WO2025041230A1 - Dispositif d'extraction de risques de chute, procédé d'extraction de risques de chute, et programme d'extraction de risques de chute - Google Patents

Dispositif d'extraction de risques de chute, procédé d'extraction de risques de chute, et programme d'extraction de risques de chute Download PDF

Info

Publication number
WO2025041230A1
WO2025041230A1 PCT/JP2023/030025 JP2023030025W WO2025041230A1 WO 2025041230 A1 WO2025041230 A1 WO 2025041230A1 JP 2023030025 W JP2023030025 W JP 2023030025W WO 2025041230 A1 WO2025041230 A1 WO 2025041230A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
subject
fall risk
risk
gait
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/JP2023/030025
Other languages
English (en)
Japanese (ja)
Inventor
貴志 山本
浩平 望月
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Mitsubishi Electric Corp
Original Assignee
Mitsubishi Electric Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Mitsubishi Electric Corp filed Critical Mitsubishi Electric Corp
Priority to PCT/JP2023/030025 priority Critical patent/WO2025041230A1/fr
Priority to JP2024558135A priority patent/JP7668970B1/ja
Publication of WO2025041230A1 publication Critical patent/WO2025041230A1/fr
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb

Definitions

  • This disclosure relates to a fall risk extraction device, a fall risk extraction method, and a fall risk extraction program.
  • Patent Document 1 describes a factor estimation system that acquires walking data from a subject, detects walking characteristics such as changes in the position or angle of the subject's joints, and analyzes the walking characteristics for each joint individually to extract the subject's risk of falling.
  • Patent Document 2 also describes a hemiplegia testing device that performs frequency analysis on the walking characteristics for each joint and uses the frequency analysis results to extract walking characteristics for hemiplegia, etc.
  • Patent Documents 1 and 2 extract the risk of falling without considering the correlation between movements in multiple parts of the body. This means that, for example, it is not possible to evaluate the risk of falling related to the interrelationship between movements in multiple parts of the body, such as the movement of the left and right legs, and there is a problem in that it is not possible to carry out appropriate follow-up observations during rehabilitation, etc.
  • the present disclosure aims to provide a fall risk extraction device, a fall risk extraction method, and a fall risk extraction program that extract a subject's risk of falling by considering the correlation between movements in multiple parts of the subject's body.
  • the fall risk extraction device disclosed herein is characterized by comprising a skeletal information extraction unit that extracts skeletal information of a subject from a video of the subject walking, a joint information extraction unit that extracts information from the skeletal information including each of the time-series position changes, angle changes, speed changes, and angular velocity changes in each joint while the subject is walking, a gait feature calculation unit that calculates gait features including correlations between the movements of multiple joints based on the information extracted by the joint information extraction unit, and a fall risk extraction unit that determines the risk of the subject falling based on the gait features including correlations between the movements of the multiple joints.
  • the device disclosed herein extracts the risk of falling by considering the correlation of joint features in walking movements, making it possible to easily and accurately monitor the progress of the coordination of movements of multiple body parts during rehabilitation of subjects at risk of falling.
  • FIG. 13A is an example of time series data of initial gait characteristics of a subject who is not at risk of falling
  • FIG. 13B is an example of time series data of initial gait characteristics of a subject who is at risk of falling.
  • 13A is an example of time series data of gait features after additive processing of a subject who is not at risk of falling
  • FIG. 13B is an example of time series data of gait features after additive processing of a subject who is at risk of falling.
  • FIG. 13A is an example of the result of frequency analysis of data from a subject who is not at risk of falling
  • FIG. 13B is an example of the result of frequency analysis of data from a subject who is at risk of falling
  • FIG. 11 is a functional configuration diagram showing a fall risk extraction device according to a second embodiment.
  • 13 is a flowchart showing an example of the operation of the fall risk extraction device according to the second embodiment.
  • 13A is an example of calculation of the double leg support period for a subject who is not at risk of falling
  • FIG. 13B is an example of calculation of the double leg support period for a subject who is at risk of falling.
  • First Embodiment 1 is a functional configuration diagram showing a fall risk extraction device 20 according to embodiment 1.
  • the fall risk extraction device 20 is connected to a camera 10 that captures an image of a walking subject, and a display unit 30 that displays the value of the fall risk extracted by the fall risk extraction device 20 to the subject, etc.
  • the camera 10 is an imaging device that captures video (moving images) of the subject.
  • the video captured by the camera 10 is input to the skeletal information extraction unit 22 via the video input unit 21 of the fall risk extraction device 20.
  • the video input unit 21 is an input/output interface 230, which will be described later, and is specifically a Universal Serial Bus (USB) terminal, an IEEE 1394 terminal, a Thunderbolt terminal, or the like.
  • USB Universal Serial Bus
  • the skeletal information extraction unit 22 extracts skeletal information of the subject's body from the input video.
  • software capable of detecting the joints of the human body from video such as "Openpose,” is used.
  • the skeletal information extraction unit 22 extracts the three-axis coordinates of the joints of the human body from the input video as skeletal information.
  • Examples of the joints extracted by the skeletal information extraction unit 22 include the skeletal joints of the head and neck, the skeletal joints of the chest, abdomen, and waist, the skeletal joints of the left and right shoulders, the skeletal joints of the left and right elbows, the skeletal joints of the left and right wrists, the skeletal joints of the left and right hands, the skeletal joints of the left and right hip joints, the skeletal joints of the left and right knees, the skeletal joints of the left and right ankles, and the skeletal joints of the left and right feet.
  • the multiple joint information preprocessing unit 23 performs preprocessing on the skeletal information extracted by the skeletal information extraction unit 22. Specifically, it calculates the walking features of multiple joints (hereinafter referred to as "initial walking features") from the skeletal information extracted by the skeletal information extraction unit 22.
  • the initial walking features are, for example, time series data of the position or angle of the skeletal joints, and data obtained by differentiating the time series data of the position or angle of the skeletal joints.
  • the differential value of the position data of the skeletal joints is velocity
  • the differential value of the angle of the skeletal joints is angular velocity.
  • the time series data of the position of the skeletal joints is the coordinate value of one axis, two axes, or three axes of the skeletal joints when time has passed during walking.
  • three axes as an example, a point in the shooting environment or a fixed point in the screen of the image is set as the origin, and the x-axis and y-axis are set horizontally from the origin, and the z-axis is set vertically.
  • the x-axis is set in the east-west direction
  • the y-axis is set in the north-south direction.
  • the angle of a skeletal joint is a quantity that represents the distance between two line segments with the same endpoint when there are line segments connecting a joint to any two other joints.
  • the time series data for the angle represents the angle value over time during walking.
  • the multiple joint information preprocessing unit 23 adds or complex-expresses the calculated multiple initial gait features between joints.
  • Adding means adding the values of the initial gait features of multiple joints at the same time in the initial gait features that are the time series data of multiple joints, and extracting time series data consisting of the added values at each time (hereinafter referred to as "gait features after addition processing").
  • Complex expression means expressing the initial gait features that are the time series data of two different joints as a complex number with the value of the initial gait feature of one joint as the real part and the value of the initial gait feature of the other joint as the imaginary part at each time, and extracting time series data consisting of complex numbers at each time (hereinafter referred to as "gait features after complex expression").
  • the time series data consisting of complex numbers at each time is displayed in a coordinate system (complex plane) of imaginary and real parts, and the complex plane can be considered to be a collection of vectors that rotate at an angular frequency related to the passage of time.
  • the post-addition processing gait features and the post-complex expression gait features are collectively referred to as "post-calculation processing gait features.”
  • the post-calculation processing gait features extracted by adding or complex expressing the initial gait features are time-series data of the subject's walking.
  • both the initial gait features and the post-calculation processing gait features are walking features, and are one of the features of the walking movement.
  • FIG. 4(a) is an example of time series data of initial gait features at the positions of both ankles of a subject who is not at risk of falling.
  • FIG. 4(b) is an example of time series data of initial gait features at the positions of both ankles of a subject who is at risk of falling.
  • FIG. 5(a) is an example of time series data of gait features after additive processing extracted by adding together the time series data of the positions of both ankles shown in FIG. 4(a) of a subject who is not at risk of falling.
  • FIG. 5(b) is an example of time series data of gait features after additive processing extracted by adding together the time series data of the positions of both ankles shown in FIG. 4(b) of a subject who is at risk of falling.
  • the changes in angle, speed, or angular velocity of both ankles may be additively processed or expressed in complex terms.
  • the frequency analysis unit 24 outputs the result of frequency analysis of the post-arithmetic processing gait features, which are information preprocessed by the multiple joint information preprocessing unit 23 (hereinafter referred to as "post-frequency analysis gait features"). Specifically, the frequency analysis unit 24 performs a Fourier transform on the post-addition processing gait features. The frequency analysis unit 24 also performs an inverse Fourier transform on the complex representation gait features. As described above, the time series data consisting of complex numbers at each time represents a collection of vectors that rotate on the complex plane at an angular frequency that changes in time, and can therefore be considered as a function of angular frequency.
  • the real time series data calculated by the inverse Fourier transform corresponds to a frequency analysis of a trajectory in complex space, and as a result, it is possible to calculate the frequency characteristics of the movements of the two joints.
  • Figure 6(a) is an example of gait features after frequency analysis, which is the result of frequency analysis of gait features after additive processing, which is time series data extracted by adding together time series data of the joint positions of a subject who is at risk of falling.
  • Figure 6(b) is an example of gait features after frequency analysis, which is the result of frequency analysis of gait features after additive processing, which is time series data extracted by adding together time series data of the joint positions of a subject who is at risk of falling. Comparing Figure 6(a) and Figure 6(b), it can be seen that there are differences in the frequency analysis results. For example, as shown in Figure 6(a), the amplitude of the subject who is not at risk of falling is larger than the amplitude of the subject who is at risk of falling, as shown in Figure 6(b).
  • the fall risk assessment unit 27 extracts the subject's risk of falling based on the post-frequency analysis gait features output from the frequency analysis unit 24 and the data stored in the memory unit 26.
  • the subject's risk of falling is an index whose value ranges from a minimum of 0 to a maximum of 100, with larger values indicating a higher risk of falling and smaller values indicating a lower risk of falling.
  • the memory unit 26 stores parameters of a calculation formula for extracting the risk of falling from the post-frequency analysis gait features.
  • the display unit 30 also displays the value of the risk of falling extracted by the fall risk assessment unit 27 to the subject, etc. The value of the risk of falling can be displayed, for example, on a display for a few seconds after the subject walks.
  • FIG. 2 is a hardware configuration diagram showing a fall risk extraction device 20 relating to embodiment 1.
  • the fall risk extraction device 20 is composed of a computer including a processor 210, a storage device 220, and an input/output interface 230.
  • the fall risk extraction device 20 may be composed of multiple computers.
  • the processor 210 is an integrated circuit (IC) that performs arithmetic processing. Specific examples of the processor 210 include a central processing unit (CPU), a digital signal processor (DSP), and a graphics processing unit (GPU).
  • the processor 210 functions as a skeletal information extraction unit 22 by running software capable of detecting joints of the human body from the aforementioned "Open pose" video, and as a multiple joint information preprocessing unit 23 by running software that adds or expresses initial walking features in complex form between joints, as a frequency analysis unit 24 by running software that performs Fourier transform and inverse Fourier transform, and as a fall risk determination unit 27 by running software that extracts the subject's fall risk.
  • the storage device 220 corresponds to the storage unit 26 shown in FIG. 1, and is composed of a volatile storage device such as a RAM (Random Access Memory), a ROM (Read Only Memory), a HDD (Hard Disk Drive), or a non-volatile storage device such as a flash memory.
  • a volatile storage device such as a RAM (Random Access Memory), a ROM (Read Only Memory), a HDD (Hard Disk Drive), or a non-volatile storage device such as a flash memory.
  • the input/output interface 230 includes the video input unit 21 shown in FIG. 1 and is a port to which an input device and an output device are connected.
  • a specific example of the input/output interface 230 is a USB terminal or the like.
  • the input device is the camera 10, as well as a touch panel, a keyboard, a mouse, etc.
  • the output device is the display 300, as well as a light, a speaker, a vibrator, etc.
  • FIG. 3 is a flowchart showing an example of the operation of the fall risk extraction device 20 according to embodiment 1.
  • the process shown in FIG. 3 is started, for example, when the subject starts walking.
  • step S101 an image of the subject captured by the camera 10 is input to the image input unit 21.
  • step S102 the skeletal information extraction unit 22 extracts skeletal information of the subject's body from the image input to the image input unit 21.
  • the multiple joint information preprocessing unit 23 calculates time series data of the positions or angles of the skeletal joints and data obtained by differentiating the time series data of the positions or angles of the skeletal joints as initial walking characteristics of the multiple joints from the skeletal information extracted by the skeletal information extraction unit 22.
  • the differentiated data is, for example, time series data of velocity or angular velocity.
  • the time series data of the positions of the skeletal joints is set to the coordinate values of one axis, two axes, or three axes of the skeletal joints as time passes during walking. Note that the coordinate values are values that have a point in the shooting environment or a fixed point on the screen of the image as the origin.
  • the angle of a skeletal joint is set to the amount that represents the distance between two line segments that have the same endpoints when there are line segments connecting one joint to any two other joints.
  • the time series data of the angle is set to the angle value as time passes during walking.
  • step S104 the multiple joint information preprocessing unit 23 adds or complex-expresses the initial gait features of the multiple joints extracted in step S103 between the joints.
  • the multiple joint information preprocessing unit 23 extracts the gait features after addition processing by adding the values of the initial gait features of all the joints at each time, which is the time series data of the multiple joints.
  • the initial gait features are used.
  • the multiple joint information preprocessing unit 23 extracts the gait features after complex expression by expressing the initial gait features, which are the time series data of the two joints, as a complex number in which the value of the initial gait feature of one joint is the real part and the value of the initial gait feature of the other joint is the imaginary part at each time.
  • the initial gait features in complex terms, the initial gait features of two or more even number of joints are used.
  • the gait features after addition processing and the gait features after complex expression are collectively referred to as the gait features after calculation processing, which is the time series data of walking.
  • the frequency analysis unit 24 performs frequency analysis on the post-calculation processing gait features obtained in the preprocessing in the multiple joint information preprocessing unit 23 to extract post-frequency analysis gait features. Specifically, a Fourier transform is performed as the frequency analysis of the post-addition processing gait features. Furthermore, an inverse Fourier transform is performed in the frequency analysis of the complex representation gait features. As described above, when a complex number is subjected to an inverse Fourier transform, real time series data can be calculated. The process of obtaining real time series data calculated by the inverse Fourier transform corresponds to a frequency analysis of a trajectory in complex space, and as a result, it becomes possible to calculate the frequency characteristics of the movements of two joints.
  • the post-frequency analysis gait features in embodiment 1 indicate an intensity distribution obtained by decomposing the post-calculation processing gait features into each frequency component contained in the gait features.
  • the fall risk assessment unit 27 refers to the parameters stored in the memory unit 26, and runs a program that reproduces a formula for assessing the fall risk described below, and assesses the fall risk from the frequency analysis results output from the frequency analysis unit 24.
  • the specific method of configuring this program is to configure a program in a programming language such as C++ or Python that performs processing similar to the formula for assessing the fall risk, and incorporate it into the software.
  • the formula for calculating the risk of falling is, for example, the following formulas (1), (2), and (3).
  • Formula (1) is a definition formula for the risk assessment gait feature V(i,n) used for fall risk assessment.
  • the risk assessment gait feature V(i,n) is the negative value of the maximum amplitude of the frequency analysis gait feature. If the multiple joint information preprocessing unit 23 outputs multiple arithmetic processing gait features, there will also be multiple frequency analysis gait features.
  • the i in the risk assessment gait feature V(i,n) is a risk assessment gait feature number that identifies each of the total number I of frequency analysis gait features.
  • all frequency analysis gait features are output when the combination of joints to be added or expressed in complex is changed, and the total number of joint combinations is I. Also, the n in the risk assessment gait feature V(i,n) is a subject number that identifies each of the total number N of subjects.
  • each time the risk assessment gait feature number i and subject number n change the negative value of the maximum amplitude of the gait feature after frequency analysis is extracted as the risk assessment gait feature V(i,n).
  • the risk assessment gait feature V(i,n) is not time-series data, but rather corresponds to one value for each variable i or n.
  • the risk of a subject falling is judged based on the magnitude of the maximum amplitude of the frequency analysis results.
  • the amplitude is large, it can be estimated that the subject has a healthy gait with no shaky gait and has a low risk of falling, whereas if the amplitude is small, it can be estimated that the subject has a weak gait with shaky gait and has a high risk of falling.
  • Equation (2) is a formula for calculating the normalized risk assessment gait feature S(i,n), which has a value between 0 and 100, by normalizing the i-th risk assessment gait feature V(i,n) across all subjects among the risk assessment gait features calculated by equation (1).
  • the normalization process using formula (2) normalizes the risk assessment gait characteristics so that, among all subjects, N in total, the maximum risk assessment gait characteristic is 100 and the minimum risk assessment gait characteristic is 0. Note that the risk assessment gait characteristic S(i,n) after normalization is not a time series, but rather corresponds to one value for each of the variables i and n.
  • Equation (3) is a formula for calculating the fall risk F(n), which is the weighted average of the normalized gait features for risk assessment S(i,n) by weighting and adding the normalized gait features for risk assessment S(i,n) extracted by equation (2) to all the normalized gait features for risk assessment S(i,n) of the total number I for each subject.
  • the weight w(i) in equation (3) is stored as a parameter in the storage unit 26.
  • a parameter capable of determining the risk of falling with higher accuracy is calculated and stored in the storage unit 26.
  • the parameters capable of determining the risk of falling with higher accuracy and stored in the storage unit 26 are created by outputting the result of the fall risk F(n) determination for all possible values of the parameter, and obtaining the parameter for which the error from the true value of the risk of falling is smaller. Note that the method for determining whether the risk of falling has been determined with high accuracy uses data from several hundred or more subjects, and uses the F value (F1 score) described later as an accuracy evaluation index.
  • TUG test is a test that comprehensively assesses walking ability, dynamic balance, agility, etc., and can determine the risk of falling from the length of the measurement time.
  • the true value f(n) of the risk of falling can also be estimated from the diagnosis by a medical therapist, but in the first embodiment, the result obtained by the TUG test, which provides a more objective numerical value, is taken as the true value f(n).
  • the fall risk F(n) used for error evaluation is calculated for all subjects numbering several hundred or more using the above formulas (1) to (3) with the initial value of weight w(i) pre-stored as a parameter in the memory unit 26.
  • the true value f(n) used for error evaluation is calculated for all subjects numbering several hundred or more using the above TUG test. In the first embodiment, if the error between the true value f(n) and the fall risk F(n) is small, it is determined as TRUE, and if it is large, it is determined as FALSE, and TRUE/FALSE is determined for all subjects.
  • TRUE/FALSE judgment whether the error is small or not is judged as follows: for each subject, if the difference between the true value f(n) and the fall risk F(n) is equal to or less than the standard error SE described below, then it is TRUE (detection correct); if it is greater than the standard error SE, then it is FALSE (detection incorrect).
  • the standard error SE is calculated as follows.
  • the fall risk detected by the fall risk extraction device 20 for the first subject is F(1)
  • the average Ea of F(1) through F(N) is calculated using the following formula (4).
  • the variance Sa2 of F(1) through F(N) is calculated by dividing the sum of squares of the deviations between F(1) through F(N) and the average Ea by the total number of subjects N using the following formula (5).
  • the standard deviation Sa of F(1) to F(N), which is the square root of the variance Sa2 is calculated, and the standard error SE is obtained by dividing the standard deviation Sa by the square root of the total number of subjects N, as shown in formula (7).
  • an F-value is calculated, which is the harmonic mean of the recall and precision rates obtained based on the binary classification of the TRUE/FALSE judgment results for all subjects using the standard error SE calculated by formula (7). Then, the accuracy of the fall risk F(n) is evaluated based on the magnitude of the calculated F-value.
  • the root mean squared error (RMSE) of the errors of all subjects shown in the following formula (8) may be calculated.
  • a machine-learned model capable of determining the similarity of data distribution may be used to evaluate the error based on the similarity of the distributions of the fall risk F(n) and the true fall risk value f(n).
  • step S107 the calculated risk of falling F(n) is displayed to the subject.
  • a numerical value F(n) for the risk of falling is displayed on the display 300.
  • the display color of the display 300 may be changed according to the value of the risk of falling F(n) to allow the subject to intuitively recognize the risk of falling.
  • a video of walking is acquired, the human skeleton is extracted from the video, the initial walking characteristics of the skeleton are extracted, and the initial walking characteristics of multiple joints are added or complex-expressed and frequency-analyzed to extract the risk of falling F(n) taking into account the correlation between the movements of multiple parts.
  • the extracted risk of falling F(n) makes it possible to evaluate the risk of falling in relation to the interlocking of the movements of multiple parts, and as a result, a more detailed diagnosis of the risk of falling can be made during medical examinations or follow-up observations of rehabilitation.
  • the negative value of the maximum amplitude of the gait feature after frequency analysis is applied to the gait feature for risk assessment V(i,n), but this is not limited to this.
  • the positive value of the maximum amplitude of the gait feature after frequency analysis may be applied to the gait feature for risk assessment V(i,n), and in such a case, the lower the value of the fall risk F(n), the higher the risk of falling.
  • the fall risk F(n) of a subject is calculated based on the gait characteristics of a plurality of joints of the subject.
  • the double support period which is the time when both feet of the subject are in contact with the ground, in addition to the gait characteristics of the plurality of joints of the subject.
  • FIG. 7 is a functional configuration diagram showing a fall risk extraction device 200 according to the second embodiment.
  • the fall risk extraction device 200 further includes a double leg support phase detection unit 25 that detects the double leg support phase of the subject from the initial gait characteristics of the joints of both legs obtained from the skeletal information extracted by the skeletal information extraction unit 22, and the fall risk determination unit 28 differs from the fall risk extraction device 20 according to the first embodiment in that it determines the risk of falling from the frequency analysis results output by the frequency analysis unit 24 and the double leg support phase output by the double leg support phase detection unit 25.
  • the other configuration is the same as that of the fall risk extraction device 20 according to the first embodiment, the same reference numerals as those of the fall risk extraction device 20 are used for the same configuration as that of the fall risk extraction device 20, and detailed description thereof will be omitted.
  • the hardware configuration of the second embodiment is the same as that of the first embodiment, detailed description thereof will be omitted.
  • the skeletal information extraction unit 22 extracts the three-axis coordinates of the joints of the human body as skeletal information from the video input via the video input unit 21, as in the first embodiment, and outputs the extracted skeletal information to the multiple joint information preprocessing unit 23 and the double-leg support period detection unit 25.
  • the double leg support phase detection unit 25 calculates the initial walking characteristics of both ankles of the subject from the skeletal information extracted by the skeletal information extraction unit 22.
  • the initial walking characteristics are, for example, time series data of the position or angle of each of the left and right ankles, and data obtained by differentiating the time series data of the position or angle of each of the left and right ankles.
  • the differential value of the ankle position data is velocity
  • the differential value of the ankle angle is angular velocity.
  • the time series data of the position of each of the left and right ankles is the coordinate values of one axis, two axes, or three axes of each of the left and right ankles as time passes during walking.
  • a point in the shooting environment or a fixed point on the screen of the image is set as the origin, and the x-axis and y-axis are set horizontally from the origin, and the z-axis is set vertically.
  • the x-axis is set in the east-west direction
  • the y-axis is set in the north-south direction.
  • the ankle angle is a quantity that represents the distance between two line segments with the same endpoint when there is a line segment connecting the left or right ankle to any two joints other than the ankle in question.
  • the time series data of the angle is the angle value over time during walking.
  • the double leg support phase of the subject is detected based on the speed of each of the left and right ankles among the initial gait characteristics of both ankles of the subject.
  • FIG. 9(a) is an example of calculating the double leg support period 50 of a subject who is not at risk of falling.
  • FIG. 9(b) is an example of calculating the double leg support period 52 of a subject who is at risk of falling.
  • the speed change of each of the right ankle and the left ankle in the direction of progress during walking is output.
  • the double leg support period 50, 52 is the time when the speed of each of the right ankle and the left ankle is equal to or less than a certain threshold value T.
  • the threshold value T is specifically determined, for example, by comparing and weighing the actual walking state of the subject and the speed of each of the right ankle and the left ankle of the subject. Comparing FIG. 9(a) and FIG. 9(b), the double leg support period 52 in FIG. 9(b) is longer than the double leg support period 50 in FIG. 9(a), suggesting that the risk of falling increases when the double leg support period is longer.
  • the initial gait features used to detect the double support phase are the ankle velocity changes in the direction of travel during walking, but time series data of the left and right directions during walking, the position of any joint in the vertical direction, acceleration, or time series data of the angle or angular velocity of the ankle joints on each side may also be used.
  • the fall risk assessment unit 28 extracts the subject's risk of falling based on the frequency analysis results output by the frequency analysis unit 24, the double support period output by the double support period detection unit 25, and the data stored in the memory unit 26.
  • the subject's risk of falling is an index whose value ranges from a minimum of 0 to a maximum of 100, with a higher value indicating an increased risk of falling and a lower value indicating a decreased risk of falling.
  • the memory unit 26 stores parameters of a calculation formula for extracting the risk of falling from gait characteristics after frequency analysis.
  • the display unit 30 also notifies the subject of the value of the fall risk extracted by the fall risk assessment unit 28 by displaying it on the display 300 or the like. For example, the display 300 may display characters such as "Fine" to indicate a low risk of falling, or "Frail" to indicate a high risk of falling.
  • FIG. 8 is a flowchart showing an example of the operation of the fall risk extraction device 200 according to embodiment 2. Steps S201, S202, S203, S204, and S205 in FIG. 8 are the same as steps S101, S102, S103, S104, and S105 in the flowchart of embodiment 1 shown in FIG. 3, and therefore detailed explanations will be omitted.
  • step S206 the double-leg supporting phase detection unit 25 extracts the initial gait characteristics of both ankles from the skeletal information extracted by the skeletal information extraction unit 22.
  • the initial gait characteristics of both ankles extracted by the double-leg supporting phase detection unit 25 are, for example, the speed of each of the left and right ankles.
  • the double-leg support phase detection unit 25 detects the double-leg support phase from the initial walking characteristics of the joints of both legs extracted in step S206.
  • the double-leg support phases 50 and 52 are defined as the times when the speeds of the right and left ankles are equal to or less than a certain threshold value T, as shown in Figures 9(a) and 9(b).
  • the fall risk assessment unit 28 refers to the parameters stored in the memory unit 26, and runs a program that reproduces a formula for assessing the fall risk, which will be described later, and assesses the fall risk from the frequency analysis results output from the frequency analysis unit 24.
  • the specific method of configuring this program is to configure a program in a programming language such as C++ or Python that performs processing similar to the formula for assessing the fall risk, and incorporate the program into the software.
  • the formulas for determining the risk of falling are, for example, the above formulas (2) and (3) and the following formula (9).
  • Formula (9) is a definition formula for the risk determination gait feature V(i,n) used in determining the risk of falling.
  • the risk determination gait feature V(i,n) is the negative value of the maximum amplitude of the gait feature after frequency analysis or the double leg support period.
  • the i in the risk determination gait feature V(i,n) is a risk determination gait feature number that identifies each of the total number I of the gait features after frequency analysis.
  • the n in the risk determination gait feature V(i,n) is a subject number that identifies each of the total number N of subjects.
  • the maximum amplitude of the frequency analysis result is used to determine the risk of falling. If the amplitude is large, the patient is able to walk normally without any gait fluctuations and the risk of falling is low. If the amplitude is small, the patient is able to walk weakly and the risk of falling is high. The risk of falling is also determined based on the magnitude of the value of the double leg support period. If the double leg support period is small, the patient is able to walk normally with smooth switching between the stance and swing legs of the left and right legs and the risk of falling is low. If the double leg support period is large, the patient is able to walk weakly and the risk of falling is high.
  • formula (2) is a formula for calculating the normalized risk assessment gait feature S(i,n) indicating a value between 0 and 100 by normalizing the i-th risk assessment gait feature V(i,n) across all subjects.
  • the risk assessment gait feature V(i,n) calculated using the above formula (9) is applied to formula (2).
  • the normalization process using formula (2) normalizes the risk assessment gait features so that, among all subjects, the total number N, the maximum risk assessment gait feature is 100 and the minimum risk assessment gait feature is 0.
  • the normalized risk assessment gait feature S(i,n) is not a set of multiple values such as a time series, but rather corresponds to one value for each of the variables i or n.
  • formula (3) calculates the fall risk F(n) by weighting and adding the normalized gait features for risk assessment S(i,n) extracted by formula (2) for each subject to all the normalized gait features for risk assessment S(i,n) of the total number I.
  • the weight w(i) in formula (3) is stored as a parameter in the storage unit 26.
  • a parameter that can determine the risk of falling with higher accuracy is calculated and stored in the storage unit 26.
  • the parameters that can determine the risk of falling with higher accuracy and are stored in the memory unit 26 are created in the same way as in the first embodiment by outputting the fall risk F(n) determination results for all possible values of the parameters and obtaining the parameters for which the error from the true value of the risk of falling is smaller.
  • the error between the fall risk F(n), which is the fall risk determination result, and the true value of the risk of falling f(n) is evaluated in the same way as in the first embodiment by calculating the root mean square error RMSE of the errors of all subjects, with the F value being the accuracy evaluation index, or by using a machine learning model that can determine the similarity of the distribution of data.
  • step S209 the calculated risk of falling F(n) is displayed to the subject. Specifically, a numerical value F(n) for the risk of falling is displayed on the display 300.
  • the display color of the display 300 may be changed according to the value of the risk of falling F(n) to allow the subject to intuitively recognize the risk of falling.
  • the risk of falling F(n) can be calculated based on the walking characteristics of the subject's multiple joints and the double support phase.
  • the calculated risk of falling F(n) makes it possible to evaluate the risk of falling while taking into account the interlocking of the movements of multiple parts of the body and the smoothness of switching between the stance leg and the swing leg of the left and right legs. As a result, a more detailed diagnosis of the risk of falling can be made during medical examinations or follow-up observations of rehabilitation.
  • the double-leg support phase detection unit 25 extracts the gait characteristics of both ankles, but this is not limited to this.
  • the multiple joint information preprocessing unit 23 may also extract the gait characteristics of both ankles.
  • the walking features including the correlation between the movements of a plurality of joints are frequency analyzed, and the weighted average of the normalized maximum amplitude of the obtained frequency characteristics is calculated to determine the risk of falling of the subject, but in the third embodiment, the risk of falling is calculated using the "correlation coefficient of features between joints.”
  • the configuration of the fall risk extraction device according to the third embodiment is the same as that of the fall risk extraction device 20 according to the first embodiment, and therefore a detailed description thereof will be omitted.
  • the hardware configuration of the second embodiment is the same as that of the first embodiment, and therefore a detailed description thereof will be omitted.
  • Joint features refers to, for example, time series data of joint displacement extracted by the multiple joint information preprocessing unit 23, or frequency spectrum data of frequency characteristics.
  • the correlation coefficient is calculated by the following procedure.
  • For feature A calculate the average Ea of A1 to An, and calculate the deviation between each of A1 to An and the average Ea. Divide the sum of the squares of each of the calculated deviations by n to calculate the variance Sa2 of A1 to An, and then calculate the standard deviation Sa, which is the square root of variance Sa2.
  • For feature B calculate the average Eb of B1 to Bn, the deviation between each of B1 to Bn and the average Eb, the variance Sb2 of B1 to Bn, and the standard deviation Sb.
  • the correlation coefficient r is obtained by dividing the covariance Sab by the product of the standard deviation Sa and the standard deviation Sb, as shown in the following formula (10).
  • one correlation coefficient r value can be calculated for each combination of two features, such as features A and B, for each subject.
  • each joint has multiple gait characteristics, such as vertical displacement, forward speed, and angular displacement around the vertical axis, multiple correlation coefficients r can be calculated using these multiple gait characteristics.
  • the correlation coefficient r has a value of -1 ⁇ r ⁇ 1, and the closer r is to 1, the more positive the correlation between features A and B is, and the closer it is to -1, the more negative the correlation between features A and B is. Furthermore, when r is close to 0, there is no correlation between features A and B.
  • the calculated correlation coefficient r is used for fall risk detection in the same way as the maximum amplitude of the frequency characteristics or the double support period. For example, assume that feature A is time series data of the displacement amount of the position of the subject's right ankle joint that is equal to or greater than a predetermined threshold, and feature B is time series data of the displacement amount of the position of the subject's left ankle joint that is equal to or greater than a predetermined threshold. If there is a difference in the displacement amount of the position of the left and right ankle joints of the subject, it can be estimated that the subject's gait is unstable.
  • the correlation coefficient r calculated for features A and B is a negative value close to -1, indicating that there is a negative correlation between the displacement amounts of the left and right ankle joints of the subject. Conversely, if there is no difference in the displacement amounts of the left and right ankle joints of the subject, it can be estimated that the subject's gait is stable. In such a case, the correlation coefficient r calculated for features A and B is a negative value close to 1, indicating that there is a positive correlation between the displacement amounts of the left and right ankle joints of the subject.
  • the "skeletal information extraction unit” in the claims corresponds to the "skeletal information extraction unit 22" described in the detailed description of the invention
  • the "joint information extraction unit” and “gait feature calculation unit” in the claims correspond to the “multiple joint information preprocessing unit 23" described in the detailed description of the invention
  • the "frequency analysis unit” in the claims corresponds to the “frequency analysis unit 24" described in the detailed description of the invention
  • the “fall risk extraction unit” in the claims corresponds to the “fall risk determination units 27, 28" described in the detailed description of the invention
  • the “double legs supporting period detection unit” in the claims corresponds to the “double legs supporting period detection unit 25" described in the detailed description of the invention.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Dentistry (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Physiology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

L'invention concerne un dispositif d'extraction de risques de chute (20) comprenant : une unité d'extraction d'informations de squelette (22) qui extrait des informations sur le squelette d'un sujet à partir d'une vidéo dans laquelle le sujet marche ; une unité de prétraitement d'informations sur plusieurs articulations (23) qui extrait, à partir des informations sur le squelette, des informations incluant des changements chronologiques de position, d'angle, de vitesse et de vitesse angulaire au niveau d'articulations respectives du sujet lors de la marche, et qui calcule, sur la base des informations extraites, des caractéristiques de marche incluant des corrélations dans des mouvements des articulations ; et une unité de détermination de risque de chute (27) qui extrait un risque de chute du sujet sur la base des caractéristiques de marche incluant les corrélations des mouvements des articulations.
PCT/JP2023/030025 2023-08-21 2023-08-21 Dispositif d'extraction de risques de chute, procédé d'extraction de risques de chute, et programme d'extraction de risques de chute Pending WO2025041230A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/JP2023/030025 WO2025041230A1 (fr) 2023-08-21 2023-08-21 Dispositif d'extraction de risques de chute, procédé d'extraction de risques de chute, et programme d'extraction de risques de chute
JP2024558135A JP7668970B1 (ja) 2023-08-21 2023-08-21 転倒リスク抽出装置、転倒リスク抽出方法、及び転倒リスク抽出プログラム

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2023/030025 WO2025041230A1 (fr) 2023-08-21 2023-08-21 Dispositif d'extraction de risques de chute, procédé d'extraction de risques de chute, et programme d'extraction de risques de chute

Publications (1)

Publication Number Publication Date
WO2025041230A1 true WO2025041230A1 (fr) 2025-02-27

Family

ID=94731829

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2023/030025 Pending WO2025041230A1 (fr) 2023-08-21 2023-08-21 Dispositif d'extraction de risques de chute, procédé d'extraction de risques de chute, et programme d'extraction de risques de chute

Country Status (2)

Country Link
JP (1) JP7668970B1 (fr)
WO (1) WO2025041230A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN121662357A (zh) * 2026-02-06 2026-03-13 中国人民解放军空军军医大学 一种基于患者实时远程监护及康复监测系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010172481A (ja) * 2009-01-29 2010-08-12 Wakayama Prefecture 下肢協調性評価システム
JP2015062654A (ja) * 2013-08-28 2015-04-09 日本電信電話株式会社 歩容推定装置とそのプログラム、転倒危険度算出装置とそのプログラム
JP2020048683A (ja) * 2018-09-25 2020-04-02 公立大学法人大阪 片麻痺の検査装置
JP2020077388A (ja) * 2018-09-28 2020-05-21 医療法人社団皓有会 運動解析装置
JP2021030051A (ja) * 2019-08-29 2021-03-01 パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカＰａｎａｓｏｎｉｃ Ｉｎｔｅｌｌｅｃｔｕａｌ Ｐｒｏｐｅｒｔｙ Ｃｏｒｐｏｒａｔｉｏｎ ｏｆ Ａｍｅｒｉｃａ 転倒リスク評価方法、転倒リスク評価装置及び転倒リスク評価プログラム
WO2022249746A1 (fr) * 2021-05-27 2022-12-01 パナソニックＩｐマネジメント株式会社 Système d'estimation de capacité physique, procédé d'estimation de capacité physique et programme

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2022092940A (ja) * 2020-12-11 2022-06-23 トヨタ紡織株式会社 推定装置、推定方法および推定プログラム

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010172481A (ja) * 2009-01-29 2010-08-12 Wakayama Prefecture 下肢協調性評価システム
JP2015062654A (ja) * 2013-08-28 2015-04-09 日本電信電話株式会社 歩容推定装置とそのプログラム、転倒危険度算出装置とそのプログラム
JP2020048683A (ja) * 2018-09-25 2020-04-02 公立大学法人大阪 片麻痺の検査装置
JP2020077388A (ja) * 2018-09-28 2020-05-21 医療法人社団皓有会 運動解析装置
JP2021030051A (ja) * 2019-08-29 2021-03-01 パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカＰａｎａｓｏｎｉｃ Ｉｎｔｅｌｌｅｃｔｕａｌ Ｐｒｏｐｅｒｔｙ Ｃｏｒｐｏｒａｔｉｏｎ ｏｆ Ａｍｅｒｉｃａ 転倒リスク評価方法、転倒リスク評価装置及び転倒リスク評価プログラム
WO2022249746A1 (fr) * 2021-05-27 2022-12-01 パナソニックＩｐマネジメント株式会社 Système d'estimation de capacité physique, procédé d'estimation de capacité physique et programme

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN121662357A (zh) * 2026-02-06 2026-03-13 中国人民解放军空军军医大学 一种基于患者实时远程监护及康复监测系统

Also Published As

Publication number Publication date
JPWO2025041230A1 (fr) 2025-02-27
JP7668970B1 (ja) 2025-04-25

Similar Documents

Publication Publication Date Title
JP7184398B2 (ja) 自動診断装置
US20220189211A1 (en) Physical activity quantification and monitoring
US12279882B2 (en) Movement disorder diagnostics from video data using body landmark tracking
Nieto-Hidalgo et al. A vision based proposal for classification of normal and abnormal gait using RGB camera
Smeragliuolo et al. Validation of the Leap Motion Controller using markered motion capture technology
US11826140B2 (en) System and method for human motion detection and tracking
Bonnet et al. Monitoring of hip and knee joint angles using a single inertial measurement unit during lower limb rehabilitation
Vun et al. Vision-based motion capture for the gait analysis of neurodegenerative diseases: A review
US20200372245A1 (en) Scoring metric for physical activity performance and tracking
Guffanti et al. Non-invasive multi-camera gait analysis system and its application to gender classification
CA2821590A1 (fr) Procede permettant d'evaluer une fonction cognitive et de predire un declin cognitif par le biais d'une evaluation quantitative du test de levee et de marche synchronisees
CN114514559A (zh) 运动识别方法、运动识别程序及信息处理装置
JP7668970B1 (ja) 転倒リスク抽出装置、転倒リスク抽出方法、及び転倒リスク抽出プログラム
Qian et al. Combining deep learning and model-based method using Bayesian Inference for walking speed estimation
Kobsar et al. Validity of a novel method to measure vertical oscillation during running using a depth camera
CN116665895A (zh) 一种帕金森运动障碍量化分析实现系统
Jia Influence of wearable biometric sensors on performance indicators of volleyball players
US20240152666A1 (en) Simulation of accelerometer data
Fazeli et al. Estimation of spatial-temporal hand motion parameters in rehabilitation using a low-cost noncontact measurement system
JP2018038752A (ja) 歩行分析方法及び歩行分析装置
WO2023280723A1 (fr) Système et procédé d'évaluation d'équilibre du corps entier
Šprdlík et al. Tremor analysis by decomposition of acceleration into gravity and inertial acceleration using inertial measurement unit
Cordero-Sánchez et al. Comparison of ground reaction forces and net joint moment predictions: Skeletal model versus artificial neural network-based approach
EP4053793A1 (fr) Système et procédé de détection et de suivi de mouvement humain
Zhao et al. Early detection of Parkinson’s disease based on IMU and machine learning

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2024558135

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2024558135

Country of ref document: JP

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23949691

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE