KR20210045555A - Method and system for evaluating elderly health status using characteristics of walk-in-place - Google Patents

Abstract

Provided are a method for evaluating a health status of the elderly using gait characteristics and a system thereof. According to an embodiment of the present invention, the method for evaluating a health status of the elderly using gait characteristics comprises: a gait data collecting unit configured to collect gait data from a target person; a gait characteristic analyzing unit configured to extract gait characteristics by analyzing the gait data; a health status learning unit configured to machine-learn the health status of the target person about the gait characteristics to build a health status evaluation model based on the gait characteristics; and a health status evaluating unit configured to evaluate a health status of a new target person by inputting gate characteristics, obtained from the new target person, to the health status evaluation model.

Description

Method and system for evaluating elderly health status using characteristics of walk-in-place}

The present invention relates to a method and system for evaluating the elderly's health status using in-place walking characteristics, and to a method and system capable of evaluating the elderly’s health status based on the in-place walking characteristics measured using simple equipment in a limited space.

As the population is rapidly aging, personal and social burdens are increasing due to medical expenses for the elderly. It is known that the health status of the elderly affects gait characteristics. Walking is a complex process in which well-balanced dynamic movements occur at the same time in several joints of the lower extremities, and high-level cognitive functions such as memory, concentration, and executive ability are involved, as well as motor and sensory skills. That is, gait characteristics are osteoarthritis of the lower extremities (ankles, knees, hips), neuromuscular diseases (eg, degenerative spinal diseases), brain dysfunction (eg, dementia), cerebrovascular It may reflect health conditions such as cerebrovascular disease (eg, stroke), cardiorespiratory diseases, metabolic disturbances, and the like. Therefore, if the health status of the elderly is evaluated using gait characteristics, it can be mainly used for early diagnosis of related diseases, symptom management, and appropriate treatment.

However, in order to analyze and quantify such walking ability, U.S. Patent Publication US 2009-0141933 uses a camera to obtain an image while walking, analyzes the movement of the leg to predict the motion, and predicts the walking cycle. Suggested. However, such a conventional technology is an expensive motion capture system (motion capture system), electromyograph (electromyograph), equipment such as force plate (force plate), free space such as a linear structure space of up to 16 meters long, and experienced experiments I need a ruler. That is, the method of evaluating walking ability using the above-described equipment not only incurs high cost due to the input of special equipment and manpower, but it is difficult for the elderly with disabilities to fully perform the above experiment, so that accurate result data can be obtained. There was a difficult drawback.

Therefore, there is a demand for a method and system capable of measuring gait in a limited space using simple equipment and evaluating the health status of the elderly.

US Patent Publication No. US 2009-0141933 A1 (June 04, 2009)

The present invention has been conceived to solve the above-described problems, and specifically, provides a method and system capable of measuring a subject's in-place walking characteristics using a simple device in a limited space and evaluating the subject's health status based thereon. .

The elderly health condition evaluation system according to an embodiment of the present invention comprises: a step in place data collection unit for collecting step data in place from a subject; An in-place walking characteristic analysis unit that analyzes the in-place walking data to extract in-place walking characteristics; A health state learning unit for constructing a health state evaluation model based on the in-place walking characteristic by machine learning the health state of the subject with respect to the in-place walking characteristic; And a health state evaluation unit that evaluates the health state of the new subject by inputting the in-place walking characteristic acquired from the new subject into the health state evaluation model.

The method for evaluating the health status of the elderly according to an embodiment of the present invention acquires in-place walking data according to in-place walking from a plurality of subjects, extracts in-place walking characteristics by analyzing the in-place walking data, and extracts in-place walking characteristics, and health based on the in-place walking characteristics. Building a state evaluation model; Collecting in-place walking data from the evaluation subject and extracting in-place walking characteristics; And evaluating the health status of the evaluation subject by inputting the in-place walking characteristics of the evaluation subject into the constructed health condition evaluation model.

The health status evaluation method and system for the elderly according to the present embodiment is based on the in-place walking data, so it does not use the expensive gait analysis system used in the past, and evaluates the health status of the subject in a method that requires less professional manpower and physical space. You can collect gait data that can be used for

By analyzing the in-place walking characteristics in a simple and automated method and using it as an input of an artificial neural network model, the time-consuming and labor-intensive process required for processing input data and feature engineering can be minimized.

By monitoring the health conditions related to senility, muscle loss (risk of falls), and cognitive dysfunction in the elderly by using the easy-to-measurable in-place walking characteristics during daily life, it is possible to provide opportunities for early diagnosis of related diseases, symptom management, and appropriate treatment. have.

1 is a block diagram of a system for evaluating an elderly health condition using an in-place walking characteristic according to an embodiment of the present invention.
2A is a diagram exemplarily showing a subject participating in the system for evaluating a health condition for the elderly and a sensor attached to the subject according to an embodiment of the present invention.
2B is an exemplary view showing a schematic configuration of a step data collection unit in place according to an embodiment of the present invention.
3 is an exemplary graph of a signal measured by attaching a 3-axis accelerometer to a foot when walking in place.
4A and 4B are exemplary diagrams in which the signal of FIG. 3 is converted into a spectrogram for analyzing the characteristics of walking in place.
5 illustrates an exemplary machine learning model of a health state learning unit of an elderly health state evaluation system according to an embodiment of the present invention.
6 is a flowchart of a method for evaluating a health condition for the elderly using an in-place walking characteristic according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. DETAILED DESCRIPTION OF THE INVENTION The detailed description to be disclosed below together with the accompanying drawings is intended to describe exemplary embodiments of the present invention, and is not intended to represent the only embodiments in which the present invention may be practiced. The following detailed description includes specific details to provide a thorough understanding of the present invention. However, one of ordinary skill in the art can appreciate that the invention may be practiced without these specific details. Specific terms used in the following description are provided to aid understanding of the present invention, and the use of these specific terms may be changed in other forms without departing from the technical spirit of the present invention.

1 is a block diagram of a system for evaluating an elderly health condition using an in-place walking characteristic according to an embodiment of the present invention. 2A is a diagram exemplarily illustrating a subject participating in the system for evaluating a health condition for the elderly and a sensor attached to the subject according to an embodiment of the present invention. 2B is an exemplary view showing a schematic configuration of a step data collection unit in place according to an embodiment of the present invention. 3 is an exemplary graph of a signal measured by attaching a 3-axis accelerometer to a foot when walking in place. 4A and 4B are exemplary diagrams in which the signal of FIG. 3 is converted into a spectrogram for analyzing the characteristics of walking in place. 5 illustrates an exemplary machine learning model of a health state learning unit of an elderly health state evaluation system according to an embodiment of the present invention.

Referring to FIG. 1, the elderly health condition evaluation system 10 includes a step in place data collection unit 100, a step in place characteristic analysis unit 110, a health state learning unit 120, a health state evaluation unit 130, and data. Includes a base portion 140.

The elderly health condition evaluation system according to the embodiments may have an aspect that is entirely hardware, or partially hardware and partially software. For example, the elderly health condition evaluation system of the present specification and each unit included therein may collectively refer to a device for sending and receiving data of a specific format and content through an electronic communication method and software related thereto. In this specification, terms such as "unit", "module", "server", "system", "device" or "terminal" refer to a combination of hardware and software driven by the hardware. Is intended to be. For example, the hardware here may be a data processing device including a central processing unit (CPU) or another processor. In addition, software driven by hardware may refer to an executing process, an object, an executable file, a thread of execution, a program, and the like.

In addition, each unit constituting the elderly health condition evaluation system is not necessarily intended to refer to separate components that are physically separated. In FIG. 1, the in-place step data collection unit 100, the in-place step characteristic analysis unit 110, the health state learning unit 120, the health state evaluation unit 130, and the database unit 140 are separate blocks separated from each other. Although shown as, this is merely functionally dividing the apparatus constituting the elderly health condition evaluation system by the operation executed by the apparatus. Therefore, depending on the embodiment, the in-place step data collection unit 100, the in-place step characteristic analysis unit 110, the health state learning unit 120, the health state evaluation unit 130, and the database unit 140 are partially or All of them may be integrated into the same device, one or more may be implemented as separate devices that are physically separated from other units, and may be components that are communicatively connected to each other in a distributed computing environment.

In addition, the target of the elderly health condition evaluation system 10 according to an embodiment of the present invention may be the elderly. Here, the elderly refers to a person who is older than a certain age, and may be defined as a person who is 65 years or older according to the Elderly Welfare Act, but is not limited thereto. The subject of the present invention may be composed of subjects under the above age, or may be composed of subjects above the above age.

The in-place walking data collection unit 100 may collect in-place walking data from the subject. The elderly health condition evaluation system 10 may include one or more sensors 101 for measuring the in-place step data.

Here, the sensor 101 may include an accelerometer or a gyroscope of a small size that is easily detachable and attached to the body of the subject H and is portable.

When collecting kinematic data of an in-place walking motion using an accelerometer or a gyroscope, the subject H participating in the elderly health condition evaluation system 10 as shown in FIG. 2A is equipped with a sensor 101 on the body. It is done. The body part of the subject H on which the sensor 101 is mounted may be a body part directly related to the in-place walking motion of the subject H. Exemplarily, the body part of the subject H to which the sensor 101 is mounted may be at least one of a pelvis, a thigh, a shank, and a foot of the subject H, It is not limited thereto.

The accelerometer may measure acceleration or impact strength of a moving object, and a gyroscope may measure a rotation angle and inclination of a rotating object. That is, the accelerometer and gyroscope attached to the body segment of the subject H may provide quantitative data on the motion occurring in the body joint during the in-place walking motion to the in-place walking data collection unit 100.

That is, the sensor 101 plays an instrumental role in evaluating the health state of the subject H by using the in-place walking characteristic. The in-place walking data collection unit 100 may receive in-place walking data of the subject H from each sensor 101. The in-place walking data collection unit 100 and the sensor 101 may be configured to transmit and receive signals wirelessly.

In some embodiments, the elderly health condition assessment system 10 may further include a camera 102 and a pressure mat 103, as shown in FIG. 2B. The camera 102 may measure the position and angle of the user's body joints. The pressure mat 103 corresponds to a limited space in which the subject's walking in place is performed. In addition, the pressure mat 103 may measure the user's plantar pressure. The user's in-place walking data (position and angle of the user's body joint) collected through the camera 102 and the in-place walking data (user's plantar pressure) collected through the pressure mat 103 are the in-place walking data collection unit 100 Can be provided on. That is, the in-place step data may include at least one of the measured kinematic data, the measured body joint position and angle of the subject, and the measured plantar pressure of the subject.

The in-place step characteristic analysis unit 110 analyzes the in-place step data received from the in-place step data collection unit 100 to extract the in-place step characteristic. Specifically, the in-place walking characteristic analysis unit 110 extracts parameters related to spatiotemporal walking elements, joint angles, and plantar pressure changes. In addition, the in-place walking characteristic analysis unit 110 may acquire images related to changes in spatiotemporal walking elements, joint angles, and plantar pressure. That is, the in-place walking characteristic may include at least one of a spatiotemporal gait factor, a parameter related to a joint angle and plantar pressure change, and an image related to a spatiotemporal gait factor, joint angle, and plantar pressure change. Here, the image may be a spectrogram showing a change in frequency and energy of a signal over time. The in-place walking characteristic analysis unit 110 may convert a time domain signal into a spectrogram through Fourier transform or wavelet transform.

4A is an exemplary diagram illustrating a spectrogram generated by applying a local Fourier transform (STFT) to the signal of FIG. 3, and FIG. 4B illustrates a continuous wavelet transform (CWT). The spectrogram generated by applying is illustrated as an example. The image representing the kinematic data of the in-place walking motion in the in-place walking characteristic analysis unit 110 may be a spectrogram, but is not limited thereto. The health state learning unit 120 constructs a health state evaluation model based on the in-place walking characteristic by machine learning the health state of the subject with respect to the in-place walking characteristic. Here, the in-place walking characteristic may include a spatiotemporal walking element analyzed by the in-place walking characteristic analysis unit 110, a parameter related to a joint angle and plantar pressure change, a spatiotemporal walking element, a joint angle, and an image related to a change in plantar pressure.

The health state learning unit 120 is a set of machine learning algorithms that attempt a high level of abstraction (a task of summarizing key contents or functions in a large amount of data or complex data) through a combination of several nonlinear transformation methods. Using deep learning, a health condition evaluation model having a structure as shown in FIG. 5 can be constructed. The deep learning model is a convolutional neural network that forms a connection pattern between neurons, similar to the structure of an animal's visual cortex, a recurrent neural network that builds up a neural network every moment over time, and a probability distribution for an input set. It may be based on a restricted Boltzmann machine that can learn, but is not limited thereto.

An input used for learning in the health state learning unit 120 is a characteristic of a step in place of each of a plurality of subjects, and an output is a state of health of each of the plurality of subjects.

Here, the health state of the subject includes the degree of senility of the subject, the degree of muscle reduction of the subject, and/or the degree of cognitive dysfunction of the subject. The degree of senility of the subject may be divided into non-frailty or frailty, or may be divided into non-frailty, pre-frailty, or frailty. The health state learning unit 120 uses the in-place walking characteristic (parameter and/or image) generated by the in-place walking characteristic analysis unit 110 as an input value, and uses the degree of senility of the subject as an output value to obtain a first health state. Build an evaluation model.

In addition, the degree of muscle reduction of the subject may be divided into non-sarcopenia or sarcopenia. The health state learning unit 120 may construct a second health state evaluation model by using the in-place walking characteristic generated by the in-place walking characteristic analysis unit 110 as an input value and using the degree of muscle reduction of the subject as an output value. have.

In addition, the degree of cognitive impairment of the subject is divided into non-cognitive impairment (non-CI) or cognitive impairment (CI), or non-cognitive impairment, mild cognitive impairment (mild CI), or moderate or higher cognitive impairment ( moderate-to-severe CI). The health state learning unit 120 may build a third health state evaluation model by using the in-place walking characteristic generated by the in-place walking characteristic analysis unit 110 as an input value and using the degree of cognitive dysfunction as an output value. .

The health state evaluation model constructed by the health state learning unit 120 includes a first health state evaluation model that determines the degree of senility of the subject, a second health state evaluation model that determines the degree of muscle reduction of the subject, and the subject's perception. It may include at least one of the third health condition evaluation models for determining the degree of dysfunction.

The health state evaluation model constructed by the health state learning unit 120 may be stored in the database unit 140. The database unit 140 may be configured to temporarily store or store information such as a health condition evaluation model and in-place walking data and/or in-place walking characteristics based thereon.

The health condition evaluation unit 130 may evaluate the health condition of a new subject by using the health condition evaluation model constructed by the health condition learning unit 120. Here, the new subject may mean another subject who has not been used to construct the model.

The in-place walking characteristic of a new subject acquired through the in-place walking data collection and in-place walking characteristic analysis process is provided to the health status evaluation unit 130 and input to the health status evaluation model, and the new subject’s health status can be output. have. Specifically, the health condition evaluation unit 130 uses the degree of senility of the new subject using the first health condition evaluation model, the degree of muscle reduction of the new subject using the second health condition evaluation model, or the third health condition evaluation model. Thus, the degree of cognitive impairment of the new subject can be evaluated.

The health state evaluation unit 130 determines the degree of senility of the new subject by dividing it into nonsenile or senility using the first health state evaluation model, or compares the degree of senility of the new subject using the first health state evaluation model. It can be judged by dividing into three categories: old age, full age or old age. The health condition evaluation unit 130 may determine the degree of muscle reduction of the new subject by dividing it into non-myelopathy or sarcopenia using the second health condition evaluation model. In addition, the health state evaluation unit 130 determines the degree of cognitive impairment of the new subject by dividing it into non-cognitive or cognitive impairment using the third health state evaluation model, or by using the third health state evaluation model. The degree of cognitive impairment of the subject can be determined by dividing it into non-cognitive impairment, mild cognitive impairment, or moderate or higher cognitive impairment.

The health condition evaluation unit 130 may provide the result output from each health condition evaluation model to a corresponding subject, and a health condition related to senility, muscle reduction (fall risk), cognitive dysfunction, etc. of the elderly based on the walking characteristics in place. Can provide monitoring of. Accordingly, an opportunity for early diagnosis of related diseases, symptom management, and appropriate treatment can be provided to the subject.

Hereinafter, a method for evaluating a health condition according to an embodiment of the present invention will be described.

6 is a flow chart of a method for evaluating a health condition according to an embodiment of the present invention. The method may be performed in the system of FIGS. 1 to 5 described above, and FIGS. 1 to 5 may be referred to for explanation of this embodiment.

Referring to FIG. 6, the method for evaluating the elderly's health using in-place walking characteristics according to an embodiment of the present invention includes the step of constructing a health state evaluation model based on the in-place walking characteristics (S100), and collecting in-place walking data from the evaluation subject. And extracting the in-place walking characteristic (S110) and inputting the in-place walking characteristic of the evaluation subject into the health state evaluation model to evaluate the health state of the evaluation subject (S120).

A health condition evaluation model based on the in-place walking characteristic is constructed (S100).

First, a plurality of subjects (H) performs a step in place for a predetermined period of time with the sensor 101 mounted on the body part, and the sensor 101 provides step-in-place data to the in-place step data collection unit 100 can do. That is, the sensor 101 plays an instrumental role in evaluating the physical condition of the subject H according to the step in place. The in-place step data collection unit 100 may receive in-place step data from each sensor 101. The sensor 101 may include an accelerometer or gyroscope of a compact size that is easily detachable and attached to the body of the subject H and is portable. In some embodiments, the elderly health condition assessment system 10 may further include a camera 102 and a pressure mat 103, as shown in FIG. 2B. The camera 102 may measure the position and angle of the user's body joints. The pressure mat 103 corresponds to a limited space in which the subject's walking in place is performed. In addition, the pressure mat 103 may measure the user's plantar pressure. The user's in-place walking data (position and angle of the user's body joint) collected through the camera 102 and the in-place walking data (user's plantar pressure) collected through the pressure mat 103 are the in-place walking data collection unit 100 Can be provided on. That is, the in-place step data may include at least one of the measured kinematic data, the measured body joint position and angle of the subject, and the measured plantar pressure of the subject.

Next, the in-place walking data is analyzed to extract in-place walking characteristics.

The in-place step characteristic analysis unit 110 analyzes the in-place step data received from the in-place step data collection unit 100 to extract the in-place step characteristic. Specifically, the in-place walking characteristic analysis unit 110 extracts parameters related to spatiotemporal walking elements, joint angles, and plantar pressure changes. In addition, the in-place walking characteristic analysis unit 110 may acquire images related to changes in spatiotemporal walking elements, joint angles, and plantar pressure. That is, the in-place walking characteristic may include at least one of a spatiotemporal gait factor, a parameter related to a joint angle and plantar pressure change, and an image related to a spatiotemporal gait factor, joint angle, and plantar pressure change. Here, the image may be a spectrogram showing a change in frequency and energy of a signal over time. The in-place walking characteristic analysis unit 110 may convert a time domain signal into a spectrogram through Fourier transform or wavelet transform.

A health state evaluation model based on the in-place walking characteristic is constructed by machine learning the health state of the subject with respect to the in-place walking characteristic.

The health state learning unit 120 constructs a health state evaluation model based on the in-place walking characteristic by machine learning the health state of the subject with respect to the in-place walking characteristic. Here, the in-place walking characteristic may include a spatiotemporal walking element analyzed by the in-place walking characteristic analysis unit 110, a parameter related to a joint angle and plantar pressure change, a spatiotemporal walking element, a joint angle, and an image related to a change in plantar pressure.

The health state learning unit 120 constructs a health state evaluation model having a structure as shown in FIG. 5 by using deep learning, which is a set of machine learning algorithms that attempt a high level of abstraction through a combination of several nonlinear converters. I can. The deep learning model is a convolutional neural network that forms a connection pattern between neurons similar to the structure of an animal's visual cortex, a recursive neural network that builds up a neural network every moment over time, and a restriction on learning a probability distribution for an input set. It may be based on a Boltzmann machine, but is not limited thereto.

An input used for learning in the health state learning unit 120 is a characteristic of a step in place of each of a plurality of subjects, and an output is a state of health of each of the plurality of subjects.

Here, the health state of the subject includes the degree of senility of the subject, the degree of muscle reduction of the subject, and/or the degree of cognitive dysfunction of the subject. The health state evaluation model constructed by the health state learning unit 120 includes a first health state evaluation model that determines the degree of senility of the subject, a second health state evaluation model that determines the degree of muscle reduction of the subject, and the subject's perception. It may include at least one of the third health condition evaluation models for determining the degree of dysfunction.

Next, the in-place walking data is collected from the evaluation target, and in-place walking characteristics are extracted (S110).

At least one of the sensor 101, the camera 102, and the pressure mat 103 may be used to acquire data about the subject's in-place walking. The in-place walking data collection unit 100 receives the in-place walking data of the evaluation subject from at least one of the sensor 101, the camera 102, and the pressure mat 103, and the in-place walking characteristic analysis unit 110 performs the in-place walking By analyzing the data, it is possible to extract the characteristics of walking in place.

The health status of the evaluation target is evaluated (S120).

The in-place walking characteristic acquired from the new subject may be provided to the health condition evaluation unit 130 and provided as an input value of the constructed health condition evaluation model, and the health condition evaluation model may provide the health condition of the new subject as an output value. I can. That is, the health state evaluation unit 130 may evaluate the health state of the new subject based on the characteristics of the new subject's walking in place. Specifically, the health condition evaluation unit 130 uses the degree of senility of the new subject using the first health condition evaluation model, the degree of muscle reduction of the new subject using the second health condition evaluation model, or the third health condition evaluation model. Thus, the degree of cognitive impairment of the new subject can be evaluated. The health condition evaluation unit 130 may provide the results according to each health condition evaluation model to the subject, and based on the in-situ walking characteristics, the health condition related to senility, muscle reduction (fall risk), cognitive dysfunction, etc. Can provide monitoring. Accordingly, an opportunity for early diagnosis of related diseases, symptom management, and appropriate treatment can be provided to the subject.

As described above, the method for evaluating the health status of the elderly according to the present exemplary embodiment can obtain the characteristics of the subject's in-place walking without using the existing expensive gait analysis system in a manner that requires less demand for professional personnel and physical space.

By analyzing the in-place walking characteristics in a simple and automated method and using it as an input of an artificial neural network model, it is possible to minimize the time-consuming and labor-intensive process required for processing input data and generating characteristics.

By monitoring the health conditions related to senility, muscle loss (risk of falls), and cognitive dysfunction in the elderly by using the easy-to-measurable in-place walking characteristics during daily life, it is possible to provide opportunities for early diagnosis of related diseases, symptom management, and appropriate treatment. have.

The operation by the method for evaluating the health condition of the elderly using the walking characteristics in place according to the embodiments described above may be implemented at least partially as a computer program and recorded on a computer-readable recording medium. A program for implementing an operation according to the method for evaluating the health condition of the elderly using the walking characteristics in place according to the embodiments is recorded, and the computer-readable recording medium includes all types of recording devices that store data that can be read by a computer. Includes. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tapes, floppy disks, and optical data storage devices. In addition, the computer-readable recording medium may be distributed over a computer system connected through a network, and computer-readable codes may be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing this embodiment may be easily understood by those of ordinary skill in the art to which this embodiment belongs.

Although described above with reference to examples, the present invention should not be construed as being limited by these examples or drawings, and those skilled in the art will be able to understand the spirit and scope of the present invention described in the following claims. It will be appreciated that various modifications and changes can be made to the present invention without departing from.

10: Elderly health condition evaluation system
100: in-place step data collection unit
110: in-situ walking characteristic analysis unit
120: Department of Health Status Learning
130: health condition evaluation unit
140: database unit

Claims (15)

In-place walking data collection unit for collecting in-place walking data from the subject;
An in-place walking characteristic analysis unit that analyzes the in-place walking data to extract in-place walking characteristics;
A health state learning unit for constructing a health state evaluation model based on the in-place walking characteristic by machine learning the health state of the subject with respect to the in-place walking characteristic; And
Elderly health condition evaluation system using the in-place walking characteristic including a health condition evaluation unit that evaluates the health condition of the new subject by inputting the in-place walking characteristic acquired from a new subject into the health condition evaluation model. The method of claim 1,
The health condition evaluation model includes a first health condition evaluation model for determining the degree of senility of the subject, a second health condition evaluation model for determining the degree of muscle reduction of the subject, and a third health condition for determining the degree of cognitive dysfunction of the subject. Including at least one of the condition evaluation model,
The health condition evaluation unit uses the first health condition evaluation model to determine the degree of senility of the new subject, the muscle reduction degree of the new subject using the second health condition evaluation model, or a new subject using the third health condition evaluation model. Elderly health status evaluation system using the in-situ walking characteristics, characterized in that to evaluate the degree of cognitive dysfunction of the. The method of claim 1,
The health state learning unit is a health state evaluation system for the elderly using in-place walking characteristics to generate the health state evaluation model using deep learning. The method of claim 1,
A sensor attached to the subject's body and measuring kinematic data according to the subject's walking motion;
A camera that measures the position and angle of the subject's body joints; And
As a limited space in which the subject's in-place walking is performed, further comprising a pressure mat for measuring the plantar pressure of the subject,
The in-place walking data includes at least one of the measured kinematic data, the measured body joint position and angle of the subject, and the measured plantar pressure of the subject. The method of claim 4,
The in-place walking characteristic is a system for evaluating the elderly health condition using the in-place walking characteristic including at least one of a spatiotemporal walking factor, a parameter related to a joint angle and plantar pressure change, and an image related to a spatiotemporal walking factor, a joint angle, and plantar pressure change. The method of claim 5,
The image is a local Fourier transform-based spectrogram, an elderly health condition evaluation system using in-place walking characteristics. The method of claim 5,
The image is a health condition evaluation system for the elderly using the in-situ walking characteristic, which is a spectrogram based on continuous wavelet transformation. Collecting in-place walking data from a plurality of subjects, analyzing the in-place walking data to extract in-place walking characteristics, and constructing a health condition evaluation model based on the in-place walking characteristics;
Collecting in-place walking data from the evaluation subject and extracting in-place walking characteristics; And
An elderly health condition evaluation method using the in-place walking characteristic comprising the step of evaluating the health condition of the evaluation subject by inputting the in-place walking characteristic of the evaluation subject into the constructed health condition evaluation model. The method of claim 8,
The health condition evaluation model includes a first health condition evaluation model for determining the degree of senility of the subject, a second health condition evaluation model for determining the degree of muscle reduction of the subject, and a third health condition for determining the degree of cognitive dysfunction of the subject. Including at least one of the condition evaluation model,
Evaluating the health status of the evaluation subject,
The degree of senility of the new subject using the first health condition evaluation model, the degree of muscle reduction of the new subject using the second health condition evaluation model, or the degree of cognitive dysfunction of the new subject using the third health condition evaluation model Elderly health condition evaluation system using the in-situ walking characteristics including evaluating. The method of claim 8,
Building a health condition evaluation model based on the in-place walking characteristic,
A method for evaluating the elderly's health using in-place walking characteristics to construct the health state evaluation model using deep learning. The method of claim 8,
The in-place walking data includes at least one of kinematic data provided from a sensor attached to the subject's body, a position and angle of the subject's body joint provided from a camera, and a plantar pressure of the subject provided from a pressure mat. A method for evaluating the health status of the elderly using gait characteristics The method of claim 11,
The in-place walking characteristic includes at least one of a spatiotemporal gait factor, a parameter related to a joint angle and plantar pressure change, and an image related to a spatio-temporal gait factor, a joint angle, and plantar pressure change. The method of claim 12,
The image is a local Fourier transform-based spectrogram, a method for evaluating the elderly's health status using in-place walking characteristics. The method of claim 12,
The image is a spectrogram based on continuous wavelet transformation, a method for evaluating the elderly's health status using in-place walking characteristics. A computer program stored in a medium to execute a method for evaluating an elderly health condition using the in-place walking characteristic of any one of claims 8 to 14 in combination with hardware.

Images