corresponding number of times in the first plan and the second plan, if the difference is less than the minimum threshold, then use the first plan as the target plan, if the difference is greater than the minimum threshold and less than the maximum threshold, then perform weighted fusion on the first and second plans to obtain the third plan as the target plan, otherwise the system continues to detect.

[0020] According to the above technical solution, step S3 further includes the following steps:

[0021] Step S31: retrieve the patient's weight and skeleton features, and retrieve the corresponding influence coefficients α and β influencing the size of the patient's muscle contraction training model in the database according to the patient's weight and skeleton features;

[0022] Step S32: retrieve the visual image of the patient during rehabilitation training, anchor the visual image of the patient's fracture position, retrieve the three-dimensional scanning data corresponding to the patient's fracture position, identify the point data in the three-dimensional point cloud, construct a three-dimensional model of the patient's fracture position according to the point data in the three-dimensional point cloud, overlap and compare the constructed three-dimensional model with the model in the database, establish a coordinate system, respectively identify the coordinates of each feature node in the two models, calculate the distance L1 and L2 between the corresponding nodes of the constructed three-dimensional model and the model in the database and the origin respectively through the distance formula, correct the distance from the constructed three-dimensional model to the origin, calculate the corrected distance L'1=α·β·L1 through the formula, where L'1 represents the distance from the corrected constructed three-dimensional model to the origin, calculate the difference between L'1 and L2, and if the difference is greater than the threshold, send a reminder to the patient through the bracelet to guide the patient to correct the wrong training method, otherwise send a reminder to the patient through the bracelet to reduce the force used;

[0023] Step S33: According to the set cycle, periodically capture images, scan and identify the edge feature nodes of the patient's body where the fracture is located, anchor the positions of the edge feature nodes and the rotational joints in the image, and mark the pixel points where the edge feature nodes and the rotational joints are located, overlap and fuse all images within the cycle, identify the marks in the pixels, if there are marks in the pixels, keep the pixels, otherwise remove the pixels, connect all edge feature nodes with the rotational joints, identify the size of the rotation angle θ1, and calculate the patient's rotation speed during rotational joint training through the formula In the formula, V represents the rotation speed of the patient during joint rotation training, and t represents the duration of the cycle. If the rotation speed is greater than the threshold, the wristband will remind the patient to reduce the rotation speed, otherwise the system will continue to detect.

[0024] According to the above technical solution, step S33 further includes the following steps:

[0025] Step S331: Identify the number of connections of muscle fibers around the fracture site of the patient, and retrieve the tensile force F of the corresponding muscle fibers in the database according to the number of connections of the muscle fibers. i , according to the joint rotation angle and the joint rotation speed, the corresponding influence coefficients θ and μ on the stretching force of the muscle fibers in the database are retrieved respectively;

[0026] Step S332: Retrieve the bone healing data at the fracture site of the patient, identify the degree of bone healing of the patient, retrieve the corresponding influence coefficient η of the bone density on the tensile force from the database according to the degree of bone healing of the patient. If the healing degree is greater than the maximum threshold, retrieve the bone density of the patient, and retrieve the corresponding tensile force G of the muscle fibers that can be tolerated from the database according to the bone density. Calculate the influence coefficient of the joint rotation angle through the formula In the formula, i = 1, 2, 3......n. Retrieve the corresponding rotation angle from the database according to the calculated influence coefficient θ. Otherwise, the system continues to detect.

[0027] According to the above technical solution, during the rehabilitation training of the patient in step S4, the system records the training duration of the patient in real time. If the training duration of the patient is greater than the threshold, the patient is reminded to rest through the bracelet. Otherwise, the system continues to detect. When the patient starts the rehabilitation training, retrieve the rehabilitation training plan planned by the system for the patient, and adjust the device according to the joint rotation angle and rotation speed in the rehabilitation training plan.

[0028] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: In the present invention, by analyzing the muscle injury characteristics at the fracture site of the patient, the corresponding rehabilitation training plan is obtained, and then the first plan given by the plan analysis model is compared, which can reduce the error of the target plan, make the plan more suitable for the patient, and thus greatly improve the accuracy of the plan. By correcting the distance from the feature nodes on the constructed three-dimensional model to the origin, the influence of the patient's body shape and skeleton size characteristics on the accuracy of the detection results can be avoided, and the accuracy of the system is greatly improved. Compare the three-dimensional model of the patient during rehabilitation training with the model set by the system, and then accurately point out the incorrect training during the patient's rehabilitation training, thereby reducing the impact of the patient's incorrect training on the patient's physical health, and thus improving the safety of the patient. By monitoring the speed of the patient's rotating joints, it is possible to avoid damage to the fracture site during the patient's training, resulting in an extension of the patient's rehabilitation period, and thus greatly improving the patient's rehabilitation efficiency. By detecting the healing degree of the patient's fracture site, the rehabilitation training plan of the patient is adjusted according to the healing degree of the patient's fracture site, making the rehabilitation training plan more personalized, and thus greatly improving the accuracy of the system. Brief Description of the Drawings

[0029] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0030] Figure 1 It is a schematic diagram of the system module composition of the present invention. Detailed Embodiments

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0032] Please refer to Figure 1 , the present invention provides a technical solution: a medical rehabilitation training system based on deep learning, including a data collection module, a training management module, and an equipment management module, characterized in that: the data collection module is used to collect the fracture diagnosis data and treatment data of patients, collect the healing condition of the fracture site of patients and the visual images of patients' rehabilitation training, the training management module is used to formulate personalized rehabilitation training plans for patients according to the patients' conditions and treatment situations, evaluate whether the patients' rehabilitation training plans meet the patients' physical conditions, and adjust according to the patients' physical conditions, the equipment management module is used to adjust the operation mode of the auxiliary equipment according to the rehabilitation training plan when adjusting the rehabilitation training plan, and the data collection module, the training management module, and the equipment management module are electrically connected to each other;

[0033] The rehabilitation training plan formulation module includes a disease analysis sub-module, a model training sub-module, and a plan formulation sub-module. The disease analysis sub-module is used to analyze the degree of damage to the muscle tissue at the injury site of the patient and the relationship between the damaged muscle tissues. The model training sub-module is used to train the plan analysis model, formulate and optimize the rehabilitation training plan according to the plan analysis model. The plan formulation sub-module is used to initially formulate the rehabilitation training plan for the patient according to the analysis results;

[0034] The training evaluation module includes a training quality detection sub-module, a training warning sub-module, and a training plan adjustment sub-module. The training quality detection sub-module is used to monitor in real time whether the indicators of the patient during the rehabilitation training meet the requirements. The training warning sub-module is used to analyze the physical state characteristics of the patient during the rehabilitation training, and issue an alarm to stop the patient's rehabilitation training according to the analysis results. The training plan adjustment sub-module is used to analyze the injury healing condition and physical characteristics of the patient, and adjust the rehabilitation training plan according to the analysis results.

[0035] The data collection module includes a patient condition comprehensive data entry module, a data acquisition module, and a sensor module. The patient condition comprehensive data entry module is used to enter the comprehensive information of the patient's condition into the system. The data acquisition module is used to collect the healing condition of the patient's condition in real time. The sensor module is used to collect the visual images of the patient during the rehabilitation training.

[0036] The training management module includes a rehabilitation training program formulation module, which is used to analyze the patient's condition data and treatment data and formulate a personalized rehabilitation training program based on the analysis results.

[0037] The training management module also includes a training evaluation module, which is used to analyze whether the patient's training meets the standards during rehabilitation training and adjust the patient's rehabilitation training plan according to the patient's rehabilitation situation.

[0038] The equipment management module includes an intelligent adjustment module and a data storage module. The intelligent adjustment module is used to adjust the operation mode of the rehabilitation training auxiliary equipment, and the data storage module is used to store the patient's historical rehabilitation training data and rehabilitation training plan.

[0039] The operation method of the medical rehabilitation training system comprises the following steps:

[0040] Step S1: Enter the patient's fracture diagnosis data and physical characteristic data into the system through the patient's condition comprehensive data entry module, collect the patient's fracture healing status and the duration of each exercise in real time during the rehabilitation training process through the data acquisition module, and collect the patient's physical characteristic parameters and visual images of the patient's training process in real time through the sensor module;

[0041] Step S2: After the patient completes the surgery, the system starts the rehabilitation training program formulation module, starts analyzing the patient's detailed data on the condition, and formulates the first program for the patient based on the analysis results;

[0042] Step S3: When the patient is undergoing rehabilitation training, the system starts the training evaluation module and begins to analyze whether the various indicators of the patient's rehabilitation training meet the standards. According to the analysis results, an alarm is issued to remind the patient to stop the rehabilitation training, and the training plan is further adjusted according to the patient's healing condition.

[0043] Step S4: When the patient is undergoing rehabilitation training, the system adjusts the working mode of the auxiliary training equipment in real time according to the training plan, and records the patient's rehabilitation training data.

[0044] Step S2 further comprises the following steps:

[0045] Step S21: Obtain historical patient fracture diagnosis data and corresponding doctor-guided rehabilitation training records in the hospital database, pre-process the data, remove duplicate, incomplete and erroneous data in the data, use the pre-processed data as sample data, use the sample data to train the solution analysis model, call the test sample and the corresponding doctor-guided solution, use the solution analysis model to analyze the model recommendation solution corresponding to the test sample, and compare the doctor-guided solution. If the similarity between the model recommendation solution and the doctor-guided solution is greater than the system-set threshold, select the model recommendation solution as the first solution, otherwise optimize the solution analysis model;

[0046] Step S22: Obtain the fracture diagnosis data of the patient, identify the fracture location of the patient, retrieve the corresponding human body structure model in the database, anchor the muscle groups around the patient's fracture location according to the patient's fracture location, identify the number of damaged muscle lines in the muscle groups, retrieve the corresponding muscle contraction training cycle and the corresponding number of times in the database as the second plan according to the location of the damaged muscle group and the number of damaged muscle lines, obtain the first plan, calculate the difference between the muscle contraction training cycle and the corresponding number of times in the first plan and the second plan. If the difference is less than the minimum threshold, then use the first plan as the target plan. If the difference is greater than the minimum threshold and less than the maximum threshold, then perform weighted fusion on the first and second plans to obtain the third plan as the target plan. Otherwise, the system continues to detect. By analyzing the muscle damage characteristics at the patient's fracture site, obtain the corresponding rehabilitation training plan, and then compare it with the first plan given by the plan analysis model, which can reduce the error of the target plan, make the plan more suitable for the patient, and thus greatly improve the accuracy of the plan.

[0047] Step S3 further includes the following steps:

[0048] Step S31: Retrieve the weight and skeleton characteristics of the patient, and retrieve the influence coefficients α and β corresponding to the size of the patient's muscle stretching and contraction training model in the database according to the weight and skeleton characteristics of the patient;

[0049] Step S32: retrieve the visual image of the patient during rehabilitation training, anchor the visual image of the patient's fracture position, retrieve the three-dimensional scanning data corresponding to the patient's fracture position, identify the point data in the three-dimensional point cloud, construct a three-dimensional model of the patient's fracture position according to the point data in the three-dimensional point cloud, overlap and compare the constructed three-dimensional model with the model in the database, establish a coordinate system, identify the coordinates of each feature node in the two models, calculate the distance L1 and L2 between the corresponding nodes of the constructed three-dimensional model and the model in the database and the origin respectively through the distance formula, correct the distance from the constructed three-dimensional model to the origin, and calculate the corrected distance L'1=α·β·L1 through the formula, where L'1 represents the distance from the corrected constructed three-dimensional model to the origin, and calculate The difference between L'1 and L2, if the difference is greater than the threshold, it means that the patient's muscle stretching training is not in place, and a reminder is sent to the patient through the bracelet to guide the patient to correct the wrong training method. Otherwise, it means that the patient's muscle stretching training is too strong, and a reminder is sent to the patient through the bracelet to reduce the force used. By correcting the distance from the feature node to the origin on the constructed three-dimensional model, the influence of the patient's body shape and skeleton size characteristics on the accuracy of the detection results can be avoided, greatly improving the accuracy of the system. The three-dimensional model of the patient during rehabilitation training is compared with the model set by the system, and then the wrong training during the patient's rehabilitation training process is accurately pointed out, thereby reducing the impact of the patient's wrong training on the patient's physical health, thereby improving the patient's safety;

[0050] Step S33: According to the set cycle, periodically capture images, scan and identify the edge feature nodes of the patient's body where the fracture is located, anchor the positions of the edge feature nodes and the rotational joints in the image, and mark the pixel points where the edge feature nodes and the rotational joints are located, overlap and fuse all images within the cycle, identify the marks in the pixels, if there are marks in the pixels, keep the pixels, otherwise remove the pixels, connect all edge feature nodes with the rotational joints, identify the size of the rotation angle θ1, and calculate the patient's rotation speed during rotational joint training through the formula In the formula, V represents the rotation speed of the patient during joint rotation training, and t represents the duration of the cycle. If the rotation speed is greater than the threshold, the bracelet will remind the patient to reduce the rotation speed. Otherwise, the system will continue to detect. By monitoring the speed of the patient's joint rotation, it can avoid the patient from causing damage to the fracture during training, resulting in a longer rehabilitation period for the patient, thereby greatly improving the patient's rehabilitation efficiency.

[0051] Step S33 further includes the following steps:

[0052] Step S331: Identify the number of connections of muscle fibers around the fracture site of the patient, and retrieve the tensile force F of the corresponding muscle fibers in the database according to the number of connections of the muscle fibers. i, according to the joint rotation angle and joint rotation speed, respectively retrieve the corresponding influence coefficients θ and μ of the stretching force on muscle fibers in the database;

[0053] Step S332: Retrieve the bone healing data of the patient's fracture site, identify the bone healing degree of the patient, and according to the bone healing degree of the patient, retrieve the corresponding influence coefficient η of the stretching force on bone density in the database. If the healing degree is greater than the maximum threshold, retrieve the bone density of the patient, and according to the bone density, retrieve the stretching force G of the muscle fibers that can be tolerated in the database, and calculate the influence coefficient of the joint rotation angle through the formula In the formula, i = 1, 2, 3......n. Retrieve the corresponding rotation angle in the database according to the calculated influence coefficient θ. Otherwise, the system continues to detect. By detecting the healing degree of the patient's fracture site, adjust the patient's rehabilitation training plan according to the healing degree of the patient's fracture site, make the rehabilitation training plan more personalized, and thus greatly improve the accuracy of the system.

[0054] In step S4, during the patient's rehabilitation training process, the system records the patient's training duration in real time. If the patient's training duration is greater than the threshold, the bracelet is used to remind the patient to rest. Otherwise, the system continues to detect. When the patient starts rehabilitation training, retrieve the rehabilitation training plan planned by the system for the patient, and adjust the device according to the joint rotation angle and rotation speed in the rehabilitation training plan.

[0055] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0056] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A medical rehabilitation training system based on deep learning, comprising a data collection module, a training management module and an equipment management module, characterized in that: The data collection module is used to collect the patient's fracture diagnosis data and treatment data, collect the patient's fracture healing status and the patient's visual images of rehabilitation training; the training management module is used to formulate a personalized rehabilitation training plan for the patient according to the patient's condition and treatment status, evaluate whether the patient's rehabilitation training plan is in line with the patient's physical condition, and adjust it according to the patient's physical condition; the equipment management module is used to adjust the operation mode of the auxiliary equipment according to the rehabilitation training plan when adjusting the rehabilitation training plan; the data collection module, the training management module and the equipment management module are electrically connected to each other; The training management module includes a rehabilitation training program formulation module, which is used to analyze the patient's condition data and treatment data and formulate a personalized rehabilitation training program based on the analysis results; The training management module also includes a training evaluation module, which is used to analyze whether the patient's training meets the standards during rehabilitation training and adjust the patient's rehabilitation training plan according to the patient's rehabilitation status; The data collection module includes a patient condition comprehensive data entry module, a data acquisition module and a sensor module. The patient condition comprehensive data entry module is used to enter the patient's condition comprehensive information into the system. The data acquisition module is used to collect the patient's condition recovery status in real time. The sensor module is used to collect visual images of the patient during rehabilitation training. The operation method of the medical rehabilitation training system comprises the following steps: Step S1: Enter the patient's fracture diagnosis data and physical characteristic data into the system through the patient's condition comprehensive data entry module, collect the patient's fracture healing status and the duration of each exercise in real time during the rehabilitation training process through the data acquisition module, and collect the patient's physical characteristic parameters and visual images of the patient's training process in real time through the sensor module; Step S2: After the patient completes the surgery, the system starts the rehabilitation training program formulation module, starts analyzing the patient's detailed data on the condition, and formulates the first program for the patient based on the analysis results; Step S3: When the patient is undergoing rehabilitation training, the system starts the training evaluation module and begins to analyze whether the various indicators of the patient's rehabilitation training meet the standards. According to the analysis results, an alarm is issued to remind the patient to stop the rehabilitation training, and the training plan is further adjusted according to the patient's healing condition. Step S4: When the patient is undergoing rehabilitation training, the system adjusts the working mode of the auxiliary training device in real time according to the training plan, and records the patient's rehabilitation training data; The step S2 further comprises the following steps: Step S21: Obtain historical patient fracture diagnosis data and corresponding doctor-guided rehabilitation training records in the hospital database, pre-process the data, remove duplicate, incomplete and erroneous data in the data, use the pre-processed data as sample data, use the sample data to train the solution analysis model, call the test sample and the corresponding doctor-guided solution, use the solution analysis model to analyze the model recommendation solution corresponding to the test sample, and compare the doctor-guided solution. If the similarity between the model recommendation solution and the doctor-guided solution is greater than the system-set threshold, select the model recommendation solution as the first solution, otherwise optimize the solution analysis model; Step S22: Obtain the fracture diagnosis data of the patient, identify the fracture position of the patient, retrieve the corresponding human body structure model in the database, anchor the muscle group around the fracture position of the patient according to the fracture position of the patient, identify the number of damaged muscle lines in the muscle group, retrieve the corresponding muscle contraction training cycle and the corresponding number of times in the database as the second plan according to the position of the damaged muscle group and the number of damaged muscle lines, obtain the first plan, calculate the difference between the muscle contraction training cycle and the corresponding number of times in the first plan and the second plan, if the difference is less than the minimum threshold, use the first plan as the target plan, if the difference is greater than the minimum threshold and less than the maximum threshold, perform weighted fusion on the first and second plans, and obtain the third plan as the target plan, otherwise the system continues to detect; The step S3 further comprises the following steps: Step S31: retrieve the patient's weight and skeleton features, and retrieve the corresponding influence coefficients α and β influencing the size of the patient's muscle contraction training model in the database according to the patient's weight and skeleton features; Step S32: retrieve the visual image of the patient during rehabilitation training, anchor the visual image of the patient's fracture position, retrieve the three-dimensional scanning data corresponding to the patient's fracture position, identify the point data in the three-dimensional point cloud, construct a three-dimensional model of the patient's fracture position according to the point data in the three-dimensional point cloud, overlap and compare the constructed three-dimensional model with the model in the database, establish a coordinate system, respectively identify the coordinates of each feature node in the two models, calculate the distance L1 and L2 between the corresponding nodes of the constructed three-dimensional model and the model in the database and the origin respectively through the distance formula, correct the distance from the constructed three-dimensional model to the origin, calculate the corrected distance L'1=α·β·L1 through the formula, where L'1 represents the distance from the corrected constructed three-dimensional model to the origin, calculate the difference between L'1 and L2, and if the difference is greater than the threshold, send a reminder to the patient through the bracelet to guide the patient to correct the wrong training method, otherwise send a reminder to the patient through the bracelet to reduce the force used; Step S33: According to the set cycle, periodically capture images, scan and identify the edge feature nodes of the patient's body where the fracture is located, anchor the positions of the edge feature nodes and the rotational joints in the image, and mark the pixel points where the edge feature nodes and the rotational joints are located, overlap and fuse all images within the cycle, identify the marks in the pixels, if there are marks in the pixels, keep the pixels, otherwise remove the pixels, connect all edge feature nodes with the rotational joints, identify the size of the rotation angle θ1, and calculate the patient's rotation speed during rotational joint training through the formula In the formula, V represents the rotation speed of the patient during joint rotation training, and t represents the duration of the cycle. If the rotation speed is greater than the threshold, the wristband will remind the patient to reduce the rotation speed, otherwise the system will continue to detect; The step S33 further comprises the following steps: Step S331: Identify the number of connections of muscle fibers around the fracture site of the patient, and retrieve the tensile force F of the corresponding muscle fibers in the database according to the number of connections of the muscle fibers. i , according to the joint rotation angle and the joint rotation speed, the corresponding influence coefficients θ and μ on the stretching force of the muscle fibers in the database are retrieved respectively; Step S332: retrieve the bone healing data of the patient's fracture, identify the patient's bone healing degree, retrieve the corresponding influence coefficient η on the tensile force of the bone density in the database according to the patient's bone healing degree, if the healing degree is greater than the maximum threshold, retrieve the patient's bone density, retrieve the corresponding tensile force G that can be borne by the muscle fiber in the database according to the bone density, and calculate the influence coefficient of the joint rotation angle by the formula Wherein, i=1,2,3...n, and the corresponding rotation angle in the database is retrieved according to the calculated influence coefficient θ, otherwise the system continues to detect.

Images