CN113223662A - Intelligent limb rehabilitation training method and system - Google Patents
Intelligent limb rehabilitation training method and system Download PDFInfo
- Publication number
- CN113223662A CN113223662A CN202110149456.4A CN202110149456A CN113223662A CN 113223662 A CN113223662 A CN 113223662A CN 202110149456 A CN202110149456 A CN 202110149456A CN 113223662 A CN113223662 A CN 113223662A
- Authority
- CN
- China
- Prior art keywords
- rehabilitation training
- target
- training plan
- target patient
- patient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Physical Education & Sports Medicine (AREA)
- Rehabilitation Tools (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a limb intelligent rehabilitation training method and a system, wherein the method comprises the following steps: acquiring age characteristics and body type characteristics of a target patient and an initial training plan made for the target patient by a doctor; acquiring limb recovery information of the target patient; determining a target rehabilitation training plan and a target action detection threshold value which are required to be executed by the target patient for the ith time according to the age characteristics, the body type characteristics, the initial training plan and the limb recovery information of the target patient; and outputting the target rehabilitation training plan and the target action detection threshold. Can revise initial rehabilitation training plan according to information such as patient's age, size, recovery condition every day, automatic generation is on the same day rehabilitation training plan and action detection threshold value, has improved the accuracy when formulating the rehabilitation training plan and detecting the action amplitude, and this rehabilitation training plan's formulation does not receive the restriction in time and place, has saved a large amount of manpowers, provides convenience for the patient.
Description
Technical Field
The invention relates to the technical field of medical instruments, in particular to a limb intelligent rehabilitation training method and system.
Background
At present, with the continuous improvement of the national living standard, people pay more and more attention to the rehabilitation training after limb injury, and particularly, at present, people with aging aggravation of the population frequently suffer from limb injury, and more people hope to recover the body function through professional rehabilitation training. In the prior art, a professional (i.e. a doctor) makes a rehabilitation training plan for a patient every day according to the body type, age and other characteristics of the patient at different periods after operation. However, in the current situation that the supply of doctors is not sufficient, it is difficult to guarantee accurate acquisition of data of each patient to make scientific and systematic analysis of the rehabilitation condition of the patient. In addition, the technical scheme is restricted by time, place and the like in the implementation process, so that the cost is high, and much inconvenience is brought to patients.
Disclosure of Invention
In order to overcome the problems in the related art, the invention discloses and provides a display control method and device of a special-shaped LED display screen.
According to a first aspect of the disclosed embodiments of the present invention, there is provided a limb intelligent rehabilitation training method, the method including:
determining a target patient needing to make an ith rehabilitation training plan, wherein the rehabilitation training plan is used for describing action names and action times needing to be completed by the target patient;
acquiring age characteristics and body shape characteristics of the target patient and an initial training plan made by a doctor for the target patient, wherein the body shape characteristics at least comprise height characteristics and weight characteristics;
acquiring limb recovery information of the target patient, wherein the limb recovery information comprises an action name and action times completed in each rehabilitation training plan when the target patient executes the (i-1) rehabilitation training plans previously;
taking the age characteristics, body type characteristics, initial training plan and limb recovery information of a target patient as the input of a pre-trained rehabilitation training plan generation model to obtain a target rehabilitation training plan which is output by the rehabilitation training plan generation model and needs to be executed by the target patient for the ith time;
taking the age characteristics, body shape characteristics, initial training plan and limb recovery information of a target patient as the input of a pre-trained detection threshold generation model to obtain a target action detection threshold of the target patient output by the detection threshold generation model, wherein the target action detection threshold is used for detecting whether the action amplitude of the target patient reaches a preset amplitude standard when the target patient executes the target rehabilitation training plan;
and outputting the target rehabilitation training plan and the target action detection threshold.
Optionally, the acquiring the limb recovery information of the target patient includes:
acquiring an acceleration component of the target patient through a three-axis acceleration sensor when the target patient is performing (i-1) times of rehabilitation training before performing;
acquiring an angular acceleration component and an angular velocity component of the target patient through a three-axis gyroscope sensor;
acquiring the atmospheric pressure of the position of the target patient through a first air pressure sensor;
acquiring an air bag air pressure value of an air bag attached to a limb of the target patient for the rehabilitation training plan through a second air pressure sensor;
and determining the action name and the action times of the target patient in each rehabilitation training plan before (i-1) times of rehabilitation training plans are executed according to the preset relation among the acceleration component, the angular velocity component, the atmospheric pressure, the air pressure value of the air bag and the action name.
Optionally, after the outputting the target rehabilitation training plan and the target action detection threshold, the method further includes:
when the target patient executes the ith rehabilitation training plan, detecting whether the action amplitude of the target action executed by the target patient each time reaches a preset amplitude standard or not according to the target action detection threshold;
and if the action amplitude reaches the preset amplitude standard, adding 1 time to the numerical value of the action times of the target action completed in the ith rehabilitation plan executed by the patient.
Optionally, the method further includes training the rehabilitation training plan generation model by using a deep learning model, where the training method is:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample population as training data;
using the training data as an input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the rehabilitation training plan generation model according to the model parameters.
Optionally, the method further includes training the detection threshold generation model by using a deep learning model, where the training method is:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample population as training data;
using the training data as an input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the detection threshold value generation model according to the model parameters.
Optionally, the method further includes:
acquiring a target action detection threshold value when the target patient executes the ith rehabilitation training plan;
determining the threshold variation amplitude of the target action detection threshold compared with all action detection thresholds corresponding to each rehabilitation training plan in the previous (i-1) rehabilitation training plans;
acquiring the exercise duration consumed by the target patient to execute the ith rehabilitation training plan;
and (3) determining the quality grade of the exercise plan corresponding to the previous (i-1) rehabilitation training plan according to a preset exercise plan quality monitoring strategy and by means of the limb recovery information, the threshold variation amplitude and the exercise duration of the target patient.
