KR102590066B1 - health care system using single inertial measuring device based on machine learning - Google Patents

Abstract

A health care system related to an example of the present invention includes a tracker that generates motion information, a terminal that receives the motion information from the tracker, and a terminal that receives the motion information from the terminal and analyzes the received motion information to provide information to the terminal user. Includes a server that generates information related to the athletic ability of, wherein the tracker uses only one IMU (Inertial Measurement Unit), and the server analyzes the first motion information generated by the one IMU, Extracting a change in the joint angle of the terminal user, extracting at least one gait parameter of the terminal user based on the extracted change in the joint angle, and determining exercise ability of the terminal user based on the extracted at least one gait parameter can be estimated.

Description

Health care system using single inertial measuring device based on machine learning}

The present invention relates to a healthcare system using a single inertial measurement device. Specifically, the motion information generated by one IMU is analyzed, the change in the user's joint angle is extracted, and the gait parameter is calculated based on the extracted joint angle change. It relates to a healthcare system that extracts and estimates the user's exercise ability based on the extracted walking parameters.

Recently, with the rapid development of electronic communication technology, interest in developing technology to manage personal health in a network environment is increasing. For example, technology is being developed to detect biometric information and provide the detected information using wearable devices or smart devices.

In addition, the user's biometric information detected in this way is not only stored in the user terminal, but is also actively used to provide feedback on the user's health management.

Meanwhile, telemedicine, which has been widely discussed recently, is based on health information accumulated during the user's own life in order to obtain user-related data. To this end, a number of personalized health management applications using wearable devices and personal terminals have been developed.

However, this conventional health care system has limitations in that the collected information, such as physical information and behavioral information, is recorded separately, and the collected information cannot be integrated and provided in the form of integrated information.

In other words, the prior art simply presented numerical values measured through the wearable device or made simple evaluations through quantitative comparison with standard values, and had limitations in collecting measured information and drawing comprehensive conclusions. .

In particular, assuming that this healthcare system is used for rehabilitation of patients who have undergone knee surgery, etc., measured results can be provided to the patient through various measurement methods, but these are mere numbers, and professional A patient without knowledge could not help but have questions about his or her current condition and how to carry out rehabilitation exercises in the future.

Therefore, when constructing a healthcare system, the related industry does not simply distribute and list the measured factors, but collects the measured information in an integrated manner and synthesizes the collected information so that appropriate feedback can be provided from experts to users. There is a demand for the introduction of a new system.

Korea Patent Publication No. 10-2018-0069445 (2018.06.25) Korean registration number 10-1718471 Korean registration number 10-1751760 Korean registration number 10-1557068

The present invention was created to solve the above-described conventional problems, and the purpose of the present invention is to provide a healthcare system using one inertial measurement device.

Specifically, the present invention analyzes motion information generated by one IMU, extracts changes in the user's joint angles, extracts gait parameters based on the extracted changes in joint angles, and analyzes the user's joint angles based on the extracted gait parameters. The purpose is to provide a healthcare system that estimates exercise capacity.

In addition, the present invention acquires TOe Off (TO) point information and HS (Heel Strike) point information on a gait cycle basis, and obtains TO (Toe Off) point information and HS (Heel Strike) point information. The purpose is to provide a system for extracting walking parameters using .

In addition, the system according to the present invention aims to transmit information for improving the terminal user's athletic ability to the terminal based on the estimated athletic ability.

Meanwhile, the technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly apparent to those skilled in the art from the description below. It will be understandable.

A healthcare system related to an example of the present invention for realizing the above-described problem includes a tracker that generates motion information; A terminal that receives the operation information from the tracker; And a server that receives the motion information from the terminal, analyzes the received motion information, and generates information related to the terminal user's athletic ability. The tracker includes only one IMU (Inertial Measurement Unit). The server analyzes the first motion information generated by the one IMU, extracts a change in the joint angle of the terminal user, and moves at least one motion information of the terminal user based on the extracted change in the joint angle. Gait parameters may be extracted, and the terminal user's exercise ability may be estimated based on the extracted at least one gait parameter.

In addition, the first motion information includes at least x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to vertical movement (pitch), rotation (roll), and deflection (yaw) movement. It can be related to one thing.

In addition, the server uses at least one of x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to pitch, rotation, and yaw movement. Thus, joint angle changes related to the terminal user's hip, knee, and ankle can be extracted.

In addition, the server estimates the joint angle and extracts the joint angle change using an LSTM (Long Short Term Memory) model, which is one of the deep learning models and can use data from the initial node for learning. You can.

Additionally, the LSTM model can use features of x-axis, y-axis, and z-axis acceleration information related to pitch, rotation, and yaw movement.

In addition, the LSTM model can additionally use features of x-axis, y-axis, and z-axis angular velocity information related to pitch, rotation, and yaw movement.

In addition, the gait parameters include stance time, swing time, stride time, cadence, step length, and gait speed. ) may include at least one of the parameters.

In addition, the server uses y-axis acceleration information and z-axis angular velocity information related to pitch, rotation, and yaw movement in units of gait cycle, TO (Toe Off). Point information and HS (Heel Strike) point information can be acquired, and the walking parameters can be extracted using the obtained TO (Toe Off) point information and HS (Heel Strike) point information.

Additionally, the server may transmit information for improving the terminal user's athletic ability to the terminal based on the estimated athletic ability.

The present invention can provide a healthcare system using one inertial measurement device.

Specifically, the present invention analyzes motion information generated by one IMU, extracts changes in the user's joint angles, extracts gait parameters based on the extracted changes in joint angles, and analyzes the user's joint angles based on the extracted gait parameters. A healthcare system that estimates exercise capacity can be provided.

In addition, the present invention acquires TOe Off (TO) point information and HS (Heel Strike) point information on a gait cycle basis, and obtains TO (Toe Off) point information and HS (Heel Strike) point information. A system for extracting walking parameters can be provided using .

Additionally, the system according to the present invention can transmit information for improving the terminal user's exercise ability to the terminal based on the estimated exercise ability.

Meanwhile, the effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

The following drawings attached to this specification illustrate a preferred embodiment of the present invention, and serve to further understand the technical idea of the present invention along with the detailed description of the invention. Therefore, the present invention is limited to the matters described in such drawings. It should not be interpreted in a limited way.
Figure 1 shows a healthcare system using a tracker according to an embodiment of the present invention.
Figure 2 shows the attachment position of the tracker according to an embodiment of the present invention.
Figure 3 shows a block diagram of a tracker according to the present invention.
Figure 4 shows a block diagram of the server unit according to the present invention.
Figure 5 shows an example of the operation of a healthcare system using one inertial measurement device according to the present invention.
Figure 6 is a flowchart explaining the operation method of a healthcare system using one inertial measurement device according to the present invention.
Figure 7 is a diagram for explaining LSTM applied to the present invention.
Figure 8 is a diagram for explaining joint angle-related gait parameters in relation to the present invention.
Figure 9 is a diagram for explaining gait parameters related to SNT (Stance Time), SWT (Swing Time), SDT (Stride Time), and steps per minute (Cadence) in relation to the present invention.
Figure 10 is a diagram for explaining walking parameters related to step length and gait speed in relation to the present invention.
FIG. 11 is a diagram illustrating an example of obtaining and using TOe Off (TO) point information and Heel Strike (HS) point information in gait cycle units.

Hereinafter, preferred embodiments of a healthcare system using a tracker according to the present invention will be described in detail with reference to the attached drawings. In the following description of the present invention, if a detailed description of a known function or configuration is judged to unnecessarily obscure the gist of the present invention, the detailed description will be omitted. Unless otherwise specified, all terms in this specification have the same general meaning as understood by a person skilled in the art to which the present invention pertains, and if there is a conflict with the meaning of the terms used in this specification, this specification Follow the definitions used in the specification.

healthcare system

The health care system (1) using a tracker proposed by the present invention allows the user to accurately check his or her condition anytime, anywhere, regardless of time and place, so that the user is scheduled for surgery or is about to undergo surgery. In the case of a patient, the patient's recovery speed can be accelerated by allowing the patient to prepare for the surgery by accurately checking his or her condition before the surgery, and by allowing the patient to check his or her recovery status and management status at a glance after the surgery. , rehabilitation costs after surgery can be reduced.

Figure 1 shows a healthcare system using a tracker according to an embodiment of the present invention.

Referring to FIG. 1, the health care system 1 using a tracker includes a tracker 10 and a server unit 30.

Referring to FIG. 1, the tracker 10 is configured to generate motion data and transmit it to the user terminal (U). When the user moves while wearing the tracker 10, the tracker 10 tracks the user's movements. Motion data is generated by accurately measuring, and the generated motion data is transmitted to the server unit 30, which will be described later, through the user terminal (U) and analyzed, allowing the user to wear the tracker 10 without the need to pay special attention. Just by moving, the user's status can be objectively converted into data for analysis, management, etc.

Figure 2 shows the attachment position of the tracker according to an embodiment of the present invention.

Referring to FIG. 2, in the present invention, only one tracker 10 is used to generate motion data about the user's knee movement.

The tracker 10 according to the present invention may be attached to the femur side or to the user's tibia side.

The use of the tracker 10 is not limited to measuring knee joints, and can also be used to measure other joint movements of the body. It is not limited to use only in humans, but can also be used in animals, etc.

The motion data measured by the tracker 10 may include the gait cycle. Preferably, the gait cycle can be divided from the TO (Toe-off) point of the leg, when the acceleration in the axis direction becomes 0 until the acceleration in the forward direction becomes 0. The TO point refers to the time when the user's foot leaves the ground. When the acceleration in the ground axis direction becomes 0, when the force in the ground axis direction becomes 0, it refers to when the user lifts the foot that was on the ground during the walking process. In other words, when the acceleration in the moving direction becomes 0, it means when the force in the moving direction becomes 0 and when the user touches the ground while walking. The user's walking repeats this process, and the present invention refers to the user's one-legged walking cycle from when the acceleration in the axis direction becomes 0 at the TO point of one leg until the acceleration in the forward direction becomes 0. By dividing this gait cycle into multiple sections and analyzing them, it is possible to make an accurate diagnosis of the user's legs.

Furthermore, in the present invention, the biomechanical characteristics occurring at the TO and HS points can be used by utilizing the acceleration value of the y-axis and the angular velocity value of the z-axis.

Specifically, regarding TO, the moment you step on the ground, the acceleration value in the y-axis direction increases significantly due to ground reaction force, and TO can be selected based on the maximum value at that point.

In addition, regarding HS, the angular velocity value in the x-axis direction decreases significantly at the transition from stance time to swing time, and HS can be selected based on the minimum value.

This will be described in detail later with reference to FIG. 11 .

The tracker 10 can be configured to enable repeated reuse, and can also be manufactured for one-time use and discarded after use.

Figure 3 shows a block diagram of a tracker according to the present invention.

Referring to FIG. 3, this tracker 10 includes a sensor unit 11 and a communication unit 13.

The sensor unit 11 is configured to measure status indicators of the worn part, such as left and right balance measurement, bending angle measurement, acceleration measurement, angular velocity measurement, angle measurement, and gait measurement, etc. It refers to a construct that measures various factors that can be identified. The sensor unit 11 includes an acceleration measurement module 111, an angular velocity measurement module 113, and an angle measurement module 115.

The acceleration measurement module 111 is a component that measures the acceleration of the area where the tracker 10 is attached, and refers to a component that measures the change in speed per unit time.

The angular velocity measurement module 113 is a configuration that measures the angular velocity of the area where the tracker 10 is attached, and refers to a configuration that measures the angle rotated in unit time based on the rotation center point, etc.

The angle measurement module 115 is a component that measures the angle of the area where the tracker 10 is attached, and refers to a component that measures the angle between changes in movement based on the rotation center point, etc.

The communication unit 13 refers to a component that connects the user terminal (U) and the tracker 10 for communication. Although it is not excluded that the user terminal (U) and the tracker 10 are connected by wired communication, preferably, the user terminal (U) and the tracker 10 may be connected by wireless communication. The communication unit 13 may include a common identification identifier.

The server unit 30 is connected to the user terminal (U), receives data from the user terminal (U), analyzes the received data, and transmits the data to the user terminal (U). As a result, data can be freely transmitted from the user terminal (U) to the server unit 30, and the server unit 30 analyzes and stores the received data when there is a request from the user terminal (U). , analyzed data, etc. can be provided to the user terminal (U). In addition, the server unit 30 can be connected to a plurality of terminals through a network, so a user terminal (U) such as a patient connects to the network and transmits data uploaded to the server unit 30 to an expert terminal (such as a doctor). E) can connect to the network and download from the server unit 30, and conversely, the user terminal (U) can download the data uploaded to the server unit 30 by the expert terminal (E). Since it is possible to download from, even if a plurality of terminals are remote from each other, easy mutual communication is possible centering on the server unit 30.

Figure 4 shows a block diagram of the server unit according to the present invention.

Referring to FIG. 4, the server unit 30 includes a data reception unit 31, a data analysis unit 33, a data storage unit 35, and a data transmission unit 37.

The data receiving unit 31 refers to a component that receives data from the user terminal (U). As described above, the server unit 30 can be connected not only to the user terminal (U) but also to a plurality of terminals, and the data reception unit 31 is preferably viewed as a configuration that receives data transmitted from the terminal. do. The data received by the data receiver 31 includes motion data transmitted from the user terminal (U) using the tracker 10, authentication data, pain data, medication data, and questionnaire data generated by the user terminal (U) itself. , Biometric data generated by biometric devices such as smart watches other than the tracker 10 may be included.

The data analysis unit 33 is connected to the data reception unit 31 and analyzes the received data. Various data transmitted from the user terminal (U) are analyzed on the server unit 30 and converted into meaningful information. Enables transmission to the user terminal (U).

Healthcare system using one inertial measurement unit (IMU)

Figure 5 shows an example of the operation of a healthcare system using one inertial measurement device according to the present invention.

Referring to FIG. 5, the tracker 10 generates operation information, and the terminal U receives the operation information from the tracker 10.

Additionally, the server 30 receives motion information from the terminal U, analyzes the received motion information, and generates information related to the exercise ability of the user of the terminal U.

At this time, the tracker 10 according to the present invention can use only one IMU (Inertial Measurement Unit).

The user's status must be determined based only on information obtained from one IMU, and the server 30 according to the present invention analyzes the first motion information generated by one IMU and extracts the change in the joint angle of the terminal user. do.

Here, the first motion information is at least one of x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to vertical movement (pitch), rotation (roll), and deflection (yaw) movement. It may be information related to .

In addition, using at least one of x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to pitch, rotation, and yaw movement, the terminal Joint angle changes related to the user's hip, knee, and ankle can be extracted.

Additionally, the server 30 may extract at least one gait parameter of the terminal user based on the extracted change in joint angle, and estimate the exercise ability of the terminal user based on the at least one extracted gait parameter.

At this time, the server 30 can estimate the joint angle and extract the joint angle change using an LSTM model, which is one of the deep learning models and can use data from the initial node for learning.

In addition, gait parameters include stance time, swing time, stride time, cadence, step length, and gait speed. It may contain at least one of the parameters.

Additionally, the server 30 may transmit information for improving the exercise ability of the user of the terminal U to the terminal, based on the estimated exercise ability.

The first technical feature of the present invention is that, unlike the existing case where joint angle measurement and gait parameter extraction are calculated using two or more IMUs, only one IMU is used to extract various gait parameters.

In addition, as a second technical feature of the present invention, hip, knee, and ankle joint angles can be extracted by using the LSTM model, one of the deep learning models, to estimate joint angles.

Additionally, as a third technical feature of the present invention, when using an existing IMU, various walking parameters extracted from estimated joint angles can be extracted, rather than walking parameters inferred through changes in acceleration gyro values.

Healthcare method using one inertial measurement device

Figure 6 is a flowchart explaining the operation method of a healthcare system using one inertial measurement device according to the present invention.

Referring to FIG. 6, a step (S1) in which the tracker 10 generates operation information is performed.

Afterwards, a step (S2) in which the terminal (U) receives operation information from the tracker (10) proceeds.

In addition, the server 30 receives motion information from the terminal U (S3), and the server 30 analyzes the first motion information generated by one IMU (Inertial Measurement Unit) included in the tracker 10. Thus, the change in the joint angle of the terminal user is extracted (S4).

Thereafter, the server 30 extracts at least one gait parameter of the terminal user based on the change in the extracted joint angle (S5), and the server 30 extracts the terminal user's gait parameter based on the extracted at least one gait parameter. The step (S6) of estimating exercise ability proceeds.

In addition, a step (S7) is performed in which the server 30 transmits information for improving the terminal user's exercise ability to the terminal based on the estimated exercise ability.

Below, the method for extracting joint angle changes and gait parameters according to the present invention will be described in more detail with reference to the drawings.

Extraction of joint angle changes

Figure 7 is a diagram for explaining the LSTM model applied to the present invention.

Referring to FIG. 7, the server 30 can estimate the joint angle and extract the joint angle change using an LSTM model, which is one of the deep learning models and can use the data of the initial node for learning. there is.

LSTM is a deep learning model, and RNN (Recurrent Neural Network), such as (a) in Figure 7, is an algorithm that uses past data for learning and is a model suitable for time series data.

Simple RNN models experience gradient vanishing (difficulty using data from initial nodes for learning) during the back-propagation process.

LSTM, as shown in (b) of Figure 7, which can prevent this and learn by leaving important information, is judged to be suitable for gait analysis, and a joint angle extraction algorithm using it can be applied.

Here, the LSTM model can use features of x-, y-, and z-axis acceleration information related to pitch, rotation, and yaw movement.

In addition, the LSTM model can additionally use features of x-axis, y-axis, and z-axis angular velocity information related to pitch, rotation, and yaw movement.

Extraction of Gait Parameters

Figure 8 is a diagram for explaining joint angle-related gait parameters in relation to the present invention.

Referring to FIG. 8, the server 30 provides x-, y-, and z-axis acceleration information and x-, y-, and z-axis angular velocities related to pitch, roll, and yaw movement. Using at least one of the information, joint angle changes related to the terminal user's hip, knee, and ankle are extracted.

Regarding the analysis method and results, referring to FIG. 8, the inventor of the present invention used the LSTM model algorithm to learn with data from 25 randomly detected people out of 30 people and then verified the effectiveness of the algorithm with the remaining 5 people. did.

In addition, as shown in Figure 8, an LSTM model with N features (IMU data - 3-axis acceleration, 3-angular velocity, rotation (Roll), movement (Pitch), and deflection (Yaw) values) was applied.

Additionally, as shown in Figure 8, the joint angles of the hip, knee, and ankle (gait parameter 1) during walking based on the LSTM model were extracted, and the accuracy of the algorithm-based predicted joint angles was confirmed using a motion analysis system.

In FIG. 8, Range Of Motion means ROM, and ROM can be derived by Equation 1 below.

Equation 1

ROM = Angle(max) - Angle (min)

Meanwhile, Figure 9 is a diagram for explaining walking parameters related to SNT, SWT, SDT, and steps per minute (Cadence) in relation to the present invention.

In Figure 9, the parameter terms are as follows.

- Gait Time - GT

- Stance Time - SNT

- Swing Time - SWT

- Stride Time - SDT

Additionally, SDT can be derived from Equation 2 below, and steps per minute (Cadence) can be derived from Equation 3 below.

Equation 2

SDT = SNT + SWT

Equation 3

Cadence = Steps/min

In addition, Figure 10 is a diagram for explaining walking parameters related to step length and gait speed in relation to the present invention.

In Figure 10, parameter terms can be organized as follows.

- Step length -

- Distal lower limb length -

- Knee angle at the time of toe off -

- Stride Time - SDT

- Gait Speed - GS

also, can be derived by the following equation 4, and GS can be derived by the following equation 5.

Equation 4

Equation 5

Referring to Figures 9 and 10, the walking parameters include stance time, swing time, stride time, cadence per minute, and step length. and at least one of a gait speed parameter.

FIG. 11 is a diagram illustrating an example of obtaining and using TOe Off (TO) point information and Heel Strike (HS) point information in gait cycle units.

Referring to FIG. 11, the server 30 uses y-axis acceleration information and z-axis angular velocity information related to pitch, rotation, and yaw movement in units of gait cycles. , TO (Toe Off) point information and HS (Heel Strike) point information can be acquired, and the walking parameters can be extracted using the obtained TO (Toe Off) point information and HS (Heel Strike) point information.

The meaning of the terms in FIG. 11 is as follows.

- SNT - Stance Time

- SWT - Swing Time

- SDT - Stride Time

-TO-Toe Off

- HS - Heel Strike

Additionally, as walking parameter 2, Cadence, SNT/SWT/SDT, etc. are used.

In relation to extracting gait parameter 2, in the present invention, TO and HS points for each gait cycle are obtained using angular velocity (z-axis) and acceleration (y-axix) values, and based on this, gait parameter 1 (SNT) is obtained. /SWT/SDT Cadence) can be obtained.

Additionally, the server 30 transmits information for improving the terminal user's exercise ability to the terminal U based on the estimated exercise ability.

Effects according to the present invention

The present invention can provide a healthcare system using a single inertial measurement device.

Specifically, the present invention analyzes the first motion information generated by one IMU, extracts the change in the user's joint angle, extracts gait parameters based on the extracted change in joint angle, and based on the extracted gait parameter A healthcare system that estimates a user's exercise ability can be provided.

In addition, the present invention acquires TOe Off (TO) point information and HS (Heel Strike) point information on a gait cycle basis, and obtains TO (Toe Off) point information and HS (Heel Strike) point information. A system for extracting walking parameters can be provided using .

Additionally, the system according to the present invention can transmit information for improving the terminal user's exercise ability to the terminal based on the estimated exercise ability.

Meanwhile, the effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

In addition, the apparatus and method described above are not limited to the configuration and method of the above-described embodiments, but all or part of the embodiments can be selectively combined so that various modifications can be made. It may be composed.

Claims (9)

A tracker that generates motion information;
A terminal that receives the operation information from the tracker; and
A server that receives the motion information from the terminal, analyzes the received motion information, and generates information related to the exercise ability of the terminal user,

The tracker uses only one IMU (Inertial Measurement Unit),

The server is,
By analyzing the first motion information generated by the one IMU, a change in the joint angle of the terminal user is extracted,
Extracting at least one gait parameter of the terminal user based on the extracted change in joint angle,
Estimating the exercise ability of the terminal user based on the extracted at least one walking parameter,

The first operation information is,
It is associated with at least one of x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to pitch, roll, and yaw movement,

The server is,
Using at least one of x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to pitch, rotation, and yaw movement, the terminal user Extracts joint angle changes related to the hip, knee, and ankle.

The server is,
Using an LSTM model, which is one of deep learning models and can use data from initial nodes for learning, estimate the joint angle and extract the joint angle change,

The LSTM model is,
It uses the features of x-axis, y-axis, and z-axis acceleration information related to pitch, roll, and yaw movement,

The LSTM model is,
Additional features of x-axis, y-axis, and z-axis angular velocity information related to pitch, roll, and yaw movement are used,

The walking parameters are,
Includes at least one of the following parameters: Stance Time, Swing Time, Stride Time, Cadence, Step Length, and Gait Speed. And

The server is,
Using y-axis acceleration information and z-axis angular velocity information related to pitch, roll, and yaw movement, TOe Off (TO) point information and HS (Heel) are generated in units of gait cycle. Strike) obtain branch information,
The walking parameters are extracted using the obtained TO (Toe Off) point information and HS (Heel Strike) point information,

The step length parameter is extracted according to Equation 1 below,

A health care system wherein the Gait Speed parameter is extracted according to Equation 2 below.

Equation 1

In Equation 1 above, means step length, means distal lower limb length, means knee angle at the time of toe off.

Equation 2

In Equation 2 above, means gait speed, means steps per minute.
According to clause 1,
The health care system is characterized in that the server transmits information for improving the exercise ability of the terminal user to the terminal based on the estimated exercise ability.
A tracker generating motion information;
A terminal receiving the operation information from the tracker; and
A server receiving the motion information from the terminal, analyzing the received motion information, and generating information related to the exercise ability of the terminal user,

The tracker uses only one IMU (Inertial Measurement Unit),

The server is,
By analyzing the first motion information generated by the one IMU, a change in the joint angle of the terminal user is extracted,
Extracting at least one gait parameter of the terminal user based on the extracted change in joint angle,
Estimating the exercise ability of the terminal user based on the extracted at least one walking parameter,

The first operation information is,
It is associated with at least one of x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to pitch, roll, and yaw movement,

The server is,
Using at least one of x-axis, y-axis, and z-axis acceleration information and x-axis, y-axis, and z-axis angular velocity information related to pitch, rotation, and yaw movement, the terminal user Extracts joint angle changes related to the hip, knee, and ankle.

The server is,
Using an LSTM model, which is one of deep learning models and can use data from initial nodes for learning, estimate the joint angle and extract the joint angle change,

The LSTM model is,
It uses the features of x-axis, y-axis, and z-axis acceleration information related to pitch, roll, and yaw movement,

The LSTM model is,
Additional features of x-axis, y-axis, and z-axis angular velocity information related to pitch, roll, and yaw movement are used,

The walking parameters are,
Includes at least one of the following parameters: Stance Time, Swing Time, Stride Time, Cadence, Step Length, and Gait Speed. And

The server is,
Using y-axis acceleration information and z-axis angular velocity information related to pitch, roll, and yaw movement, TO (Toe Off) point information and HS (Heel) are generated in units of gait cycle. Strike) Obtain branch information,
The walking parameters are extracted using the obtained TO (Toe Off) point information and HS (Heel Strike) point information,

The step length parameter is extracted according to Equation 1 below,

A healthcare method, characterized in that the gait speed parameter is extracted according to Equation 2 below.

Equation 1

In Equation 1 above, means step length, means distal lower limb length, means knee angle at the time of toe off.

Equation 2

In Equation 2 above, means gait speed, means steps per minute. delete delete delete delete delete delete