KR102395937B1 - Health state prediction method and system based on gait time-frequency analysis - Google Patents

Abstract

A method and system for predicting health status based on walking time-frequency analysis are provided. A health state prediction system based on walking time-frequency analysis according to an embodiment of the present invention includes: a walking characteristic signal acquisition unit for receiving a walking characteristic signal according to the subject's gait from at least one sensor attached to the subject's body; a gait characteristic signal converter converting the received gait characteristic signal into a spectrogram using time-frequency analysis; a health state learning unit for machine learning the health state of the subject with respect to the spectrogram to construct a health state model based on the walking characteristic signal; and a health state predictor for predicting the health state of the new subject by inputting the spectrogram obtained from the new subject into the health state model.

Description

Health state prediction method and system based on gait time-frequency analysis

The present invention relates to a method and system for predicting a health state based on a walking time-frequency analysis, and a method for acquiring a user's walking characteristics through simple equipment and predicting a user's health state based on a walking time-frequency analysis and systems.

Gait (gait) is the most basic means of human mobility and is closely related to individual factors (gender, age), health factors, and pathological factors (trauma, neurological/musculoskeletal/psychiatric diseases). Therefore, it is possible to diagnose and monitor the degree of senility, musculoskeletal system and neurological diseases of a subject through quantitative evaluation of gait, to determine the effect of systematic treatment, and to use it for surgical decision making and postoperative review.

In addition, the walking ability may be an index indicating the exercise ability of the subject, and in order to monitor the exercise ability and health status of the subject, a gait analysis is absolutely necessary. In addition, the gait analysis data can be used as basic data for the development and performance improvement of walking aids and walking aids.

In order to analyze and quantify such walking ability, US Patent Publication No. US 2009-0141933 proposes a technique for acquiring an image while walking using a camera and analyzing the movement of the leg to predict motion and predict the gait cycle. . However, this prior art requires equipment such as an expensive camera, an expensive ground reaction force, a pressure sensor, an electromyography sensor, and an experienced experimenter. That is, the walking ability evaluation method using the above-described equipment not only caused a lot of cost as many devices and manpower were input, but also the subject with poor mobility could not perform all of the above-described experiments, so it was difficult to take accurate result data. there was

Accordingly, there is a need for a method and system capable of acquiring characteristics according to the subject's gait with only a simple device and predicting the subject's health status according to the obtained information.

US Patent Publication US 2009-0141933 A1 (June 04, 2009)

The present invention has been devised to solve the above problems, and specifically, a method and system capable of acquiring a characteristic signal according to the subject's gait with only a simple device, and predicting the health status of the subject according to the obtained gait characteristic signal to provide.

A health state prediction system based on walking time-frequency analysis according to an embodiment of the present invention includes: a walking characteristic signal acquisition unit for receiving a walking characteristic signal according to the subject's gait from at least one sensor attached to the subject's body; a gait characteristic signal converter converting the received gait characteristic signal into a spectrogram using time-frequency analysis; a health state learning unit for machine learning the health state of the subject with respect to the spectrogram to construct a health state model based on the walking characteristic signal; and a health state prediction unit for predicting the health state of the new subject by inputting the spectrogram obtained from the new subject into the health state model.

In a health state prediction method based on walking time-frequency analysis according to an embodiment of the present invention, a gait characteristic signal according to gait from a plurality of subjects is received from at least one sensor attached to the subject's body, and the received gait characteristic converting a signal into a spectrogram using time-frequency analysis, machine learning the health state of the subject with respect to the spectrogram, and constructing a health state model based on the gait characteristic signal; obtaining a spectrogram from the subject to be evaluated; and predicting the health status of the subject to be evaluated by inputting the spectrogram of the subject to be evaluated into the constructed health state model.

A computer program according to an embodiment of the present invention is stored in a medium in combination with hardware to execute the method for predicting a health state based on the walking time-frequency analysis.

The health state prediction method and system according to the present embodiment can acquire characteristics according to the subject's gait using only a simple device such as an inertial measurement sensor, and convert the acquired gait characteristic signal into a spectrogram to obtain a health state based on the gait characteristic signal It is used for model building.

In addition, the health state prediction method and system according to the present embodiment can easily predict the health state of a new subject based on the walking characteristics of the new subject acquired with a simple device through the constructed health state model.

1 is a block diagram of a health state prediction system based on walking time-frequency analysis according to an embodiment of the present invention.
2 is a diagram exemplarily illustrating a subject participating in a health state prediction system according to an embodiment of the present invention and a sensor attached to the subject.
3A is an exemplary graph of a gait characteristic signal provided by a sensor.
3B and 3C are exemplary diagrams of spectrograms obtained by converting a gait characteristic signal using time-frequency analysis.
4A is an exemplary image in which a spectrogram is merged with another spectrogram.
4B is an exemplary image in which the image size of the spectrogram is adjusted and the adjusted images are merged.
5 is a diagram illustrating an exemplary machine learning model of a health state learning unit of a health state prediction system according to an embodiment of the present invention.
6 is a graph illustrating an analysis result according to the exemplary model of FIG. 5 .
7 is a flowchart of a method for predicting a health state based on walking time-frequency analysis according to an embodiment of the present invention.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. DETAILED DESCRIPTION The detailed description set forth below in conjunction with the appended 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 in order to provide a thorough understanding of the present invention. However, one skilled in the art will recognize that the present invention may be practiced without these specific details. Specific terms used in the following description are provided to help the understanding of the present invention, and the use of these specific terms may be changed to other forms without departing from the technical spirit of the present invention.

1 is a block diagram of a health state prediction system based on walking time-frequency analysis according to an embodiment of the present invention. 2 is a diagram exemplarily illustrating a subject participating in a health state prediction system according to an embodiment of the present invention and a sensor attached to the subject. 3A is an exemplary graph of a gait characteristic signal provided from a sensor, and FIGS. 3B and 3C are exemplary diagrams of a spectrogram obtained by converting the gait characteristic signal using time-frequency analysis. 4A is an exemplary image in which a spectrogram and another spectrogram are merged. 4B is an exemplary image in which the image size of the spectrogram is adjusted and the adjusted images are merged.

1 to 4B , the health state prediction system 10 includes a walking characteristic signal acquisition unit 100 , a walking characteristic signal conversion unit 110 , a health state learning unit 120 , and a health state prediction unit 130 . and a database unit 140 .

A health state prediction system according to embodiments may be entirely hardware, or may have aspects that are partly hardware and partly software. For example, the health state prediction system of the present specification and each unit included therein may collectively refer to a device for exchanging data in a specific format and content in an electronic communication method and software related thereto. As used herein, terms such as “unit”, “module”, “server”, “system”, “device” or “terminal” refer to a combination of hardware and software driven by the hardware. it is intended to be For example, the hardware herein may be a data processing device including a CPU or other processor. In addition, software driven by hardware may refer to a running process, an object, an executable file, a thread of execution, a program, and the like.

In addition, each part constituting the health state prediction system is not intended to necessarily refer to physically distinct separate components. In FIG. 1 , the walking characteristic signal acquisition unit 100 , the walking characteristic signal conversion unit 110 , the health state learning unit 120 , the health state prediction unit 130 , and the database unit 140 are separate blocks separated from each other. However, this is only functionally dividing the devices constituting the health state prediction system by the operations executed by the corresponding devices. Accordingly, depending on the embodiment, the walking characteristic signal acquisition unit 100, the walking characteristic signal conversion unit 110, the health state learning unit 120, the health state prediction unit 130, and the database unit 140 are partially or All of them may be integrated in the same single device, one or more may be implemented as separate devices physically separated from other units, or components may be communicatively connected to each other in a distributed computing environment.

The gait characteristic signal acquisition unit 100 receives a gait characteristic signal according to the subject's gait from at least one sensor attached to the subject's body. As shown in FIG. 2 , the subject H participating in the health state prediction system 10 is equipped with a sensor S on the body. The body part of the subject H to which the sensor S is mounted may be a body part having a direct relationship to the walking of the subject H. That is, the body part of the subject H on which the sensor S is mounted may provide not only the driving force of walking, but also a function of maintaining the walking by providing dynamic safety during walking. Illustratively, the body part of the subject H to which the sensor S is mounted is the pelvis (Pelvis), thigh (Right Thigh, Left Thigh), shin (Right Shank, Left Shank) and foot (Right) of the subject (H). Foot, Left Foot), but is not limited thereto. As described above, gait (gait) is the most basic means of human movement, and characteristics such as gender, age, health factor, pathological factor, and exercise ability may be reflected and appear, and these characteristics Accordingly, the walking characteristic signal collected by the sensor S may appear differently.

The subject H walks a predetermined distance with the sensor S mounted on a body part, and the sensor S may provide a signal that changes according to the walking to the walking characteristic signal acquisition unit 100 . That is, the sensor S performs a tool role in evaluating the physical state of the subject H according to walking. The gait characteristic signal acquisition unit 100 may receive a gait characteristic signal that changes according to the gait of the subject H from each sensor S. The walking characteristic signal acquisition unit 100 and the sensor S may be configured to wirelessly transmit and receive signals.

Here, the sensor (S) may be configured in a small size to be easily removable, attachable, and portable to the body of the subject (H). The sensor S may include an inertial measurement unit (IMU). The inertial measurement sensor (IMU) includes a gyroscope, an accelerometer, and a geomagnetic sensor that measures free movement in a three-dimensional space, and can measure the speed, direction, and acceleration of the subject H. That is, the inertial measurement sensor senses acceleration information and angular velocity information that are changed according to the walking of the subject H at a position attached to the body of the subject H, and provides the walking characteristic signal acquisition unit 100 as a walking characteristic signal. can

The walking characteristic signal may include acceleration information for each of the X-axis, Y-axis, and Z-axis and angular velocity information for each of the X-axis, Y-axis, and Z-axis provided from the inertial measurement sensor (IMU). Here, the X-axis and the Y-axis define a vertical plane perpendicular to the floor surface on which the subject H is located, and the Z-axis may be defined as an axis perpendicular to the vertical plane, but is not limited thereto.

The walking characteristic signal conversion unit 110 may convert the walking characteristic signal received by the walking characteristic signal acquisition unit 100 into a spectrogram using time-frequency analysis. The walking characteristic signal provided from the inertial measurement sensor (IMU) may be a time domain signal, data showing that the amplitude of the signal changes with time, as exemplarily shown in FIG. 3A . 3A is a graph showing angular velocity information for each of the X-axis, Y-axis, and Z-axis.

The walking characteristic signal converter 110 may convert the amplitude data according to time into a time-frequency (TF) domain signal through time-frequency analysis, and the time-frequency (TF) domain signal is It can be created like a single image. A spectrogram is an image converted to intuitively grasp a wave from the human eye, and combines the characteristics of a waveform and a spectrum. Accordingly, the time-frequency (TF) domain signal may be input data suitable for deep learning than the time domain signal. The spectrogram represents a spectral sequence having time on one axis and frequency on the other axis, and intensity of frequency components may be classified by color and displayed together.

The walking characteristic signal converter 110 may convert the time domain signal into a spectrogram through a Fourier transform or a wavelet transform. Exemplarily, the walking characteristic signal converter 110 may generate a spectrogram through a local Fourier transform (Short-time Fourier transform, STFT), but is not limited thereto, and a Stockwell transform or a continuous way A continuous wavelet transform (CWT) may also be applied. Figure 3b shows an exemplary spectrogram generated through a local Fourier transform (Short-time Fourier transform, STFT), Figure 3c is a spectrogram generated through a continuous wavelet transform (Continuous wavelet transform, CWT) It is shown by way of example. Here, the gait characteristic signal may be converted into a spectrogram of a smoother image through continuous wavelet transform (CWT). The spectrogram of such a smooth image is more suitable for deep learning because it can prevent the loss of frequency and temporal information that occurs depending on the temporal resolution setting in the local Short-time Fourier transform (STFT). It may be more suitable input data.

The walking characteristic signal conversion unit 110 converts acceleration information for each of the X-axis, Y-axis, and Z-axis provided from the inertial measurement sensor (IMU) and angular velocity information for each of the X-axis, Y-axis, and Z-axis into a spectrogram, respectively. can be converted Exemplarily, the angular velocity signal in the time domain for each of the X-axis, Y-axis, and Z-axis shown in FIG. 3A is an X-axis angular velocity spectrogram, Y-axis angular velocity spectrogram, and Z-axis angular velocity spectrogram as shown in FIG. 3B . each can be converted into grams.

Also, the walking characteristic signal conversion unit 110 may adjust the spectrogram so that the converted spectrogram may be input to the health state learning unit 120 to be described later. Here, the adjustment of the spectrogram may be a process of adjusting the image size of the spectrogram to fit the size of input data of the health state model built by the health state learning unit 120 or merging the image with other spectrograms. 4A is an exemplary image merged with another spectrogram, and FIG. 4B is an exemplary image obtained by adjusting the image size of the spectrogram and merging the adjusted image.

As shown in Figure 4a, the size of the acceleration spectrogram and the angular velocity spectrogram along each axis (X-axis, Y-axis, or Z-axis) is adjusted to be the same, and the image adjusted to fit the size of the input data of the health state model is merged. can In addition, the acceleration spectrogram and the angular velocity spectrogram along each axis (X-axis, Y-axis, or Z-axis) are merged, placed in a direction parallel to the same plane, and as shown in FIG. 4B, an input data image having a certain size and resolution can create Here, the input data image may be generated corresponding to each sensor S. That is, each signal generated by one sensor S is converted into a spectrogram, and may be merged into one image to generate an input data image of a predetermined size.

The health state learning unit 120 machine-learns the health state of the subject with respect to the spectrogram to construct a health state model based on the gait characteristic signal.

The health state learning unit 120 can learn an abstract model that sets a spectrogram obtained from a plurality of subjects as an input value, sets the health state of each of the plurality of subjects as an output value, and determines a relationship between them. there is.

The health state learning unit 120 uses deep learning to perform machine learning so that the human brain discovers patterns in numerous data and then imitates the information processing method of classifying objects so that the computer can distinguish objects through deep learning. It is possible to build a health state model based on the gait characteristic signal. The health state learning unit 120 may build an abstract model that determines a relationship between the input spectrogram and the health state of the subject provided as a result. The health state learning unit 120 may include a machine learning model based on a convolutional neural network, and a deep neural network model in which multiple hidden layers exist between an input layer and an output layer; A convolutional neural network model that forms a connection pattern between neurons similar to the structure of an animal's visual cortex, a recurrent neural network model that builds up a neural network at every moment over time, and a probability distribution for an input set. You can use a deep learning model of one of the available restricted Boltzmann machines. However, the above-described method is only an example, and the machine learning method according to an embodiment of the present invention is not limited thereto.

The health state learner 120 may perform machine learning by using the spectrogram generated by the walking characteristic signal converter 110 as an input value and using the health state of the subject as an output value. Here, the health state of the subject may include current state information on the pathological element of the subject. Here, the pathological element of the subject may include information on trauma, neurological diseases, musculoskeletal diseases, and psychiatric diseases. For example, the pathological element may include information on the subject's current state regarding at least one of deformity, muscle weakness, sensory loss, pain, and movement control impairment. . The pathological component of the subject may consist of numerical values for the presence and severity of the disease, but is not limited thereto.

In some embodiments, the subject's health status may also include an assessment of the subject's overall athletic ability. That is, it may also include evaluation of strength, agility, and endurance that can define basic athletic ability.

Also, in some embodiments, the health state model may be generated in response to a sensor S attached to a body part of the subject. Illustratively, the health state model corresponding to each of the pelvis (Pelvis), thigh (Right Thigh, Left Thigh), shin (Right Shank, Left Shank) and foot (Right Foot, Left Foot) of the subject (H) is individually can be created However, the present invention is not limited thereto, and the input value input when constructing the health state model may be data obtained by merging signals generated from each body part or signal-converted spectrograms, and the health state model is attached to a plurality of body parts. It may be built based on a plurality of sensors.

In addition, in some embodiments, the health state model may be constructed differently according to gender or differently according to age in consideration of a subject's unique factors (gender, age).

Through this machine learning process, the health state learner 120 may build a health state model based on the walking characteristic signal. The constructed health state model may be stored in the database unit 140 . The database unit 140 may be configured to temporarily store or store basic data such as a health state model and a spectrogram and/or a gait characteristic signal based thereon.

The health state prediction unit 130 may predict the health state of a new subject by using the health state model built by the health state learning unit 120 . Here, the new subject may refer to another subject not used in the construction of the model. A new subject also attaches the sensor (S) to a part of his or her body and walks a certain distance. The gait characteristic signal acquisition unit 100 may receive a gait characteristic signal according to the gait of a new subject from the sensor S, and the gait characteristic signal conversion unit 110 may convert the gait characteristic signal into a spectrogram. The transformed spectrogram of the new subject may be provided to the health state prediction unit 130 and provided as an input value of the health state model, and the health state model may provide the health state of the new subject as an output value. That is, the health state prediction unit 130 may provide prediction data for the new subject's pathological factors and exercise ability based on the spectrogram of the new subject.

5 is a diagram illustrating an exemplary machine learning model structure of a health state learning unit of a health state prediction system according to an embodiment of the present invention.

A machine learning model based on a convolutional neural network consists of a plurality of convolution steps and a plurality of pooling steps. It can learn automatically, and image analysis and discrimination can be performed according to the learned filter. The output value in the exemplary model of FIG. 5 is to determine whether the subject's walking ability is normal (Normal foot, Abnormal foot) and whether or not an athlete has different athletic ability from the general public (Athlete foot). According to an output value and a result to be determined, a plurality of convolution steps and a plurality of pooling steps may be configured differently from FIG. 5 .

In addition, the machine learning model according to FIG. 5 may be built based on sensors mounted on various body parts. i) Feet, ii) Thigh, iii) Shank, iv) Pelvis and Feet, v) Pelvis and Thigh, vi) Pelvis and Pelvis and Shank) and vii) built a machine learning model based on sensors mounted on the pelvis, feet, thighs and shins (Full sensors). 6 is a graph illustrating an analysis result according to the exemplary model of FIG. 5 . Specifically, FIG. 6 is a graph showing the accuracy according to the attachment position of the sensor, which is the basis for the construction of such a health state model. After the model is built, the spectrogram obtained from the new subjects is input to the model, and the output value of the model for the new subjects and the actual subject's state (Normal foot, Abnormal foot, Athlete foot) are compared with the verification process is performed. The higher the accuracy (Accuracy, %) shown in FIG. 6, the more the output value of the model and the actual state of the subject are the same result.

As shown in FIG. 6 , it can be confirmed that the accuracy increases when using a plurality of sensors attached to different locations rather than building a model using a single sensor, and when using all sensors (Full sensors), the highest It can be seen that the accuracy is indicated. In addition, even if a single sensor is used, it can be seen that when a health prediction model is built based on the sensor mounted on the foot, which is most directly related to walking, it shows higher accuracy than other body parts.

As such, the health state prediction system according to an embodiment of the present invention can acquire characteristics according to the subject's gait using only a simple device such as an inertial measurement sensor, and converts the acquired gait characteristic signal into a spectrogram to add the gait characteristic signal to the gait characteristic signal. It is used to build a health condition prediction model based on it.

In addition, the health state prediction system according to an embodiment of the present invention can easily predict the health state of a new subject based on the walking characteristics of the new subject acquired with a simple device through the constructed health state model.

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

7 is a flowchart of a method for predicting a health state according to an embodiment of the present invention. The method may be performed in the system of FIGS. 1 to 4B described above, and FIGS. 1 to 4B may be referred to for explanation in this embodiment.

Referring to FIG. 7 , the health state prediction method based on walking time-frequency analysis according to an embodiment of the present invention includes the steps of constructing a health state model based on a gait characteristic signal (S100), and obtaining a spectrogram from the subject to be evaluated. It includes a step (S110) and a step (S120) of predicting the health status of the subject to be evaluated.

A health state model based on the gait characteristic signal is constructed (S100).

First, a gait characteristic signal according to gait from a plurality of subjects is received from at least one sensor attached to the subject's body.

The subject H walks a predetermined distance with the sensor S mounted on a body part, and the sensor S may provide a signal that changes according to the walking to the walking characteristic signal acquisition unit 100 . That is, the sensor S performs a tool role in evaluating the physical state of the subject H according to walking. The gait characteristic signal acquisition unit 100 may receive a gait characteristic signal that changes according to the gait of the subject H from each sensor S. The sensor (S) may be of a small size so as to be easily removable, attachable, and portable to the body of the subject (H). The sensor S may include an inertial measurement sensor IMU. The inertial measurement sensor includes a gyroscope, an accelerometer, and a geomagnetic sensor that measures free movement in a three-dimensional space, and can measure the speed, direction, and acceleration of the subject H. The walking characteristic signal may include acceleration information for each of the X-axis, Y-axis, and Z-axis and angular velocity information for each of the X-axis, Y-axis, and Z-axis provided from the inertial measurement sensor (IMU).

Next, the received gait characteristic signal is converted into a spectrogram using time-frequency analysis.

The walking characteristic signal conversion unit 110 may convert the walking characteristic signal received by the walking characteristic signal acquisition unit 100 into a spectrogram using time-frequency analysis. The walking characteristic signal conversion unit 110 may convert the amplitude data according to time into a time-frequency (TF) domain signal through time-frequency analysis, and this time-frequency (TF) domain signal is a single image and can be created together. The walking characteristic signal converter 110 may generate a spectrogram through a local Fourier transform (Short-time Fourier transform, STFT), but is not limited thereto, and a Stockwell transform or a continuous wavelet transform (Continuous). wavelet transform (CWT) may also be applied. Here, the gait characteristic signal may be converted into a spectrogram of a smoother image through continuous wavelet transform (CWT). The spectrogram of such a smooth image is more suitable for deep learning because it can prevent the loss of frequency and temporal information that occurs depending on the temporal resolution setting in the local Short-time Fourier transform (STFT). It may be more suitable input data.

The walking characteristic signal conversion unit 110 converts acceleration information for each of the X-axis, Y-axis, and Z-axis provided from the inertial measurement sensor (IMU) and angular velocity information for each of the X-axis, Y-axis, and Z-axis into a spectrogram, respectively. can be converted Also, the walking characteristic signal conversion unit 110 may adjust the spectrogram so that the converted spectrogram may be input to the health state learning unit 120 to be described later. Here, the adjustment of the spectrogram may be a process of adjusting the image size of the spectrogram to fit the size of input data of the health state model built by the health state learning unit 120 or merging the image with other spectrograms.

By machine learning the health state of the subject with respect to the spectrogram, a health state model based on the walking characteristic signal is constructed.

Building a health state model based on the walking characteristic signal may include performing machine learning by using the spectrogram as an input value and using the health state of the subject as an output value. In addition, the step of building a health state model based on the gait characteristic signal includes a convolutional neural network-based machine learning model comprising a plurality of convolution steps and a plurality of pooling steps. can be built

The health state learning unit 120 can learn an abstract model that sets a spectrogram obtained from a plurality of subjects as an input value, sets the health state of each of the plurality of subjects as an output value, and determines a relationship between them. there is. The health state learner 120 may perform machine learning by using the spectrogram generated by the walking characteristic signal converter 110 as an input value and using the health state of the subject as an output value. Here, the health state of the subject may include current state information on the pathological element of the subject. Here, the pathological element of the subject may include information on trauma, neurological diseases, musculoskeletal diseases, and psychiatric diseases. For example, the pathological element may include information on the subject's current state regarding at least one of deformity, muscle weakness, sensory loss, pain, and movement control impairment. . The pathological component of the subject may consist of numerical values for the presence and severity of the disease, but is not limited thereto.

Next, a spectrogram is obtained from the subject to be evaluated (S110).

The subject to be evaluated also attaches the sensor (S) to a part of the person's body and walks a certain distance. The gait characteristic signal acquisition unit 100 may receive a gait characteristic signal according to the gait of the subject to be evaluated from the sensor S, and the gait characteristic signal conversion unit 110 may convert the gait characteristic signal into a spectrogram.

The health status of the subject to be evaluated is predicted (S120).

The transformed spectrogram of the new subject may be provided to the health state prediction unit 130 and provided as an input value of the constructed health state model, and the health state model may provide the health state of the new subject as an output value. That is, the health state prediction unit 130 may provide prediction data for the new subject's pathological factors and exercise ability based on the spectrogram of the new subject.

As such, the health state prediction method according to an embodiment of the present invention can acquire characteristics according to the subject's gait using only a simple device such as an inertial measurement sensor, and converts the acquired gait characteristic signal into a spectrogram to obtain the gait characteristic signal. It is used to build a health condition model based on it.

In addition, the health state prediction method according to an embodiment of the present invention can easily predict the health state of a new subject based on the walking characteristics of the new subject acquired with a simple device through the constructed health state model.

The operation of the method for predicting a health state based on walking time-frequency analysis according to the embodiments described above may be at least partially implemented as a computer program and recorded in a computer-readable recording medium. A program for implementing an operation by a health state prediction method based on walking time-frequency analysis according to embodiments is recorded and a computer-readable recording medium is any type of recording device in which computer-readable data is stored includes Examples of the computer-readable recording medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. In addition, the computer-readable recording medium may be distributed in a network-connected computer system, and the computer-readable code may be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the present embodiment may be easily understood by those skilled in the art to which the present embodiment belongs.

Although the above has been described with reference to the embodiments, the present invention should not be construed as being limited by these embodiments or drawings, and those skilled in the art will appreciate the spirit and scope of the present invention described in the claims below. It will be understood that various modifications and variations of the present invention can be made without departing from the scope of the present invention.

10: Health state prediction system based on walking time-frequency analysis
100: walking characteristic signal acquisition unit
110: walking characteristic signal conversion unit
120: Health State Learning Department
130: health state prediction unit
140: database unit

Claims (15)

a gait characteristic signal acquisition unit for receiving a gait characteristic signal according to the subject's gait from at least one or more inertial measurement sensors attached to the lower body including the subject's feet;
a gait characteristic signal converter converting the received gait characteristic signal into a spectrogram using time-frequency analysis;
a health state learning unit configured to construct a health state model based on the walking characteristic signal by performing machine learning using the spectrogram as an input value and using the health state and exercise ability of the subject as output values; and
A health state prediction unit for predicting the health state of the new subject by inputting the spectrogram obtained from the new subject into the health state model,
The time-frequency analysis is a continuous wavelet transform (CWT), and the spectrogram is generated as a smooth image through continuous wavelet transform,
The walking characteristic signal includes acceleration information for each axis and angular velocity information for each axis provided from the inertial measurement sensor,
The acceleration information for each axis is converted into an acceleration spectrogram for each axis, and the angular velocity information for each axis is converted into an angular velocity spectrogram for each axis,
The acceleration spectrogram for each axis and the angular velocity spectrogram for each axis are merged after each size is adjusted corresponding to the size of the input value,
The subject's health status includes current status information about the subject's pathological component,
The exercise ability is a walking time-frequency analysis-based health state prediction system, characterized in that it includes information on the subject's muscle strength, agility, endurance, and whether an athlete has an athletic ability different from that of the general public (Athlete foot). delete delete delete According to claim 1,
The health state learning unit generates the health state model through a convolutional neural network-based machine learning model consisting of a plurality of convolution steps and a plurality of pooling steps - a walking time-frequency analysis based on health State Prediction System. delete delete A health state prediction method based on walking time-frequency analysis performed in a health state prediction system based on walking time-frequency analysis, the method comprising: a gait characteristic signal according to gait from a plurality of subjects to the lower body including the subject's feet Receives from at least one attached inertial measurement sensor, converts the received gait characteristic signal into a spectrogram using time-frequency analysis, uses the spectrogram as an input value, and sets the health status of the subject as an output value constructing a health state model based on the gait characteristic signal by performing machine learning using
obtaining a spectrogram from the subject to be evaluated; and
and inputting the spectrogram of the subject to be evaluated into the built-up health state model to predict the health state and exercise ability of the subject to be evaluated,
The time-frequency analysis is a continuous wavelet transform (CWT), and the spectrogram is generated as a smooth image through continuous wavelet transform,
The acceleration information for each axis is converted into an acceleration spectrogram for each axis, and the angular velocity information for each axis is converted into an angular velocity spectrogram for each axis,
The acceleration spectrogram for each axis and the angular velocity spectrogram for each axis are merged after each size is adjusted corresponding to the size of the input value,
The subject's health status includes current status information about the subject's pathological component,
The exercise ability is a walking time-frequency analysis-based health state prediction method, characterized in that it includes information about the subject's muscle strength, agility, endurance, and whether an athlete having an athletic ability different from that of the general public (Athlete foot). delete delete delete 9. The method of claim 8,
The step of constructing a health state model based on the walking characteristic signal comprises:
A health state prediction method based on walking time-frequency analysis for constructing the health state model through a machine learning model based on a convolutional neural network comprising a plurality of convolution steps and a plurality of pooling steps. delete delete A computer program stored on a medium in combination with hardware to execute the method for predicting health conditions based on walking time-frequency analysis according to claim 8 or 12.

Images