According to a second aspect of the disclosed embodiments of the present invention, there is provided a limb intelligent rehabilitation training system, the system comprising: the system comprises a target patient determination module, an information determination module, a motion data acquisition device, a rehabilitation training plan generation module, a detection threshold generation module and an output module, wherein the target patient determination module is connected with the information determination module, the information determination module is connected with the motion data acquisition device, the motion data acquisition device is connected with the rehabilitation training plan generation module, the rehabilitation training plan generation module is connected with the detection threshold generation module, and the detection threshold generation module is connected with the output module;
the target patient determination module is used for determining a target patient needing to make an ith rehabilitation training plan, and the rehabilitation training plan is used for describing action names and action times needing to be completed by the target patient;
the information determining module is used for acquiring the age characteristics and the body type characteristics of the target patient and an initial training plan made by a doctor for the target patient, wherein the body type characteristics at least comprise height characteristics and weight characteristics;
the motion data acquisition equipment is used for acquiring limb recovery information of the target patient, and the limb recovery information comprises action names and action times finished in each rehabilitation training plan when the target patient executes the (i-1) rehabilitation training plans previously;
the rehabilitation training plan generating module is used for taking the age characteristics, the body type characteristics, the initial training plan and the limb recovery information of the target patient as the input of a pre-trained rehabilitation training plan generating model so as to obtain a target rehabilitation training plan which is output by the rehabilitation training plan generating model and needs to be executed by the target patient for the ith time;
the detection threshold generation module is used for taking the age characteristics, the body type characteristics, the initial training plan and the limb recovery information of a target patient as the input of a pre-trained detection threshold generation model so as to obtain a target action detection threshold of the target patient output by the detection threshold generation model, wherein the target action detection threshold is used for detecting whether the action amplitude of the target patient reaches a preset amplitude standard or not when the target patient executes the target rehabilitation training plan;
and the output module is used for outputting the target rehabilitation training plan and the target action detection threshold.
Optionally, the motion data acquiring device is configured to:
acquiring an acceleration component of the target patient through a three-axis acceleration sensor when the target patient is performing (i-1) times of rehabilitation training before performing;
acquiring an angular acceleration component and an angular velocity component of the target patient through a three-axis gyroscope sensor;
acquiring the atmospheric pressure of the position of the target patient through a first air pressure sensor;
acquiring an air bag air pressure value of an air bag attached to a limb of the target patient for the rehabilitation training plan through a second air pressure sensor;
and determining the action name and the action times of the target patient in each rehabilitation training plan before (i-1) times of rehabilitation training plans are executed according to the preset relation among the acceleration component, the angular velocity component, the atmospheric pressure, the air pressure value of the air bag and the action name.
Optionally, the system further includes: the motion detection device is connected with the output module, and the frequency counting module is connected with the motion detection device;
the motion detection device is used for detecting whether the action amplitude of the target action executed by the target patient each time reaches a preset amplitude standard or not according to the target action detection threshold when the target patient executes the ith rehabilitation training plan;
and the frequency counting module is used for adding 1 time to the numerical value of the frequency of the target action finished in the ith rehabilitation plan executed by the patient if the action amplitude reaches a preset amplitude standard.
Optionally, the rehabilitation training plan generating module is configured to:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample population as training data;
using the training data as an input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the rehabilitation training plan generation model according to the model parameters.
Optionally, the detection threshold generating module is configured to:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample population as training data;
using the training data as an input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the detection threshold value generation model according to the model parameters.
Optionally, the system further includes: the device comprises a detection threshold generating module, a change amplitude determining module, a movement duration determining module and a quality grade determining module, wherein the detection threshold generating module is connected with the change amplitude determining module, the change amplitude determining module is connected with the movement duration determining module, and the movement duration determining module is connected with the quality grade determining module;
a detection threshold acquisition module, configured to acquire a target action detection threshold when the target patient executes the ith rehabilitation training plan;
a variation amplitude determination module, configured to determine a variation amplitude of the target motion detection threshold compared to threshold variation amplitudes of all motion detection thresholds corresponding to each of the previous (i-1) rehabilitation training plans;
a movement duration determination module, configured to obtain a movement duration consumed by the target patient to execute the ith rehabilitation training plan;
and the quality grade determining module is used for determining the quality grade of the motion plan corresponding to the previous (i-1) times of rehabilitation training plans according to a preset motion plan quality monitoring strategy and through the limb recovery information, the threshold value change amplitude and the motion duration of the target patient.
In summary, the present invention provides a limb intelligent rehabilitation training method and system, the method includes: acquiring age characteristics and body type characteristics of a target patient and an initial training plan made for the target patient by a doctor; acquiring limb recovery information of the target patient; taking the age characteristics, body type characteristics, initial training plan and limb recovery information of the target patient as the input of a pre-trained rehabilitation training plan generation model to obtain the target rehabilitation training plan which is output by the rehabilitation training plan generation model and needs to be executed by the target patient for the ith time; taking the age characteristics, body type characteristics, initial training plan and limb recovery information of the target patient as the input of a pre-trained detection threshold generation model to obtain a target action detection threshold of the target patient output by the detection threshold generation model; and outputting the target rehabilitation training plan and the target action detection threshold. Can revise initial rehabilitation training plan according to information such as patient's age, size, recovery condition every day, automatic generation is on the same day rehabilitation training plan and action detection threshold value, has improved the accuracy when formulating the rehabilitation training plan and detecting the action amplitude, and this rehabilitation training plan's formulation does not receive the restriction in time and place, has saved a large amount of manpowers, provides convenience for the patient.
Additional features and advantages of the present disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart illustrating a method of limb intelligent rehabilitation training, according to an exemplary embodiment;
FIG. 2 is a flow chart of a method of patient information acquisition according to the method shown in FIG. 1;
FIG. 3 is a flow chart of another limb intelligent rehabilitation training method according to FIG. 1;
FIG. 4 is a flow chart of a method of determining a quality level of an exercise plan according to the method shown in FIG. 1;
FIG. 5 is a block diagram illustrating the architecture of a limb intelligent rehabilitation training system according to an exemplary embodiment;
fig. 6 is a block diagram of another limb intelligent rehabilitation training system shown in fig. 5.
Detailed Description
The following detailed description of the disclosed embodiments will be made in conjunction with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
Fig. 1 is a flowchart illustrating a limb intelligent rehabilitation training method according to an exemplary embodiment, as shown in fig. 1, the method including:
before introducing the limb intelligent rehabilitation training method provided by the present disclosure, a target application scenario related to each embodiment in the present disclosure is first introduced, where the target application scenario includes an electronic device capable of detecting motion data of a patient and outputting information to the patient, and the electronic device may be a wearable hardware device (a pressure sensor, an acceleration sensor, and a gyroscope sensor that detect the motion data of the patient are disposed in the hardware device).
In step 101, a target patient for which an ith rehabilitation training plan needs to be made is determined.
Wherein, the rehabilitation training plan is used for describing the action name and the action times of the target patient.
For example, in a general case, after a patient has limb damage, a doctor may make a rehabilitation training plan for the patient according to the degree of limb damage and personal physical signs of the patient, and the patient executes a set of rehabilitation training plans every day or every preset time period according to the order, so as to achieve the purposes of promoting limb recovery and maintaining physical functions. It will be appreciated that as the patient continues to execute the rehabilitation program, the patient's physical function is continually recovering, and therefore the rehabilitation program needs to be updated periodically to accommodate the patient's current and up-to-date limb recovery.
In the embodiment disclosed in the present invention, the case where the patient needs to perform the rehabilitation training plan the ith time is taken as an example, i may be a numerical value of any integer from 1 to n, and represents the number of times the patient performs the rehabilitation training (or, if the patient performs the rehabilitation training plan once a day at a fixed frequency, i may also be used to represent the number of days). After determining the target patient who needs to perform the ith rehabilitation training plan, the ith rehabilitation training plan is prepared for the target patient according to the personal signs (e.g., age characteristics, body type characteristics, muscle tightness, etc.) and the limb recovery degree of the target patient through the following steps 102 and 106.
In step 102, the age characteristics, body type characteristics and the initial training plan of the target patient are obtained.
Wherein the body shape characteristics at least comprise height characteristics and weight characteristics.
For example, the age, height, weight, and the like of the target patient all affect the degree of completion of each target action when the target patient executes the ith rehabilitation training plan, and if the training intensity of the target rehabilitation training plan is high, the target patient may also have damage to the limbs of the patient, and if the training intensity of the target rehabilitation training plan is low, the target patient does not have quick recovery of the limbs of the patient. Therefore, a target rehabilitation training plan suitable for the personal signs of the target patient needs to be established for the target patient according to the age characteristic, the height characteristic, the weight characteristic and the initial training plan established by the doctor for the target patient.
In step 103, limb recovery information of the target patient is acquired.
Wherein the limb recovery information comprises the name of the action and the number of actions completed in each rehabilitation training plan when the target patient previously performed (i-1) rehabilitation training plans.
For example, it can be understood that each target patient needs to refer to the completion degree of each (i-1) rehabilitation training plan (i.e. the limb rehabilitation information of the target patient) before executing the current (i-th) rehabilitation training plan when making each rehabilitation training plan in addition to the personal signs such as the age characteristics and the body shape characteristics of the target patient. If the completion degree of the previous i-1 rehabilitation training plans is not expected, the limb recovery situation of the target patient is not ideal, so that the training intensity needs to be properly reduced when the ith rehabilitation training plan is made; if the completion degree of the previous i-1 times of rehabilitation training plans is ideal, the limb recovery condition of the target patient is proved to be good, so that the training intensity is properly enhanced when the i-th time of rehabilitation training plan is made.
In step 104, the age characteristics, body type characteristics, initial training plan and limb recovery information of the target patient are used as the input of a pre-trained rehabilitation training plan generation model to obtain the target rehabilitation training plan which is output by the rehabilitation training plan generation model and needs to be executed by the target patient for the ith time.
For example, after the relevant information such as the age characteristic, the body type characteristic, the limb recovery information, and the like of the target patient is acquired through the above step 101-103, in order to accurately make the target rehabilitation training plan for the target patient according to the relevant information, in the embodiment of the present disclosure, a technical scheme of deep learning and deep training model training is adopted, a rehabilitation training plan generation model is trained in advance, the age characteristic, the body type characteristic, the initial training plan, and the limb recovery information of the target patient are used as the input of the rehabilitation training plan generation model, and the output result of the rehabilitation training plan generation model is acquired, and is the target rehabilitation training plan that the target patient needs to execute for the ith time.
In step 105, the age characteristics, body type characteristics, initial training plan and limb recovery information of the target patient are used as the input of the pre-trained detection threshold generation model to obtain the target motion detection threshold of the target patient output by the detection threshold generation model.
The target action detection threshold is used for detecting whether the action amplitude of the target patient reaches a preset amplitude standard when the target patient executes the target rehabilitation training plan.
For example, it can be understood that, when a patient performs a rehabilitation training plan each time, it is required to monitor whether the motion amplitude of each training motion performed by the patient reaches a preset amplitude standard according to a preset motion detection threshold, and as the limb recovery condition of the patient changes continuously, the value of the motion detection threshold also needs to be adjusted to improve the accuracy of detecting whether the motion amplitude of the training motion reaches the preset amplitude standard. In the embodiment of the present disclosure, a deep learning and a technical scheme of training a deep learning model are also adopted, a detection threshold generation model is trained in advance, and the age characteristics, the body type characteristics, the initial training plan and the limb recovery information of a target patient are used as the inputs of the detection threshold generation model, so as to obtain the output result of the detection threshold generation model, where the output result is a target motion detection threshold.
In step 106, the target rehabilitation training program and the target motion detection threshold are output.
For example, by outputting the target rehabilitation training plan and the target action detection threshold value through the wearable hardware device in the embodiment of the present disclosure, it is convenient for the patient to execute the corresponding target action according to the target rehabilitation training plan output on the screen of the wearable hardware device. Meanwhile, the wearable hardware equipment collects the motion data of the patient in the process that the patient executes the target action and detects whether the action amplitude of the target action of the patient reaches a preset amplitude standard according to the target action detection threshold.
In summary, according to the technical scheme disclosed by the invention, the initial rehabilitation training plan can be corrected every day according to the information of the age, the body type, the recovery condition and the like of the patient, the rehabilitation training plan and the action detection threshold value in the same day are automatically generated, the accuracy of the formulated rehabilitation training plan and the action detection amplitude is improved, the generation of the rehabilitation training plan is not limited by time and place, a large amount of manpower is saved, and convenience is provided for the patient.
Fig. 2 is a flowchart of a patient information acquisition method according to fig. 1, and as shown in fig. 2, the step 103 includes:
in step 1031, when the target patient is performing the previous (i-1) rehabilitation exercises, the acceleration component of the target patient is acquired by the three-axis acceleration sensor.
In step 1032, the angular acceleration component and the angular velocity component of the target patient are acquired by the three-axis gyroscope sensors.
In step 1033, atmospheric pressure at the location of the target patient is obtained via a first barometric pressure sensor.
In step 1034, a bladder pressure value of the bladder in engagement with the limb of the target patient undergoing the rehabilitation training plan is obtained via the second pressure sensor.
Illustratively, the wearable hardware device in the embodiments of the disclosure is provided with a three-axis sensor, a three-axis gyroscope sensor, and a first air pressure sensor, which can be respectively used to obtain acceleration components, angular velocity components, and atmospheric pressure at a position of a target patient on three preset axes (i.e., an X axis, a Y axis, and a Z axis determined according to a preset coordinate system) of a limb end moving during rehabilitation training of the patient. And the wearable device is also provided with an air bag on the side attached to the limb of the target patient, and an air pressure sensor positioned in the wearable device acquires the air pressure value of the air bag. The pose information of the extremity of the patient can be determined by combining the three-axis acceleration sensor, the three-axis gyroscope sensor, the first air pressure sensor and the second air pressure sensor, so that whether the action amplitude of the patient reaches a preset amplitude standard or not is judged.
In step 1035, the name of the action and the number of actions performed by the target patient in each rehabilitation training program when performing the previous (i-1) rehabilitation training programs are determined according to the preset relationship among the acceleration component, the angular velocity component, the atmospheric pressure, the air pressure value of the air bag and the action name.
Illustratively, by a preset pose calculation method, the limb end pose of each action in each rehabilitation training plan when the patient executes the previous (i-1) rehabilitation training plans is calculated respectively, so as to judge whether the action reaches a preset amplitude standard according to the limb end pose.
In addition, after the position and posture of the limb end of the patient are acquired through the wearable hardware equipment, whether the patient falls down or not can be monitored in real time according to the position and posture information, so that the patient is monitored to fall down and alarm information is sent, and the falling down patient is prevented from having accidents.
Fig. 3 is a flowchart of another limb intelligent rehabilitation training method according to fig. 1, and as shown in fig. 3, after step 106, the method further includes:
in step 107, when the target patient executes the i-th rehabilitation training plan, whether the action amplitude of the target action executed by the target patient each time reaches a preset amplitude standard is detected according to the target action detection threshold.
In step 108, if the motion amplitude reaches the preset amplitude standard, 1 is added to the value of the motion frequency of the target motion completed in the i-th rehabilitation plan executed by the patient.
For example, it can be understood that, each time the amplitude of the action performed by the patient reaches the preset amplitude standard, 1 time of the specified action completed by the patient in the current rehabilitation training plan is recorded (correspondingly, if the amplitude of the action performed by the patient does not reach the preset amplitude standard, 0 time of the specified action completed by the patient in the current rehabilitation training plan is recorded), and when the current rehabilitation training plan is completed, the times of the joint completion of the actions in the current rehabilitation training plan of the patient will be displayed on the wearable hardware device.
In addition, the embodiment of the disclosure of the present invention further includes training the rehabilitation training plan generation model by using a deep learning model, where the deep learning model at least includes: the method comprises a convolutional neural network model and a cyclic neural network model, wherein the training method comprises the following steps: determining a sample population, the sample population comprising a preset number of patients; acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample group as training data; taking the training data as the input of the deep learning model; when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training; determining a model parameter corresponding to the current loss value; and obtaining the rehabilitation training plan generation model according to the model parameters.
It can be understood that the rehabilitation training plan generating model is updated every preset time period, the latest sample group in the time period is taken as a sampling group, and the deep learning model is retrained to obtain the latest rehabilitation training plan generating model.
Similarly, the embodiment of the disclosure of the present invention further includes training the detection threshold generation model by using a deep learning model, wherein the training method includes: determining a sample population, the sample population comprising a preset number of patients; acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample group as training data; taking the training data as the input of the deep learning model; when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training; determining a model parameter corresponding to the current loss value; and obtaining the detection threshold value generation model according to the model parameters. The detection threshold generation model also needs to be updated every other preset time period, the latest sample group in the time period is used as a sampling group, and the deep learning model is retrained to obtain the latest detection threshold generation model.
In addition, an embodiment of the present disclosure further provides a method for generating a detection threshold, that is: an initial range of detection thresholds is first determined, for example: when the patient executes a rehabilitation training plan (the rehabilitation training plan can complete the specified action 10 times for the specified patient), acquiring the motion data of the patient and the pose information of the extremity of the limb when executing each specified action through the three-axis acceleration sensor, the three-axis gyroscope sensor and the air pressure sensor. In the initial range of the detection threshold, the numerical value of each initial threshold is used as the detection threshold to judge that the action amplitude reaches the specified action amplitude threshold in a plurality of times in the current rehabilitation training plan (for example, the detection threshold a is 3 and is traversed until the detection threshold a is 8), and the times that the corresponding action amplitude reaches the specified action amplitude threshold under each numerical value of the detection threshold is recorded; the number of times that the motion amplitude reaches the specified motion amplitude threshold value obtained each time is subtracted from the number 10 (the rehabilitation training plan can finish the specified motion for 10 times for the specified patient), and the obtained initial detection threshold value a corresponding to the minimum difference value is used as the optimal detection threshold value when the patient executes the rehabilitation training plan.
Fig. 4 is a flow chart of a method for determining a quality level of an exercise plan according to fig. 1, as shown in fig. 4, the method further comprising:
in step 401, a target motion detection threshold for the target patient executing the i-th rehabilitation training program is obtained.
In step 402, the threshold variation amplitude of the target motion detection threshold is determined compared with all motion detection thresholds corresponding to each of the previous (i-1) times of rehabilitation training plans.
Illustratively, after the target motion detection threshold of the i-th rehabilitation training plan of the target patient is output through the step 106, the target motion detection threshold is obtained, and the target motion detection threshold is compared with the motion detection threshold corresponding to each previous rehabilitation training plan, so as to obtain the threshold variation range of the motion detection threshold from the 1 st time of executing the rehabilitation training plan of the target patient to the i-th time of executing the rehabilitation training plan.
In step 403, the exercise duration consumed by the target patient to execute the i-th rehabilitation training program is obtained.
For example, when the target patient executes the ith rehabilitation training plan, on the premise that the motion reaches the preset amplitude standard, the motion duration consumed by completing the target motion number corresponding to the ith rehabilitation training plan can be used for describing the difficulty of the ith rehabilitation training plan for the target patient. It can be understood that the longer the exercise time consumed by the target patient, the higher the difficulty of the i-th rehabilitation training plan for the target patient, that is, the poor rehabilitation effect of the target patient caused by executing the previous (i-1) rehabilitation training plans, that is, the poor quality of the previous (i-1) rehabilitation training plans.
In step 404, according to a preset motion plan quality monitoring strategy, determining a motion plan quality grade corresponding to the previous (i-1) rehabilitation training plans through the limb recovery information, the threshold variation amplitude and the motion duration of the target patient.
Illustratively, in order to determine the quality level of the rehabilitation training plan of the previous (i-1) times, an exercise plan quality monitoring strategy is preset, and the strategy is used for specifying the corresponding relation between the quality level of the rehabilitation training plan and the limb recovery information, the threshold change amplitude and the exercise duration of the target patient. The rehabilitation training program typically includes at least one predetermined quality level, such as high, medium, and low.
In conclusion, according to the technical scheme disclosed by the invention, the initial rehabilitation training plan can be corrected every day according to the information of the age, the body type, the recovery condition and the like of the patient, the rehabilitation training plan and the action detection threshold value in the same day are automatically generated, the accuracy of making the rehabilitation training plan and detecting the action amplitude is improved, the making of the rehabilitation training plan is not limited by time and place, a large amount of manpower is saved, and convenience is provided for the patient.
Fig. 5 is a block diagram illustrating a limb intelligent rehabilitation training system according to an exemplary embodiment, where the system 500 includes: a target patient determination module 510, an information determination module 520, a motion data collection device 530, a rehabilitation training plan generation module 540, a detection threshold generation module 550, and an output module 560, the target patient determination module 510 being connected to the information determination module 520, the information determination module 520 being connected to the motion data collection device 530, the motion data collection device 530 being connected to the rehabilitation training plan generation module 540, the rehabilitation training plan generation module 540 being connected to the detection threshold generation module 550, the detection threshold generation module 550 being connected to the output module 560;
a target patient determination module 510, configured to determine a target patient who needs to make an ith rehabilitation training plan, where the rehabilitation training plan is used to describe the name and number of actions that the target patient needs to complete;
an information determining module 520, configured to obtain an age characteristic, a body type characteristic and an initial training plan formulated by a doctor for the target patient, where the body type characteristic includes at least a height characteristic and a weight characteristic;
a motion data collecting device 530 for acquiring limb recovery information of the target patient, the limb recovery information including a name of an action and a number of actions performed by the target patient in each rehabilitation training plan when the target patient performed (i-1) times of rehabilitation training plans previously;
a rehabilitation training plan generating module 540, configured to use the age characteristics, body shape characteristics, initial training plan, and limb recovery information of the target patient as inputs of a pre-trained rehabilitation training plan generating model, so as to obtain a target rehabilitation training plan, which is output by the rehabilitation training plan generating model and needs to be executed by the target patient for the ith time;
a detection threshold generation module 550, configured to use the age characteristics, body shape characteristics, initial training plan, and limb recovery information of the target patient as inputs of a pre-trained detection threshold generation model to obtain a target motion detection threshold of the target patient output by the detection threshold generation model, where the target motion detection threshold is used to detect whether the motion amplitude of the target patient reaches a preset amplitude standard when the target patient executes the target rehabilitation training plan;
and an output module 560, configured to output the target rehabilitation training plan and the target motion detection threshold.
Optionally, the motion data collecting device 530 is configured to:
acquiring an acceleration component of the target patient through a three-axis acceleration sensor when the target patient is performing (i-1) times of rehabilitation training before the target patient is performed;
acquiring an angular acceleration component and an angular velocity component of the target patient through a three-axis gyroscope sensor;
acquiring the atmospheric pressure of the position of the target patient through a first air pressure sensor;
acquiring an air bag air pressure value of an air bag attached to a limb of the target patient for the rehabilitation training plan through a second air pressure sensor;
and determining the action name and the action times of the target patient in each rehabilitation training plan before (i-1) times of rehabilitation training plans are executed according to the preset relation among the acceleration component, the angular velocity component, the atmospheric pressure, the air pressure value of the air bag and the action name.
Fig. 6 is a block diagram illustrating another limb intelligent rehabilitation training system according to fig. 5, and as shown in fig. 6, the system 600 further includes: a motion detection device 570 and a count module 580, the motion detection device 570 being coupled to the output module 560, the count module 580 being coupled to the motion detection device 570;
the motion detection device 570 is configured to detect whether the motion amplitude of the target motion performed by the target patient each time reaches a preset amplitude standard according to the target motion detection threshold when the target patient performs the ith rehabilitation training plan;
the number counting module 580 is configured to add 1 to the number of actions of the target action completed in the ith rehabilitation plan executed by the patient if the action amplitude reaches the preset amplitude standard.
Optionally, the rehabilitation training plan generating module 540 is configured to:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample group as training data;
taking the training data as the input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the rehabilitation training plan generation model according to the model parameters.
Optionally, the detection threshold generating module 550 is configured to:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample group as training data;
taking the training data as the input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the detection threshold value generation model according to the model parameters.
Optionally, the system further includes: the device comprises a detection threshold generating module, a change amplitude determining module, a movement duration determining module and a quality grade determining module, wherein the detection threshold generating module is connected with the change amplitude determining module, the change amplitude determining module is connected with the movement duration determining module, and the movement duration determining module is connected with the quality grade determining module;
a detection threshold acquisition module, configured to acquire a target action detection threshold when the target patient executes the ith rehabilitation training plan;
a variation amplitude determination module, configured to determine a variation amplitude of the target motion detection threshold compared to threshold variation amplitudes of all motion detection thresholds corresponding to each of the previous (i-1) rehabilitation training plans;
a movement duration determination module, configured to obtain a movement duration consumed by the target patient to execute the ith rehabilitation training plan;
and the quality grade determining module is used for determining the quality grade of the motion plan corresponding to the previous (i-1) times of rehabilitation training plans according to a preset motion plan quality monitoring strategy and through the limb recovery information, the threshold value change amplitude and the motion duration of the target patient.
In conclusion, according to the technical scheme disclosed by the invention, the initial rehabilitation training plan can be corrected every day according to the information of the age, the body type, the recovery condition and the like of the patient, the rehabilitation training plan and the action detection threshold value in the same day are automatically generated, the accuracy of making the rehabilitation training plan and detecting the action amplitude is improved, the making of the rehabilitation training plan is not limited by time and place, a large amount of manpower is saved, and convenience is provided for the patient.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that, in the foregoing embodiments, various features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various combinations that are possible in the present disclosure are not described again.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.
Claims (10)
1. A limb intelligent rehabilitation training method is characterized by comprising the following steps:
determining a target patient needing to make an ith rehabilitation training plan, wherein the rehabilitation training plan is used for describing action names and action times needing to be completed by the target patient;
acquiring age characteristics and body shape characteristics of the target patient and an initial training plan made by a doctor for the target patient, wherein the body shape characteristics at least comprise height characteristics and weight characteristics;
acquiring limb recovery information of the target patient, wherein the limb recovery information comprises an action name and action times completed in each rehabilitation training plan when the target patient executes the (i-1) rehabilitation training plans previously;
taking the age characteristics, body type characteristics, initial training plan and limb recovery information of a target patient as the input of a pre-trained rehabilitation training plan generation model to obtain a target rehabilitation training plan which is output by the rehabilitation training plan generation model and needs to be executed by the target patient for the ith time;
taking the age characteristics, body shape characteristics, initial training plan and limb recovery information of a target patient as the input of a pre-trained detection threshold generation model to obtain a target action detection threshold of the target patient output by the detection threshold generation model, wherein the target action detection threshold is used for detecting whether the action amplitude of the target patient reaches a preset amplitude standard when the target patient executes the target rehabilitation training plan;
and outputting the target rehabilitation training plan and the target action detection threshold.
2. The limb intelligent rehabilitation training method according to claim 1, wherein the acquiring limb rehabilitation information of the target patient comprises:
acquiring an acceleration component of the target patient through a three-axis acceleration sensor when the target patient is performing (i-1) times of rehabilitation training before performing;
acquiring an angular acceleration component and an angular velocity component of the target patient through a three-axis gyroscope sensor;
acquiring the atmospheric pressure of the position of the target patient through a first air pressure sensor;
acquiring an air bag air pressure value of an air bag attached to a limb of the target patient for the rehabilitation training plan through a second air pressure sensor;
and determining the action name and the action times of the target patient in each rehabilitation training plan before (i-1) times of rehabilitation training plans are executed according to the preset relation among the acceleration component, the angular velocity component, the atmospheric pressure, the air pressure value of the air bag and the action name.
3. The limb intelligent rehabilitation training method according to claim 1, wherein after said outputting the target rehabilitation training plan and the target motion detection threshold, the method further comprises:
when the target patient executes the ith rehabilitation training plan, detecting whether the action amplitude of the target action executed by the target patient each time reaches a preset amplitude standard or not according to the target action detection threshold;
and if the action amplitude reaches the preset amplitude standard, adding 1 time to the numerical value of the action times of the target action completed in the ith rehabilitation plan executed by the patient.
4. The limb intelligent rehabilitation training method according to claim 1, further comprising training the rehabilitation training plan generation model using a deep learning model, wherein the training method is as follows:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample population as training data;
using the training data as an input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the rehabilitation training plan generation model according to the model parameters.
5. The limb intelligent rehabilitation training method according to claim 1, further comprising training the detection threshold generation model by using a deep learning model, wherein the training method comprises:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample population as training data;
using the training data as an input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the detection threshold value generation model according to the model parameters.
6. The limb intelligent rehabilitation training method according to claim 1, further comprising:
acquiring a target action detection threshold value when the target patient executes the ith rehabilitation training plan;
determining the threshold variation amplitude of the target action detection threshold compared with all action detection thresholds corresponding to each rehabilitation training plan in the previous (i-1) rehabilitation training plans;
acquiring the exercise duration consumed by the target patient to execute the ith rehabilitation training plan;
and (3) determining the quality grade of the exercise plan corresponding to the previous (i-1) rehabilitation training plan according to a preset exercise plan quality monitoring strategy and by means of the limb recovery information, the threshold variation amplitude and the exercise duration of the target patient.
7. A limb intelligent rehabilitation training system, the system comprising: the system comprises a target patient determination module, an information determination module, a motion data acquisition device, a rehabilitation training plan generation module, a detection threshold generation module and an output module, wherein the target patient determination module is connected with the information determination module, the information determination module is connected with the motion data acquisition device, the motion data acquisition device is connected with the rehabilitation training plan generation module, the rehabilitation training plan generation module is connected with the detection threshold generation module, and the detection threshold generation module is connected with the output module;
the target patient determination module is used for determining a target patient needing to make an ith rehabilitation training plan, and the rehabilitation training plan is used for describing action names and action times needing to be completed by the target patient;
the information determining module is used for acquiring the age characteristics and the body type characteristics of the target patient and an initial training plan made by a doctor for the target patient, wherein the body type characteristics at least comprise height characteristics and weight characteristics;
the motion data acquisition equipment is used for acquiring limb recovery information of the target patient, and the limb recovery information comprises action names and action times finished in each rehabilitation training plan when the target patient executes the (i-1) rehabilitation training plans previously;
the rehabilitation training plan generating module is used for taking the age characteristics, the body type characteristics, the initial training plan and the limb recovery information of the target patient as the input of a pre-trained rehabilitation training plan generating model so as to obtain a target rehabilitation training plan which is output by the rehabilitation training plan generating model and needs to be executed by the target patient for the ith time;
the detection threshold generation module is used for taking the age characteristics, the body type characteristics, the initial training plan and the limb recovery information of a target patient as the input of a pre-trained detection threshold generation model so as to obtain a target action detection threshold of the target patient output by the detection threshold generation model, wherein the target action detection threshold is used for detecting whether the action amplitude of the target patient reaches a preset amplitude standard or not when the target patient executes the target rehabilitation training plan;
and the output module is used for outputting the target rehabilitation training plan and the target action detection threshold.
8. The system of claim 7, wherein the motion data acquisition device is configured to:
acquiring an acceleration component of the target patient through a three-axis acceleration sensor when the target patient is performing (i-1) times of rehabilitation training before performing;
acquiring an angular acceleration component and an angular velocity component of the target patient through a three-axis gyroscope sensor;
acquiring the atmospheric pressure of the position of the target patient through a first air pressure sensor;
acquiring an air bag air pressure value of an air bag attached to a limb of the target patient for the rehabilitation training plan through a second air pressure sensor;
and determining the action name and the action times of the target patient in each rehabilitation training plan before (i-1) times of rehabilitation training plans are executed according to the preset relation among the acceleration component, the angular velocity component, the atmospheric pressure, the air pressure value of the air bag and the action name.
9. The system of claim 7, further comprising: the motion detection device is connected with the output module, and the frequency counting module is connected with the motion detection device;
the motion detection device is used for detecting whether the action amplitude of the target action executed by the target patient each time reaches a preset amplitude standard or not according to the target action detection threshold when the target patient executes the ith rehabilitation training plan;
and the frequency counting module is used for adding 1 time to the numerical value of the frequency of the target action finished in the ith rehabilitation plan executed by the patient if the action amplitude reaches a preset amplitude standard.
10. The system of claim 7, wherein the rehabilitation training plan generation module is configured to:
determining a sample population, the sample population comprising a preset number of patients;
acquiring age characteristics, body type characteristics, an initial training plan and limb recovery information of each patient in the sample population as training data;
using the training data as an input of the deep learning model;
when the loss value of the loss function of the deep learning model is smaller than a preset threshold value, stopping training;
determining a model parameter corresponding to the current loss value;
and obtaining the rehabilitation training plan generation model according to the model parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110149456.4A CN113223662B (en) | 2021-02-03 | 2021-02-03 | Intelligent limb rehabilitation training method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110149456.4A CN113223662B (en) | 2021-02-03 | 2021-02-03 | Intelligent limb rehabilitation training method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113223662A true CN113223662A (en) | 2021-08-06 |
| CN113223662B CN113223662B (en) | 2023-07-28 |
Family
ID=77084543
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110149456.4A Active CN113223662B (en) | 2021-02-03 | 2021-02-03 | Intelligent limb rehabilitation training method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113223662B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113707268A (en) * | 2021-08-27 | 2021-11-26 | 温州医科大学慈溪生物医药研究院 | Rehabilitation training evaluation method and system |
| CN114203275A (en) * | 2021-12-16 | 2022-03-18 | 江苏海洋大学 | Recovery state analysis system for rehabilitation training movement |
| CN115410690A (en) * | 2022-11-02 | 2022-11-29 | 山东宝德龙健身器材有限公司 | Rehabilitation training information management system and method |
| WO2023236414A1 (en) * | 2022-06-06 | 2023-12-14 | 郑州安杰莱智能科技有限公司 | Ankle joint rehabilitation training apparatus |
| CN117789920A (en) * | 2024-02-28 | 2024-03-29 | 吉林大学第一医院 | Nursing management system and method based on data mining |
| CN118053544A (en) * | 2024-02-01 | 2024-05-17 | 江苏医药职业学院 | Exercise injury rehabilitation training method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101340951A (en) * | 2005-10-19 | 2009-01-07 | 伯弗曼斯健康***有限公司 | Systems and methods for administering an exercise program |
| CN109102857A (en) * | 2018-05-31 | 2018-12-28 | 杭州同绘科技有限公司 | A kind of intelligence limb rehabilitation training system and method |
| CN110164524A (en) * | 2019-04-29 | 2019-08-23 | 北京国润健康医学投资有限公司 | A kind of hemiplegia patient training mission adaptive matching method and its system |
| CN110993057A (en) * | 2019-12-10 | 2020-04-10 | 上海金矢机器人科技有限公司 | Rehabilitation training system and method based on cloud platform and lower limb rehabilitation robot |
-
2021
- 2021-02-03 CN CN202110149456.4A patent/CN113223662B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101340951A (en) * | 2005-10-19 | 2009-01-07 | 伯弗曼斯健康***有限公司 | Systems and methods for administering an exercise program |
| CN109102857A (en) * | 2018-05-31 | 2018-12-28 | 杭州同绘科技有限公司 | A kind of intelligence limb rehabilitation training system and method |
| CN110164524A (en) * | 2019-04-29 | 2019-08-23 | 北京国润健康医学投资有限公司 | A kind of hemiplegia patient training mission adaptive matching method and its system |
| CN110993057A (en) * | 2019-12-10 | 2020-04-10 | 上海金矢机器人科技有限公司 | Rehabilitation training system and method based on cloud platform and lower limb rehabilitation robot |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113707268A (en) * | 2021-08-27 | 2021-11-26 | 温州医科大学慈溪生物医药研究院 | Rehabilitation training evaluation method and system |
| CN114203275A (en) * | 2021-12-16 | 2022-03-18 | 江苏海洋大学 | Recovery state analysis system for rehabilitation training movement |
| WO2023236414A1 (en) * | 2022-06-06 | 2023-12-14 | 郑州安杰莱智能科技有限公司 | Ankle joint rehabilitation training apparatus |
| EP4309638A4 (en) * | 2022-06-06 | 2024-01-24 | Zhengzhou Angelexo Intelligent Technologies Co., Ltd. | Ankle joint rehabilitation training apparatus |
| CN115410690A (en) * | 2022-11-02 | 2022-11-29 | 山东宝德龙健身器材有限公司 | Rehabilitation training information management system and method |
| CN118053544A (en) * | 2024-02-01 | 2024-05-17 | 江苏医药职业学院 | Exercise injury rehabilitation training method and system |
| CN118053544B (en) * | 2024-02-01 | 2024-07-23 | 江苏医药职业学院 | Exercise injury rehabilitation training method and system |
| CN117789920A (en) * | 2024-02-28 | 2024-03-29 | 吉林大学第一医院 | Nursing management system and method based on data mining |
| CN117789920B (en) * | 2024-02-28 | 2024-05-28 | 吉林大学第一医院 | Nursing management system and method based on data mining |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113223662B (en) | 2023-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113223662B (en) | Intelligent limb rehabilitation training method and system | |
| CN110334573B (en) | Human motion state discrimination method based on dense connection convolutional neural network | |
| CN105233485A (en) | A bodybuilding-assisting motion mode correction system and method | |
| KR101446653B1 (en) | Health and rehabilitation game apparatus, system and game method | |
| CN104841117A (en) | Counting method and system of sports times based on mobile terminal acceleration sensor | |
| CN205198906U (en) | A motion pattern corrects system for assisting body -building | |
| CN117012362B (en) | Adaptive data identification method, system, equipment and storage medium | |
| CN108958482B (en) | Similarity action recognition device and method based on convolutional neural network | |
| Bevilacqua et al. | Automatic classification of knee rehabilitation exercises using a single inertial sensor: A case study | |
| CN105411617A (en) | Respiratory gating equipment and method as well as MCU (Microcontroller Unit) | |
| CN106422274A (en) | Multi-sensor-based assessment system for yoga | |
| CN113749652A (en) | Muscle movement monitoring method and device | |
| CN206121113U (en) | Yoga evaluation system based on multisensor | |
| CN113856132A (en) | Rope skipping action recognition system and method and rope skipping | |
| US20240032820A1 (en) | System and method for self-learning and reference tuning activity monitor | |
| CN110880371A (en) | Human body posture testing and health monitoring management system | |
| JP2020137801A (en) | Device for estimating posture of human or the like | |
| CN210091521U (en) | Percussion training auxiliary device | |
| CN108742538B (en) | Physical sign measuring method based on big data and artificial intelligence and medical robot system | |
| CN108606795B (en) | Human body action data management system | |
| CN114392126B (en) | Disabled children's hand cooperation training system | |
| Chen | Gait feature extraction from inertial body sensor networks for medical applications | |
| CN108461117A (en) | A kind of heart information management system | |
| Joukov et al. | Generalized hebbian algorithm for wearable sensor rotation estimation | |
| CN116269266B (en) | AI-based old people health monitoring method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |