WO2022216102A1 - System and method for determining severe stage of parkinson's syndrome on basis of gait data - Google Patents

System and method for determining severe stage of parkinson's syndrome on basis of gait data Download PDF

Info

Publication number
WO2022216102A1
WO2022216102A1 PCT/KR2022/005088 KR2022005088W WO2022216102A1 WO 2022216102 A1 WO2022216102 A1 WO 2022216102A1 KR 2022005088 W KR2022005088 W KR 2022005088W WO 2022216102 A1 WO2022216102 A1 WO 2022216102A1
Authority
WO
WIPO (PCT)
Prior art keywords
pedestrian
parkinson
syndrome
data
length
Prior art date
Application number
PCT/KR2022/005088
Other languages
French (fr)
Korean (ko)
Inventor
강성우
Original Assignee
인하대학교 산학협력단
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 인하대학교 산학협력단 filed Critical 인하대학교 산학협력단
Publication of WO2022216102A1 publication Critical patent/WO2022216102A1/en

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
    • A61B5/40—Detecting, measuring or recording for evaluating the nervous system
    • A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
    • A61B5/4082—Diagnosing or monitoring movement diseases, e.g. Parkinson, Huntington or Tourette
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording 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, mobility of a limb
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording 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, mobility of a limb
    • A61B5/112—Gait analysis
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording 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, mobility of a limb
    • A61B5/1126—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb using a particular sensing technique
    • A61B5/1128—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb using a particular sensing technique using image analysis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment

Definitions

  • the present invention relates to a system and method for determining the severity of Parkinson's syndrome based on gait data, and more particularly, to a system and method for determining the severity of Parkinson's syndrome based on gait data including a photographing unit and a server.
  • Parkinson's disease is a neurodegenerative disease characterized by the death of nerve cells that secrete a neurotransmitter called dopamine in the brain.
  • Parkinson's syndrome is also called Parkinson's complex syndrome, which means that the symptoms of Parkinson's disease are combined with other symptoms.
  • Parkinson's disease which can be treated with drugs
  • the overall course is different from Parkinson's disease because it does not respond well to drug treatment and the disease progresses quickly.
  • Korean Patent Application No. 10-2017-0122419 discloses an apparatus and a measurement method for quantitative measurement of freezing of gait in Parkinson's disease patients, but the technique is described in Parkinson's disease patient's gait freezing. In addition to the inconvenience of having to attach the device to the patient's ankle to measure There is a problem.
  • an embodiment of the present invention provides a server for determining a severe stage by receiving the photographing unit for photographing the side of the pedestrian and the photographic data taken by the photographing unit when the pedestrian walks forward,
  • the server is an extractor for extracting the skeleton data of the pedestrian from the photographed photo data, Calculate the average deviation / pelvic length of the difference between the z-coordinates of each knee of the pedestrian based on
  • a calculation unit that calculates the average deviation/pelvic length of the difference in the z-coordinate of each ankle of the pedestrian and the average deviation/pelvic length of the difference in the z-coordinate of each knee of the pedestrian and the difference in the z-coordinate of each ankle of the pedestrian based on
  • It may include a gait data-based Parkinson's syndrome severity stage determination system including a determination unit that determines whether the Parkinson's syndrome state of the pedestrian is included within a preset range from the average deviation/pelvic length.
  • the calculation unit from the skeleton data Calculating the walking speed / leg length based on the gait data-based Parkinson's syndrome severe stage determination system that determines whether the state of the pedestrian's Parkinson's syndrome is within a preset range from the walking speed / leg length have.
  • the calculation unit from the skeleton data Calculates the average foot length of both feet of a walking pedestrian / walker's foot viewed from the side based on It may include a gait data-based Parkinson's syndrome severity stage determination system that determines whether it is included within a preset range.
  • the calculation unit from the skeleton data Calculates the average of the bending angles of both knees while walking, and the determination unit determines whether the Parkinson's syndrome state of the pedestrian is included within a preset range from the average of the bending angles of both knees while walking.
  • the determination unit is the average deviation/pelvic length of the difference between the walking speed/leg length calculated by the calculation unit, the z-coordinate of each knee of the pedestrian, and the average deviation/pelvic length of the difference between the z-coordinate of each ankle of the pedestrian .
  • the server may include a gait data-based Parkinson's syndrome severe stage determination system including a learning unit for deep learning the Parkinson's determination data.
  • a method for determining the severity of Parkinson's syndrome based on gait data comprising a third step of calculating with Parkinson's discrimination data and a fourth step of determining whether the parkinson's syndrome of a pedestrian is higher than a preset standard based on the deep-learned Parkinson's discrimination data may include
  • the third step is from the skeleton data, Comprising the step of calculating the average deviation / pelvic length of the difference of the z-coordinate of each knee of the pedestrian based on It may include a method for determining the severity of Parkinson's syndrome based on gait data including the step of determining whether it is included within the criteria.
  • the third step is from the skeleton data, Further comprising the step of calculating the average deviation / pelvic length of the difference in the z-coordinate of each ankle of the pedestrian based on It may include a method for determining the severity of Parkinson's syndrome based on gait data further comprising the step of determining whether it is included within the criteria.
  • the third step is from the skeleton data
  • the step of calculating the walking speed / leg length based on the gait data further comprising the step of determining whether the fourth step is included within a preset standard from the walking speed / leg length Determination of the severe stage of Parkinson's syndrome methods may be included.
  • the third step is from the skeleton data, Further comprising the step of calculating the average foot length of both feet of the walking pedestrian / foot length of the pedestrian viewed from the side based on It may include a method for determining the severity of Parkinson's syndrome based on gait data further comprising the step of determining whether it is included within the preset criteria from the gait data.
  • the third step is from the skeleton data, Further comprising the step of calculating the average of the bending angle of both knees while walking based on based methods for determining the severity of Parkinson's syndrome.
  • An embodiment of the present invention aims to provide a system and method for determining the severity of Parkinson's syndrome based on gait data.
  • An embodiment of the present invention can determine the severity of Parkinson's syndrome patients without performing various functional tests, such as autonomic nervous system examination or brain MRI, only with the gait of the Parkinson's syndrome patient by photographing the walking state of the Parkinson's syndrome patient.
  • an embodiment of the present invention uses the calculated average deviation/pelvic length of the z-coordinate of each knee of the pedestrian, which is the calculated detection factor 1, and the calculated average deviation/pelvic length of the z-coordinate of each ankle, which is the calculated detection factor 2, of the pedestrian.
  • one embodiment of the present invention can determine the Parkinson's syndrome severe stage of the Parkinson's syndrome patient using the calculated detection factor 3, walking speed/leg length.
  • an embodiment of the present invention can determine the Parkinson's syndrome severe stage of the Parkinson's syndrome patient using the average foot length/walker's foot length of both feet of a walking pedestrian viewed from the side of the calculated detection element 4 .
  • the severity of Parkinson's syndrome of a patient with Parkinson's syndrome can be determined using the average of the bending angles of both knees while walking, which is the calculated detection element 5.
  • an embodiment of the present invention has an effect that it is possible to more accurately determine the severity of the Parkinson's syndrome patient using five detection elements.
  • the determined content is transmitted not only to the patient but also to the medical staff, so that the condition of the Parkinson's syndrome patient can be quickly checked and prompt treatment is possible.
  • FIG. 1 is a conceptual diagram illustrating a situation in which a pedestrian walking forward is photographed from the side based on the gait data-based Parkinson's syndrome severity stage determination system according to an embodiment of the present invention.
  • FIG. 2 is a diagram specifically illustrating the configuration of a gait data-based Parkinson's syndrome severity stage determination system according to an embodiment of the present invention.
  • FIG. 3 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 1 used in the gait data-based parkinson's syndrome severe stage determination system and method according to an embodiment of the present invention.
  • FIG. 4 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed in order to calculate the detection element 2 used in the system and method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
  • FIG. 5 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed in order to calculate the detection element 3 used in the gait data-based parkinson's syndrome severity stage determination system and method according to an embodiment of the present invention.
  • FIG. 6 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed in order to calculate the detection element 4 used in the system and method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
  • FIG. 7 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 5 used in the gait data-based parkinson's syndrome severe stage determination system and method according to an embodiment of the present invention.
  • FIG. 8 is a flowchart of a method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
  • the camera used in the present invention may include a Time of Flight (ToF) module inside the camera.
  • ToF Time of Flight
  • the Time of Flight (ToF) module emits light toward the subject, and calculates a time for the emitted light to be reflected back after hitting the subject as a distance.
  • photo data refers to a frame-by-frame scene of an image being reported by a pedestrian taken by using the camera used in the present invention.
  • Skeleton data is obtained by extracting a total of 31 characteristics by designating each point of the pedestrian's skeleton as one point from the photographed photo data.
  • 31 characteristics included in the extracted skeleton data include nose, both eyes, both ears, head, neck, elbow, wrist, chest, waist, pelvis, hip, knee, foot, etc. For the example, only nine characteristics are used, including points of the pelvis, both hips, both knees, both ankles, and both feet.
  • FIGS. 1 and 2 are conceptual diagram schematically illustrating a gait data-based Parkinson's syndrome severe stage determination system according to an embodiment of the present invention
  • FIG. 2 is a configuration of a gait data-based Parkinson's syndrome severe stage determination system according to an embodiment of the present invention It is a drawing showing in detail.
  • the system for determining the severity of Parkinson's syndrome based on gait data which is an embodiment of the present invention, is a system for determining the severity of a Parkinson's syndrome patient by using the photographed data taken from the side of the Parkinson's syndrome patient walking in front.
  • the gait data-based Parkinson's syndrome severity stage determination system may include a photographing unit 100 , a server 200 , and a terminal 300 .
  • the photographing unit 100 may include a camera for photographing the side of the pedestrian toward the front.
  • the server 200 may include an extracting unit 210 , a calculating unit 220 , a learning unit 230 , and a determining unit 240 .
  • the terminal 300 may include a display unit 310 .
  • the display unit 310 may include a display panel that displays information about the Parkinson's syndrome patient p on the screen. However, the present invention is not limited thereto, and the display unit 310 may include various configurations supporting the output of information.
  • the recording unit 100, the server 200, the terminal 300 is DMR and WiFi, low energy Bluetooth (BLE), Bluetooth (Bluetooth), 3G (3rd Generation), 3GPP (3rd Generation Partnership Project), LTE (Long Term) Evolution), LTE-A, 4G (4th Generation), and 5G (5th Generation) may perform wireless communication with an external device through at least one communication module.
  • BLE low energy Bluetooth
  • Bluetooth Bluetooth
  • 3G 3rd Generation
  • 3GPP 3rd Generation Partnership Project
  • LTE Long Term) Evolution
  • LTE-A Long Term Evolution
  • 4G (4th Generation) 4G (4th Generation
  • 5G (5th Generation) may perform wireless communication with an external device through at least one communication module.
  • the extraction unit 210 may extract skeleton data of a pedestrian from the photo data captured by the photographing unit 100 .
  • the calculation unit 220 may calculate five detection elements to be used to determine the severity of the Parkinson's syndrome patient based on the skeleton data of the pedestrian extracted by the extraction unit.
  • the calculated five detection elements are referred to as Parkinson's discrimination data.
  • the five detection elements will be described later.
  • the learning unit 230 can perform deep learning from the Parkinson's determination data in the calculating unit 230, and the determining unit 240 determines the severe stage of Parkinson's syndrome of a pedestrian through a preset criterion based on the deep-learned Parkinson's determination data. can be discerned.
  • the side of the pedestrian can be photographed using the camera c.
  • the camera (c) located on the side of the Parkinson's syndrome patient (p) photographs the walking of the Parkinson's syndrome patient. Only the skeleton data of the lower body to be used in the calculation of the Parkinson's syndrome patient (p) is extracted from the photographed photo data.
  • the server 200 calculates Parkinson's determination data, which will be described later, by using the extracted skeleton data. The process of calculating the Parkinson's discrimination data will be described later.
  • the calculated Parkinson's discrimination data becomes deep learning together with Parkinson's discrimination data of other Parkinson's syndrome patients in the learning unit 230 .
  • the deep-learning Parkinson's discriminant data determines whether the discriminant 240 is higher than a preset standard and determines the Parkinson's syndrome severe stage of the Parkinson's syndrome patient (p).
  • the determined severe stage of the Parkinson's syndrome patient (p) may be displayed to the Parkinson's syndrome patient (p) through the display unit 310, and may be transmitted to the medical staff as well as the Parkinson's syndrome patient (p).
  • FIG. 3 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 1 used in the gait data-based parkinson's syndrome severe stage determination system and method according to an embodiment of the present invention
  • FIG. 4 is the present invention.
  • FIG. 1 It is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate a detection element 2 used in a system and method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention
  • Figure 5 is an embodiment of the present invention
  • FIG. 1 It is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 3 used in the gait data-based Parkinson's syndrome severe stage determination system and method according to the present invention
  • FIG. It is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 4 used in the system and method for determining the severe stage of the syndrome
  • FIG. 7 is a system for determining the stage of severe Parkinson's syndrome based on gait data according to an embodiment of the present invention. And it is a conceptual diagram schematically illustrating a situation of photographing a pedestrian in order to calculate the detection element 5 used in the method.
  • Detection element 1 is the average deviation/pelvic length of the difference between the z-coordinates of each knee of the pedestrian.
  • the expression for calculating detection element 1 is to be. at is the z-coordinate of the left knee, is the z-coordinate of the right knee. denotes the time the pedestrian walked the preset distance. is the pelvic length of the pedestrian (p) being photographed.
  • the reason for dividing the average deviation of the difference in the z-coordinate of each knee of the pedestrian (p) by the pelvic length of the pedestrian (p) is that each pedestrian (p) is tall, so there is a difference in walking speed, and the pelvis is wide. This is to standardize because there is a difference in the way the knee is opened by default.
  • the reason that only the z-coordinate of each knee of the pedestrian p is used in the equation for calculating the detection element 1 is that the x-coordinate is a value that changes to the pedestrian p moving forward or backward, making it difficult to characterize the walking knee. It was not included by judgment, and the y-coordinate of each knee is also a value that changes when the body coordinates of the pedestrian (p) move up and down, so it is not included like the x-coordinate.
  • the gait of Parkinson's syndrome patients differs from that of ordinary people in that they walk with their knees apart. In order to use the characteristic of walking with the knees apart, the z-coordinate must be used in the three-dimensional coordinate system shown in FIG. 2 .
  • the x-coordinate is a value that changes when the pedestrian walks back and forth
  • the y-coordinate is a value that changes when the pedestrian moves up and down. Therefore, since a pedestrian walking toward the front is photographed from the side, the width of the knee can be expressed as a z-coordinate. Therefore, the z-coordinate of each knee of the pedestrian (p) is used.
  • the detection factor 2 is the average deviation/pelvic length of the difference between the z-coordinates of each ankle of the pedestrian.
  • the expression for calculating detection factor 2 is to be. It is similar to detection element 1, but with the difference that detection element 1 uses the z-coordinate of the pedestrian (p)'s knee, and detection element 2 uses the z-coordinate of the pedestrian's (p) ankle. of the above formula at is the z-coordinate of the left ankle is the z-coordinate of the right ankle. denotes the time the pedestrian walked the preset distance. of the above formula is the pelvic length of the pedestrian (p) being photographed.
  • the reason for dividing is the same as the reason for dividing the average deviation of the z-coordinate of each knee of the walking pedestrian (p) by the pelvis length of the walking pedestrian (p) in the equation for calculating the detection factor 1.
  • This is to standardize because there is a difference, and the difference is in the fact that the knee is basically widened and the distance between the ankles is farther due to the wide pelvis.
  • the difference between walking with patients with Parkinson's syndrome and the general public is the characteristic of walking with the knees apart. As a result, the knees are spread apart, and the distance between the ankles is also farther away.
  • the z-coordinate that can represent perspective should be used based on the camera (c) photographed from the side of the pedestrian (p).
  • the detection factor 3 is walking speed/leg length.
  • the expression for calculating detection element 3 is to be. at is the x-coordinate of the waist of the pedestrian (p), denotes the time the pedestrian walked the preset distance. In addition, is the leg length of the pedestrian (p). As described above, the x-coordinate is a value that changes when the pedestrian moves back and forth, and becomes a coordinate indicating the distance the pedestrian walked from the starting point to the ending point. Therefore, the expression for calculating the detection factor 3 at is the total distance traveled by the pedestrian during the recording divided by the time walked, so it means the speed at which the pedestrian walked. cast The reason for dividing by ' is to standardize because the size of the stride varies according to the length of the leg and the speed changes even when walking the same distance.
  • the detection factor 4 is the average foot length / foot length of both feet of a walking pedestrian when viewed from the side.
  • the expression for calculating detection factor 4 is to be.
  • the x-coordinate of the left foot is the x-coordinate of the left ankle.
  • the x-coordinate of the right foot is the x-coordinate of the right ankle.
  • the formula for calculating the detection element 4 means the footsteps of pedestrians, denotes the time the pedestrian walked the preset distance.
  • the foot length of the left foot viewed from the side is can be calculated with The front point of the pedestrian's left foot viewed from the side ( ) at the ankle point of the left foot ( ) to get the foot length of the left foot viewed from the side.
  • the foot length of the right foot viewed from the side is can be calculated with The front point of the pedestrian's right foot viewed from the side ( ) at the ankle point of the right foot ( ) to get the foot length of the right foot viewed from the side.
  • the detection element 5 is the average flexion angle of both knees during walking.
  • the expression for calculating detection element 5 is to be.
  • the formula for calculating the detection element 5 is the bending angle of the knee of the left leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the left hip, left knee, and left ankle.
  • the bending angle of the knee of the right leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the right hip, right knee, and right ankle. denotes the time the pedestrian walked the preset distance.
  • FIG. 8 is a flowchart of a method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
  • the method for determining the severity of Parkinson's syndrome based on gait data includes a first step (s100) of deep learning the Parkinson's determination data, a second step of photographing the side of the pedestrian when the pedestrian walks forward (s200), a third step (s300) of extracting skeleton data of a pedestrian from the photographed photo data, and calculating the parkinson's discrimination data from the skeleton data (s300), based on the deep learning Parkinson's discrimination data higher than a preset standard It may include a fourth step (s400) of determining the severity of the pedestrian's Parkinson's syndrome by determining the
  • the first step (s100) of deep learning the Parkinson's identification data is a step in which the Parkinson's identification data of other Parkinson's syndrome patients can be deep-learned before photographing the gait of the Parkinson's syndrome patient to determine the severe stage.
  • the Parkinson's discrimination data of other Parkinson's syndrome patients is deep learning.
  • the Parkinson's syndrome patient to determine the severe stage walks the preset distance of the present invention, and photographs the Parkinson's syndrome patient with a camera from the side. It is possible to more quickly identify the Parkinson's severe stage of Parkinson's syndrome patients from the captured data and the deep-learned Parkinson's identification data.
  • the second step (s200) of photographing the side of the pedestrian is a step in which the side of the pedestrian can be photographed with a camera including a ToF module when the pedestrian walks a preset distance.
  • the third step (s300) of extracting the pedestrian's skeleton data from the photographed photo data and calculating the parkinson's determination data from the skeleton data may include the step (s310) of extracting the pedestrian's skeleton data from the photographed photograph data. have.
  • the third step (s300) of extracting the skeleton data of the pedestrian from the photographed photo data and calculating the Parkinson's discrimination data from the skeleton data is the step of calculating the detection element 1 (s321), the steps of calculating the detection element 2 (s322), calculating the detection element 3 (s323), calculating the detection element 4 (s324), and calculating the detection element 5 (s325) may include.
  • the step of calculating the detection element 1 is from the skeleton data, This is a step of calculating the average deviation/pelvic length of the difference between the z-coordinates of each knee of the pedestrian based on .
  • the equation for calculating detection element 1 is the z-coordinate of the left knee, is the z-coordinate of the right knee.
  • the length of the pedestrian's pelvis is the reason for dividing the average deviation of the difference in the z-coordinate of each knee of a walking pedestrian by the pedestrian's pelvis length. This is to standardize because Parkinson's syndrome patients walk with their knees apart when walking due to pain in the knee and back.
  • the detection factor 1 is calculated by measuring the z-coordinates of both knees of a walking ordinary person, the difference between the two knees will be constant.
  • the degree of divergence of the pedestrian's knee compared to the pedestrian's pelvis length in the detection element 1 calculated using the z-coordinate of the walking pedestrian's knee.
  • the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 1 calculated in the step (s321) of calculating the detection element 1. have. Unlike the gait of the general public, the gait of patients with Parkinson's syndrome has the characteristic of walking with the knees spread apart.
  • the detection element 1 calculated using the z-coordinate of the walking pedestrian's knee contains the degree of divergence between the knees compared to the pelvic length. Therefore, it is possible to determine the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the degree of divergence between the knees compared to the pelvic length contained in the detection element 1.
  • Calculating the detection element 2 (s322) From the skeleton data, This is a step of calculating the average deviation/pelvic length of the difference between the z-coordinates of each ankle of the pedestrian based on .
  • the z-coordinate of the left ankle is the z-coordinate of the right ankle.
  • the length of the pedestrian's pelvis is the length of the pedestrian's pelvis.
  • the reason for dividing is the same as the reason for dividing the average deviation of the z-coordinate of each knee of a walking pedestrian by the length of the pedestrian's pelvis in the equation for calculating the detection factor 1. This is to standardize because there is a difference in the distance between the ankles because the knee is basically widened by the wide one.
  • the difference between walking with patients with Parkinson's syndrome and the general public is the characteristic of walking with the knees apart. As a result, the knees are spread apart, and the distance between the ankles is also farther away. Therefore, in the detection element 2 calculated using the z-coordinate of the walking pedestrian's ankle, the degree of divergence between the pedestrian's ankles compared to the pedestrian's pelvis length can be known.
  • the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 2 calculated in the step (s322) of calculating the detection element 2.
  • the detection element 2 calculated using the z-coordinate of the ankle of the walking pedestrian contains the degree of divergence between the ankles compared to the pelvic length. Therefore, it is possible to discriminate the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the degree of divergence between the ankles compared to the pelvic length contained in the detection element 2.
  • the step of calculating the detection element 3 is from the skeleton data, It is a step of calculating walking speed/leg length based on . at is the x-coordinate of the waist of the pedestrian (p), denotes the time the pedestrian walked the preset distance. In addition, is the leg length of the pedestrian (p). cast The reason for dividing by ' is to standardize because the size of the stride varies according to the length of the leg and the speed changes even when walking the same distance. As mentioned earlier, patients with Parkinson's syndrome have the characteristic of walking with their knees apart due to back and knee pain. However, in addition to the characteristic of walking with the knees apart, patients with Parkinson's syndrome cannot walk quickly due to pain. Therefore, even a Parkinson's syndrome patient with a large stride may have a different walking speed than an ordinary person with the same stride.
  • the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 3 calculated in the step (s323) of calculating the detection element 3. have. Unlike the gait of the general public, the gait of patients with Parkinson's syndrome is characterized by a slow pace. Therefore, it is possible to discriminate the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the walking speed compared to the leg length contained in the detection element 3 .
  • the step of calculating the detection element 4 is from the skeleton data, This is a step of calculating the foot length of the pedestrian / the foot length of the pedestrian while walking as viewed from the side.
  • the x-coordinate of the left foot is the x-coordinate of the left ankle.
  • the x-coordinate of the right foot is the x-coordinate of the right ankle.
  • the formula for calculating the detection element 4 means the footsteps of pedestrians, denotes the time the pedestrian walked the preset distance.
  • the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 4 calculated in the step (s324) of calculating the detection element 4. have.
  • the foot length seen from the side in the walking of ordinary people is the same as that of ordinary people.
  • the foot length viewed from the side in the gait of the Parkinson's syndrome patient is different from the foot length of the Parkinson's syndrome patient.
  • the detection element 4 is calculated using the foot length viewed from the side and the foot length of the pedestrian. Therefore, it is possible to determine the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the foot length viewed from the side when walking compared to the foot length contained in the detection element 4 .
  • the step of calculating the detection element 5 is from the skeleton data, This is a step of calculating the average of each knee bend while walking.
  • the formula for calculating the detection element 5 is the bending angle of the knee of the left leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the left hip, left knee, and left ankle.
  • the bending angle of the knee of the right leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the right hip, right knee, and right ankle. denotes the time the pedestrian walked the preset distance.
  • the difference between walking with patients with Parkinson's syndrome and the general public is the characteristic of walking with the knees apart. Therefore, the more severe the Parkinson's syndrome, the more the knee is spread to both sides and the more the knee is bent. Therefore, it is possible to know the degree to which the pedestrian bends the knee when walking in the detection element 5 calculated using the average of the bending angle of the knee.
  • the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 5 calculated in the step (s325) of calculating the detection element 5. have.
  • the knee flexion angle is different. Therefore, with the angle formed by the hip, knee, and ankle of the Parkinson's syndrome patient when walking contained in the detection element 5, the pedestrian's Parkinson's syndrome severe stage can be determined in the pedestrian's Parkinson's syndrome severe stage (s400).
  • the fourth step (s400) of determining the severe stage of Parkinson's syndrome of a pedestrian extracts the pedestrian's skeleton data from the photographed photo data, and calculates the parkinson's discrimination data from the skeleton data as the Parkinson's discrimination data calculated in the third step (s300) By determining whether it is higher than a preset criterion based on the data, it is possible to determine the severity of Parkinson's syndrome in the pedestrian.
  • the above-described five detection elements are also used as respective indicators for determining the pedestrian's Parkinson's syndrome severe stage in the fourth step (s400) of determining the pedestrian's Parkinson's syndrome severe stage.
  • all five detection elements may be used to determine the pedestrian's Parkinson's syndrome severe stage.
  • Parkinson's syndrome patient (p) there is a patient with Parkinson's syndrome (p).
  • the Parkinson's syndrome patient (p) walks a predetermined distance forward.
  • the side of the Parkinson's syndrome patient (p) is photographed.
  • Skeleton data of a Parkinson's syndrome patient (p) is extracted from the photographed photo data, and only nine characteristics located in the lower body are used for the extracted skeleton characteristics.
  • Patients with Parkinson's syndrome walk with their knees wide apart as the severity increases. The more you spread your knees, the more your ankles point outward, the more you bend your knees, and the slower you walk.
  • five detection elements that can recognize characteristics such as knee, ankle, and speed are designated and calculated.
  • the five calculated detection elements of the Parkinson's syndrome patient (p) are compared with the detection elements of other Parkinson's syndrome patients to determine whether they are higher than a preset standard or not, thereby determining the severity of the Parkinson's syndrome patient (p).

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Surgery (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Physics & Mathematics (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Physiology (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Neurology (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • Neurosurgery (AREA)
  • Developmental Disabilities (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The present invention relates to a system and method for determining the severe stage of Parkinson's syndrome on the basis of gait data. More specifically, the present invention relates to a system and method for determining the severe stage of Parkinson's syndrome on the basis of gait data, wherein the system includes a photographing unit and a server. In an embodiment of the present invention, the severe stage of the Parkinson's syndrome patient can be further accurately determined using five detection elements calculated using the characteristics of patients with Parkinson's syndrome, walking with his/her knees apart.

Description

보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법 System and method for determining the severity of Parkinson's syndrome based on gait data
본 발명은 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에 관한 발명으로 구체적으로 촬영부와 서버를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에 관한 발명이다.The present invention relates to a system and method for determining the severity of Parkinson's syndrome based on gait data, and more particularly, to a system and method for determining the severity of Parkinson's syndrome based on gait data including a photographing unit and a server.
파킨슨병은 뇌에서 도파민이라는 신경전달 물질을 분비하는 신경 세포들이 죽어감으로써 도파민 부족으로 인해 느린 운동, 정지 시 떨림, 근육 강직, 질질 끌며 걷기, 굽은 자세와 같은 여러 증세를 나타내는 신경퇴행성 질환이다. 반면에 파킨슨 증후군은 파킨슨 복합체증후군이라고도 불리며, 파킨슨병 증상에 다른 증상들이 복합적으로 있다는 의미이기도 하다.Parkinson's disease is a neurodegenerative disease characterized by the death of nerve cells that secrete a neurotransmitter called dopamine in the brain. On the other hand, Parkinson's syndrome is also called Parkinson's complex syndrome, which means that the symptoms of Parkinson's disease are combined with other symptoms.
또한, 약물 치료가 가능한 파킨슨병과 달리 약물치료에도 반응이 좋지 않고, 병의 진행도 빨라서 전반적인 경과가 파킨슨 병과는 다르다.In addition, unlike Parkinson's disease, which can be treated with drugs, the overall course is different from Parkinson's disease because it does not respond well to drug treatment and the disease progresses quickly.
따라서, 빠른 약물치료 및 수술치료를 위해 파킨슨 증후군 환자의 중증 단계를 판별해야 할 필요성이 있다.Therefore, there is a need to determine the severity of Parkinson's syndrome patients for rapid drug treatment and surgical treatment.
이와 관련하여 한국 특허 출원번호 제10-2017-0122419호에서는 파킨슨병 환자에서 보행동결(freezing of gait)의 정량적 측정을 위한 장치 및 측정방법에 대해서 게시하고 있지만, 상기 기술은 파킨슨 병 환자의 보행 동결을 측정하기 위해, 환자의 발목에 장치를 부착하고 생활해야한다는 번거로움도 있을뿐더러, 파킨슨 병을 포함하고, 다른 증상까지 포함하여 병의 진행도도 빠른 파킨슨 증후군 환자들의 중증 단계를 판별하지 못한다는 문제점이 있다.In this regard, Korean Patent Application No. 10-2017-0122419 discloses an apparatus and a measurement method for quantitative measurement of freezing of gait in Parkinson's disease patients, but the technique is described in Parkinson's disease patient's gait freezing. In addition to the inconvenience of having to attach the device to the patient's ankle to measure There is a problem.
따라서 상술된 문제점을 해결하기 위한 기술이 필요하게 되었다.Therefore, there is a need for a technique for solving the above-mentioned problems.
한편 전술한 배경기술은 발명자가 본 발명의 도출을 위해 보유하고 있었거나, 본 발명의 도출 과정에서 습득한 기술 정보로서, 반드시 본 발명의 출원 전에 일반 공중에게 공개된 공지기술이라 할 수는 없다.On the other hand, the above-mentioned background art is technical information possessed by the inventor for the derivation of the present invention or acquired in the process of derivation of the present invention, and cannot necessarily be said to be a known technique disclosed to the general public prior to the filing of the present invention.
전술한 목적을 해결하기 위해서 본 발명의 일 실시예는 보행자가 전방을 향해 보행할 때, 보행자의 측면을 촬영하는 촬영부 및 상기 촬영부에서 촬영된 사진 데이터를 전달받아 중증단계를 판별하는 서버, 상기 서버는 촬영된 사진 데이터로부터 보행자의 스켈레톤 데이터를 추출하는 추출부,
Figure PCTKR2022005088-appb-img-000001
를 기준으로 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이를 연산하고,
Figure PCTKR2022005088-appb-img-000002
에 기준으로 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이를 연산하는 연산부 및 상기 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이 및 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이로부터 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하는 판별부를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템을 포함할 수 있다.In order to solve the above object, an embodiment of the present invention provides a server for determining a severe stage by receiving the photographing unit for photographing the side of the pedestrian and the photographic data taken by the photographing unit when the pedestrian walks forward, The server is an extractor for extracting the skeleton data of the pedestrian from the photographed photo data,
Figure PCTKR2022005088-appb-img-000001
Calculate the average deviation / pelvic length of the difference between the z-coordinates of each knee of the pedestrian based on
Figure PCTKR2022005088-appb-img-000002
A calculation unit that calculates the average deviation/pelvic length of the difference in the z-coordinate of each ankle of the pedestrian and the average deviation/pelvic length of the difference in the z-coordinate of each knee of the pedestrian and the difference in the z-coordinate of each ankle of the pedestrian based on It may include a gait data-based Parkinson's syndrome severity stage determination system including a determination unit that determines whether the Parkinson's syndrome state of the pedestrian is included within a preset range from the average deviation/pelvic length.
또한, 상기 연산부는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000003
를 기준으로 보행속도/다리길이를 연산하고, 상기 판별부는 상기 보행속도/다리길이로부터 보행자의 파킨슨 증후군 상태 상태가 기 설정된 범위내에 포함되는지 판별하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템을 포함할 수 있다.In addition, the calculation unit from the skeleton data,
Figure PCTKR2022005088-appb-img-000003
Calculating the walking speed / leg length based on the gait data-based Parkinson's syndrome severe stage determination system that determines whether the state of the pedestrian's Parkinson's syndrome is within a preset range from the walking speed / leg length have.
또한, 상기 연산부는 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000004
를 기준으로 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발길이를 연산하고, 상기 판별부는 상기 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발길이로부터 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템을 포함할 수 있다.In addition, the calculation unit from the skeleton data,
Figure PCTKR2022005088-appb-img-000004
Calculates the average foot length of both feet of a walking pedestrian / walker's foot viewed from the side based on It may include a gait data-based Parkinson's syndrome severity stage determination system that determines whether it is included within a preset range.
또한, 상기 연산부는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000005
에 기준으로 보행중인 양 무릎의 굽힘 각 평균을 연산하고, 상기 판별부는 보행중인 양 무릎의 굽힘 각 평균으로부터 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템을 포함할 수 있다.In addition, the calculation unit from the skeleton data,
Figure PCTKR2022005088-appb-img-000005
Calculates the average of the bending angles of both knees while walking, and the determination unit determines whether the Parkinson's syndrome state of the pedestrian is included within a preset range from the average of the bending angles of both knees while walking. may include
또한, 상기 판별부는 상기 연산부에서 연산된 상기 보행속도/다리길이, 상기 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이, 상기 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이, 상기 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발길이 및 상기 보행중인 양 무릎의 굽힘 각 평균의 값인 파킨슨 판별 데이터를 이용하여 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하고, 상기 서버는 상기 파킨슨 판별 데이터를 딥러닝 하는 학습부를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템을 포함할 수 있다.In addition, the determination unit is the average deviation/pelvic length of the difference between the walking speed/leg length calculated by the calculation unit, the z-coordinate of each knee of the pedestrian, and the average deviation/pelvic length of the difference between the z-coordinate of each ankle of the pedestrian , Using the Parkinson's determination data, which is the average foot length of both feet of the walking pedestrian viewed from the side / the pedestrian's foot length, and the average value of the bending of both knees while walking, it is determined whether the pedestrian's Parkinson's syndrome is within a preset range, and , The server may include a gait data-based Parkinson's syndrome severe stage determination system including a learning unit for deep learning the Parkinson's determination data.
또한, 파킨슨 판별 데이터를 딥러닝하는 제1단계, 보행자가 전방을 향해 보행할 때, 보행자의 측면을 촬영하는 제2단계, 촬영된 사진데이터로부터 보행자의 스켈레톤 데이터를 추출하고, 상기 스켈레톤 데이터로부터 상기 파킨슨 판별 데이터로 연산하는 제3단계 및 딥러닝 된 상기 파킨슨 판별데이터에 기초하여 기 설정된 기준보다 높은 지 판별하여 보행자의 파킨슨 증후군을 판별하는 제4단계를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법을 포함할 수 있다.In addition, the first step of deep learning the Parkinson's determination data, the second step of photographing the side of the pedestrian when the pedestrian walks forward, extracting the pedestrian's skeleton data from the photographed photo data, and from the skeleton data A method for determining the severity of Parkinson's syndrome based on gait data, comprising a third step of calculating with Parkinson's discrimination data and a fourth step of determining whether the parkinson's syndrome of a pedestrian is higher than a preset standard based on the deep-learned Parkinson's discrimination data may include
또한, 상기 제3단계는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000006
를 기준으로 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이를 연산하는 단계를 포함하고, 상기 제 4단계는 상기 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법을 포함할 수 있다.In addition, the third step is from the skeleton data,
Figure PCTKR2022005088-appb-img-000006
Comprising the step of calculating the average deviation / pelvic length of the difference of the z-coordinate of each knee of the pedestrian based on It may include a method for determining the severity of Parkinson's syndrome based on gait data including the step of determining whether it is included within the criteria.
또한, 상기 제3단계는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000007
를 기준으로 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이를 연산하는 단계를 더 포함하고, 제 4단계는 상기 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법을 포함할 수 있다.In addition, the third step is from the skeleton data,
Figure PCTKR2022005088-appb-img-000007
Further comprising the step of calculating the average deviation / pelvic length of the difference in the z-coordinate of each ankle of the pedestrian based on It may include a method for determining the severity of Parkinson's syndrome based on gait data further comprising the step of determining whether it is included within the criteria.
또한, 상기 제3단계는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000008
를 기준으로 보행속도/다리길이를 연산하는 단계를 더 포함하고, 상기 제 4단계는 상기 보행속도/다리길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법을 포함할 수 있다.In addition, the third step is from the skeleton data,
Figure PCTKR2022005088-appb-img-000008
The step of calculating the walking speed / leg length based on the gait data further comprising the step of determining whether the fourth step is included within a preset standard from the walking speed / leg length Determination of the severe stage of Parkinson's syndrome methods may be included.
또한, 상기 제3단계는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000009
를 기준으로 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발 길이를 연산하는 단계를 더 포함하고, 상기 제 4단계는 상기 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발 길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법을 포함할 수 있다.In addition, the third step is from the skeleton data,
Figure PCTKR2022005088-appb-img-000009
Further comprising the step of calculating the average foot length of both feet of the walking pedestrian / foot length of the pedestrian viewed from the side based on It may include a method for determining the severity of Parkinson's syndrome based on gait data further comprising the step of determining whether it is included within the preset criteria from the gait data.
또한, 상기 제3단계는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000010
를 기준으로 보행중인 양 무릎의 굽힘 각 평균을 연산하는 단계를 더 포함하고, 상기 제 4단계는 상기 보행중인 양 무릎의 굽힘 각 평균으로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법을 포함할 수 있다.In addition, the third step is from the skeleton data,
Figure PCTKR2022005088-appb-img-000010
Further comprising the step of calculating the average of the bending angle of both knees while walking based on based methods for determining the severity of Parkinson's syndrome.
본 발명의 일실시예는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법을 제공하는 데에 목적이 있다. An embodiment of the present invention aims to provide a system and method for determining the severity of Parkinson's syndrome based on gait data.
본 발명의 일실시예는 파킨슨 증후군 환자의 보행 중인 모습을 촬영하여, 파킨슨 증후군 환자의 걸음걸이만으로도 자율신경계 검사나 뇌 MRI 등 다양한 기능 검사를 안하고 파킨슨 증후군 환자들의 중증 단계를 판별할 수 있다.An embodiment of the present invention can determine the severity of Parkinson's syndrome patients without performing various functional tests, such as autonomic nervous system examination or brain MRI, only with the gait of the Parkinson's syndrome patient by photographing the walking state of the Parkinson's syndrome patient.
또한, 본 발명의 일실시예는 연산된 검출요소 1인 보행자의 각 무릎의 z좌표의 평균편차/골반길이와 연산된 검출요소 2인 보행자의 각 발목의 z좌표의 평균편차/골반길이를 이용하여 파킨슨 증후군 환자의 파킨슨 증후군 중증단계를 판별할 수 있다.In addition, an embodiment of the present invention uses the calculated average deviation/pelvic length of the z-coordinate of each knee of the pedestrian, which is the calculated detection factor 1, and the calculated average deviation/pelvic length of the z-coordinate of each ankle, which is the calculated detection factor 2, of the pedestrian. Thus, it is possible to determine the severity of Parkinson's syndrome in patients with Parkinson's syndrome.
또한, 본 발명의 일실시예는 연산된 검출요소 3인 보행속도/다리길이를 이용하여 파킨슨 증후군 환자의 파킨슨 증후군 중증단계를 판별할 수 있다.In addition, one embodiment of the present invention can determine the Parkinson's syndrome severe stage of the Parkinson's syndrome patient using the calculated detection factor 3, walking speed/leg length.
또한, 본 발명의 일실시예는 연산된 검출요소 4인 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발길이를 이용하여 파킨슨 증후군 환자의 파킨슨 증후군 중증단계를 판별할 수 있다.In addition, an embodiment of the present invention can determine the Parkinson's syndrome severe stage of the Parkinson's syndrome patient using the average foot length/walker's foot length of both feet of a walking pedestrian viewed from the side of the calculated detection element 4 .
또한, 본 발명의 일실시예는 연산된 검출요소 5인 보행중인 양 무릎의 굽힘 각 평균을 이용하여 파킨슨 증후군 환자의 파킨슨 증후군 중증단계를 판별할 수 있다.In addition, according to an embodiment of the present invention, the severity of Parkinson's syndrome of a patient with Parkinson's syndrome can be determined using the average of the bending angles of both knees while walking, which is the calculated detection element 5.
또한, 본 발명의 일실시예는 5가지의 검출요소를 이용하여 보다 파킨슨 증후군 환자의 중증 단계를 더욱 정확하게 판별할 수 있다는 효과가 있다.In addition, an embodiment of the present invention has an effect that it is possible to more accurately determine the severity of the Parkinson's syndrome patient using five detection elements.
또한, 파킨슨 증후군 환자의 중증 단계를 판별하여, 판별된 내용을 환자뿐만 아니라 의료진에게도 전송하여, 파킨슨 증후군 환자의 상태를 빠르게 확인하여 신속한 치료가 가능하다. In addition, by determining the severity of the Parkinson's syndrome patient, the determined content is transmitted not only to the patient but also to the medical staff, so that the condition of the Parkinson's syndrome patient can be quickly checked and prompt treatment is possible.
본 발명에서 얻을 수 있는 효과는 이상에서 언급한 효과들로 제한되지 않으며, 언급하지 않은 또 다른 효과들은 아래의 기재로부터 본 발명이 속하는 분야에서 통상의 지식을 가진 자에게 명확하게 이해될 수 있을 것이다.The effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those of ordinary skill in the art from the following description. .
도 1은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템에 기초하여 전방을 향해 걷고 있는 보행자를 측면에서 촬영하는 상황을 도시한 개념도이다.1 is a conceptual diagram illustrating a situation in which a pedestrian walking forward is photographed from the side based on the gait data-based Parkinson's syndrome severity stage determination system according to an embodiment of the present invention.
도 2는 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템의 구성을 구체적으로 도시화한 도면이다.2 is a diagram specifically illustrating the configuration of a gait data-based Parkinson's syndrome severity stage determination system according to an embodiment of the present invention.
도 3은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 1을 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이다.3 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 1 used in the gait data-based parkinson's syndrome severe stage determination system and method according to an embodiment of the present invention.
도 4는 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 2를 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이다.4 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed in order to calculate the detection element 2 used in the system and method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
도 5는 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소3을 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이다.5 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed in order to calculate the detection element 3 used in the gait data-based parkinson's syndrome severity stage determination system and method according to an embodiment of the present invention.
도 6은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 4를 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이다.6 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed in order to calculate the detection element 4 used in the system and method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
도 7은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 5를 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이다.7 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 5 used in the gait data-based parkinson's syndrome severe stage determination system and method according to an embodiment of the present invention.
도 8은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법에 대한 순서도이다.8 is a flowchart of a method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
아래에서는 첨부한 도면을 참조하여 본 발명이 속하는 기술 분야에서 통상의 지식을 가진 자가 용이하게 실시할 수 있도록 본 발명의 실시 예를 상세히 설명한다. 그러나 본 발명은 여러 가지 상이한 형태로 구현될 수 있으며 여기에서 설명하는 실시 예에 한정되지 않는다. 그리고 도면에서 본 발명을 명확하게 설명하기 위해서 설명과 관계없는 부분은 생략하였으며, 명세서 전체를 통하여 유사한 부분에 대해서는 유사한 도면 부호를 붙였다.Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily carry out the present invention. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.
명세서 전체에서, 어떤 부분이 다른 부분과 “연결”되어 있다고 할 때, 이는 “직접적으로 연결” 되어 있는 경우뿐만 아니라 그 중간에 다른 소자를 두고 “전기적으로 연결”되어 있는 경우도 포함한다. 또한, 어떤 부분이 어떤 구성요소를 “포함” 한다고 할 때, 이는 특별히 반대되는 기재가 없는 한 다른 구성요소를 제외하는 것이 아니라 다른 구성요소를 더 포함할 수 있는 것을 의미한다. Throughout the specification, when a part is "connected" with another part, it includes not only the case of being "directly connected" but also the case of being "electrically connected" with another element in the middle. In addition, when a part "includes" a certain component, this means that other components may be further included, rather than excluding other components, unless otherwise stated.
본 발명에서 사용되는 카메라는 카메라 내부에 ToF(Time of Flight; 비행시간) 모듈이 포함할 수 있다. 카메라가 촬영을 하면 상기 ToF(Time of Flight) 모듈은 피사체를 향해 빛을 발사해, 발사한 빛이 피사체에 부딪힌 후 반사되어 돌아오는 시간을 거리로 계산할 수 있다.The camera used in the present invention may include a Time of Flight (ToF) module inside the camera. When the camera takes a picture, the Time of Flight (ToF) module emits light toward the subject, and calculates a time for the emitted light to be reflected back after hitting the subject as a distance.
본 발명에서 사진데이터란 본 발명에서 사용되는 카메라를 이용하여 촬영된 보행자가 보해중인 영상의 프레임 단위의 장면을 뜻한다. 촬영된 사진데이터에서 보행자의 뼈대의 각 지점을 하나의 포인트로 지정하여 총 31개의 특성을 추출한 것이 스켈레톤 데이터이다. 추출된 스켈레톤 데이터가 포함하고 있는 31개의 특성에는 코, 양쪽 눈, 양쪽 귀, 머리, 목, 팔꿈치, 팔목, 가슴, 허리, 골반, 엉덩이, 무릎, 발 등이 포함되어 있으며, 본 발명의 일실시예는 골반, 양쪽 엉덩이, 양쪽 무릎, 양쪽 발목, 양쪽 발의 지점을 포함한 9개의 특성만 사용된다.In the present invention, photo data refers to a frame-by-frame scene of an image being reported by a pedestrian taken by using the camera used in the present invention. Skeleton data is obtained by extracting a total of 31 characteristics by designating each point of the pedestrian's skeleton as one point from the photographed photo data. 31 characteristics included in the extracted skeleton data include nose, both eyes, both ears, head, neck, elbow, wrist, chest, waist, pelvis, hip, knee, foot, etc. For the example, only nine characteristics are used, including points of the pelvis, both hips, both knees, both ankles, and both feet.
이하에서는 도 1 및 도 2에 기초하여 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템에 대해서 구체적으로 설명하기로 한다. 도1은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템을 개략적으로 도시한 개념도이고, 도2는 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템의 구성을 구체적으로 도시화한 도면이다.Hereinafter, a detailed description will be given of a system for determining the severity of Parkinson's syndrome based on gait data based on FIGS. 1 and 2 . 1 is a conceptual diagram schematically illustrating a gait data-based Parkinson's syndrome severe stage determination system according to an embodiment of the present invention, and FIG. 2 is a configuration of a gait data-based Parkinson's syndrome severe stage determination system according to an embodiment of the present invention It is a drawing showing in detail.
본 발명의 일 실시예인 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템은 정면으로 걷는 파킨슨 증후군 환자의 측면으로 촬영하여 촬영된 데이터를 이용하여 파킨슨 증후군 환자의 중증 단계를 판별하는 시스템이다.The system for determining the severity of Parkinson's syndrome based on gait data, which is an embodiment of the present invention, is a system for determining the severity of a Parkinson's syndrome patient by using the photographed data taken from the side of the Parkinson's syndrome patient walking in front.
본 발명의 일 실시예인 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템은 촬영부(100), 서버(200), 단말(300)을 포함할 수 있다.The gait data-based Parkinson's syndrome severity stage determination system according to an embodiment of the present invention may include a photographing unit 100 , a server 200 , and a terminal 300 .
촬영부(100)은 전방을 향해 보행자의 측면을 촬영하는 카메라를 포함할 수 있다. 서버(200)는 추출부(210), 연산부(220), 학습부(230) 및 판별부(240)를 포함할 수 있다. 단말(300)은 디스플레이부(310)을 포함할 수 있다. 디스플레이부(310)은 파킨슨 증후군 환자(p)의 정보를 화면에 표시하는 디스플레이 패널(display panel)등을 포함할 수 있다. 다만, 이에 한정되지 않고 디스플레이부(310)는 정보의 출력을 지원하는 다양한 구성을 포함할 수 있다. 또한, 촬영부(100), 서버(200), 단말(300)은 DMR과 WiFi, 저에너지블루투스(BLE), 블루투스(Bluetooth), 3G(3rd Generation), 3GPP(3rdGeneration Partnership Project), LTE(Long Term Evolution), LTE-A, 4G(4th Generation), 5G(5th Generation) 중 적어도 하나의 통신 모듈을 통해 외부장치와 무선 통신할 수 있다.The photographing unit 100 may include a camera for photographing the side of the pedestrian toward the front. The server 200 may include an extracting unit 210 , a calculating unit 220 , a learning unit 230 , and a determining unit 240 . The terminal 300 may include a display unit 310 . The display unit 310 may include a display panel that displays information about the Parkinson's syndrome patient p on the screen. However, the present invention is not limited thereto, and the display unit 310 may include various configurations supporting the output of information. In addition, the recording unit 100, the server 200, the terminal 300 is DMR and WiFi, low energy Bluetooth (BLE), Bluetooth (Bluetooth), 3G (3rd Generation), 3GPP (3rd Generation Partnership Project), LTE (Long Term) Evolution), LTE-A, 4G (4th Generation), and 5G (5th Generation) may perform wireless communication with an external device through at least one communication module.
추출부(210)는 촬영부(100)에서 촬영된 사진 데이터에서 보행자의 스켈레톤 데이터를 추출할 수 있다. 연산부(220)는 추출부에서 추출한 보행자의 스켈레톤 데이터에 기초하여 파킨슨 증후군 환자의 중증 단계를 판별하기 위해 사용될 검출요소 5가지를 연산할 수 있다. 이하, 연산 된 5가지의 검출요소는 파킨슨 판별 데이터라 한다. 검출요소 5가지에 대해서는 후술하기로 한다. 학습부(230)는 상기 연산부(230)에서 파킨슨 판별 데이터로부터 딥러닝을 할 수 있고, 판별부(240)는 딥러닝 된 파킨슨 판별 데이터를 바탕으로 기 설정된 기준을 통해 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있다. 본 발명의 일 실시예에 따르면, 보행자(p)는 전방을 향해(x축) 보행을 하면, 보행자의 측면을 카메라(c)를 이용하여 촬영할 수 있다.The extraction unit 210 may extract skeleton data of a pedestrian from the photo data captured by the photographing unit 100 . The calculation unit 220 may calculate five detection elements to be used to determine the severity of the Parkinson's syndrome patient based on the skeleton data of the pedestrian extracted by the extraction unit. Hereinafter, the calculated five detection elements are referred to as Parkinson's discrimination data. The five detection elements will be described later. The learning unit 230 can perform deep learning from the Parkinson's determination data in the calculating unit 230, and the determining unit 240 determines the severe stage of Parkinson's syndrome of a pedestrian through a preset criterion based on the deep-learned Parkinson's determination data. can be discerned. According to an embodiment of the present invention, when the pedestrian p walks forward (x-axis), the side of the pedestrian can be photographed using the camera c.
구체적으로, 파킨슨 증후군 환자(p)가 기 설정된 거리를 정면을 향해 보행을 하면, 파킨슨 증후군 환자(p)의 측면에 위치한 카메라(c)가 파킨슨 증후군 환자의 보행을 촬영한다. 이 촬영된 사진데이터 중에서 파킨슨 증후군 환자(p)의 연산에 사용될 하반신의 스켈레톤 데이터만 추출한다. 추출된 스켈레톤 데이터를 이용하여 서버(200)에서 후술할 파킨슨 판별 데이터를 연산한다. 파킨슨 판별데이터를 연산하는 과정은 후술하기로 한다. 연산 된 파킨슨 판별 데이터는 학습부(230)에서 또 다른 파킨슨 증후군 환자들의 파킨슨 판별데이터와 함께 딥러닝이 된다. 딥러닝된 파킨슨 판별데이터는 판별부(240)에서 기 설정된 기준보다 높은 지 판별하고 파킨슨 증후군 환자(p)의 파킨슨 증후군 중증 단계를 판별한다. 판별된 파킨슨 증후군 환자(p)의 중증 단계는 디스플레이부(310)을 통해 파킨슨 증후군 환자(p)에게 보여지고, 파킨슨 증후군 환자(p)뿐만 아니라 의료진한테도 전달될 수 있다.Specifically, when the Parkinson's syndrome patient (p) walks a predetermined distance to the front, the camera (c) located on the side of the Parkinson's syndrome patient (p) photographs the walking of the Parkinson's syndrome patient. Only the skeleton data of the lower body to be used in the calculation of the Parkinson's syndrome patient (p) is extracted from the photographed photo data. The server 200 calculates Parkinson's determination data, which will be described later, by using the extracted skeleton data. The process of calculating the Parkinson's discrimination data will be described later. The calculated Parkinson's discrimination data becomes deep learning together with Parkinson's discrimination data of other Parkinson's syndrome patients in the learning unit 230 . The deep-learning Parkinson's discriminant data determines whether the discriminant 240 is higher than a preset standard and determines the Parkinson's syndrome severe stage of the Parkinson's syndrome patient (p). The determined severe stage of the Parkinson's syndrome patient (p) may be displayed to the Parkinson's syndrome patient (p) through the display unit 310, and may be transmitted to the medical staff as well as the Parkinson's syndrome patient (p).
이하에서는 도 3내지 도 7에 기초하여 파킨슨 판별 데이터에 대해서 구체적으로 설명하기로 한다. 도 3은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 1을 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이고, 도 4는 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 2를 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이고, 도5는 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소3을 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이고, 도 6은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 4를 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이고, 도 7은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템 및 방법에서 사용되는 검출요소 5를 연산하기 위해 보행자를 촬영하는 상황을 개략적으로 도시한 개념도이다.Hereinafter, the Parkinson's discrimination data will be described in detail based on FIGS. 3 to 7 . 3 is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 1 used in the gait data-based parkinson's syndrome severe stage determination system and method according to an embodiment of the present invention, and FIG. 4 is the present invention. It is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate a detection element 2 used in a system and method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention, Figure 5 is an embodiment of the present invention It is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 3 used in the gait data-based Parkinson's syndrome severe stage determination system and method according to the present invention, and FIG. It is a conceptual diagram schematically illustrating a situation in which a pedestrian is photographed to calculate the detection element 4 used in the system and method for determining the severe stage of the syndrome, and FIG. 7 is a system for determining the stage of severe Parkinson's syndrome based on gait data according to an embodiment of the present invention. And it is a conceptual diagram schematically illustrating a situation of photographing a pedestrian in order to calculate the detection element 5 used in the method.
검출요소 1은 보행자의 각 무릎의 z 좌표의 차이의 평균편차/골반길이이다. 검출요소1을 연산하는 식은
Figure PCTKR2022005088-appb-img-000011
이다.
Figure PCTKR2022005088-appb-img-000012
에서
Figure PCTKR2022005088-appb-img-000013
는 왼쪽 무릎의 z좌표를,
Figure PCTKR2022005088-appb-img-000014
는 오른쪽 무릎의 z좌표를 의미한다.
Figure PCTKR2022005088-appb-img-000015
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다.
Figure PCTKR2022005088-appb-img-000016
는 촬영되는 보행자(p)의 골반 길이이다. 보행자(p)의 골반길이로 보행중인 보행자(p)의 각 무릎의 z좌표의 차의 평균편차를 나누는 이유는 촬영되는 보행자(p)마다 키가 커서 보행 속도의 차이가 나기도 하고, 골반이 넓은 것에 의해 기본적으로 무릎이 벌어지는 것에 차이가 나기 때문에 표준화하기 위해서이다. 또한, 상기 검출요소 1을 연산하는 식에서 보행자(p)의 각 무릎의 z좌표만 사용하는 이유는 x좌표는 보행자(p)가 전진하거나 후진하는 것으로 바뀌는 값으로 보행중인 무릎의 특징을 짓기에는 어렵다고 판단하여 포함되지 않았고, 각 무릎의 y 좌표 또한 보행자(p)의 신체좌표가 위 아래로 움직일 때 변화하는 값이므로 x 좌표와 마찬가지로 포함되지 않는다. 파킨슨 증후군 환자의 보행에서 일반인의 보행과 다른 점은 무릎을 벌리고 걷는다는 특징이 있다. 무릎을 벌리고 걷는 특징을 이용하기 위해서 도2의 표시된 3차원 좌표계에서 z좌표를 이용해야 한다. 앞에서 서술했듯이, x좌표는 보행자가 앞 뒤로 걸을 때 바뀌는 값이며, y좌표는 보행자가 위 아래 움직일 때 변화하는 값이므로 z좌표가 카메라를 기준으로 하여금 가깝거나 먼 거리를 나타내는 값이 된다. 따라서 전면을 향해 보행하는 보행자를 측면에서 촬영하기 때문에 무릎의 벌어지는 값은 z좌표로 나타낼 수 있는 것이다. 따라서 보행자(p)의 각 무릎의 z좌표를 이용하는 것이다. Detection element 1 is the average deviation/pelvic length of the difference between the z-coordinates of each knee of the pedestrian. The expression for calculating detection element 1 is
Figure PCTKR2022005088-appb-img-000011
to be.
Figure PCTKR2022005088-appb-img-000012
at
Figure PCTKR2022005088-appb-img-000013
is the z-coordinate of the left knee,
Figure PCTKR2022005088-appb-img-000014
is the z-coordinate of the right knee.
Figure PCTKR2022005088-appb-img-000015
denotes the time the pedestrian walked the preset distance.
Figure PCTKR2022005088-appb-img-000016
is the pelvic length of the pedestrian (p) being photographed. The reason for dividing the average deviation of the difference in the z-coordinate of each knee of the pedestrian (p) by the pelvic length of the pedestrian (p) is that each pedestrian (p) is tall, so there is a difference in walking speed, and the pelvis is wide. This is to standardize because there is a difference in the way the knee is opened by default. In addition, the reason that only the z-coordinate of each knee of the pedestrian p is used in the equation for calculating the detection element 1 is that the x-coordinate is a value that changes to the pedestrian p moving forward or backward, making it difficult to characterize the walking knee. It was not included by judgment, and the y-coordinate of each knee is also a value that changes when the body coordinates of the pedestrian (p) move up and down, so it is not included like the x-coordinate. The gait of Parkinson's syndrome patients differs from that of ordinary people in that they walk with their knees apart. In order to use the characteristic of walking with the knees apart, the z-coordinate must be used in the three-dimensional coordinate system shown in FIG. 2 . As mentioned earlier, the x-coordinate is a value that changes when the pedestrian walks back and forth, and the y-coordinate is a value that changes when the pedestrian moves up and down. Therefore, since a pedestrian walking toward the front is photographed from the side, the width of the knee can be expressed as a z-coordinate. Therefore, the z-coordinate of each knee of the pedestrian (p) is used.
검출요소2는 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이이다. 검출요소 2를 연산하는 식은
Figure PCTKR2022005088-appb-img-000017
이다. 검출요소 1과 비슷하지만 다른 점은 검출요소 1은 보행자(p)의 무릎의 z좌표를 사용하는 식이고, 검출요소 2는 보행자(p)의 발목의 z 좌표를 사용한다. 상기 식의
Figure PCTKR2022005088-appb-img-000018
에서
Figure PCTKR2022005088-appb-img-000019
는 왼쪽 발목의 z좌표를
Figure PCTKR2022005088-appb-img-000020
는 오른쪽 발목의 z좌표를 의미한다.
Figure PCTKR2022005088-appb-img-000021
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다. 상기 식의
Figure PCTKR2022005088-appb-img-000022
는 촬영되는 보행자(p)의 골반 길이이다. 검출요소 2를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000023
로
Figure PCTKR2022005088-appb-img-000024
를 나누는 이유도 검출요소 1를 연산하는 식에서 보행중인 보행자(p)의 각 무릎의 z좌표의 평균편차를 보행자(p)의 골반 길이로 나누는 이유와 동일하게 촬영되는 보행자 마다 키가 커서 보행 속도의 차이가 나고, 골반이 넓은 것에 의해 기본적으로 무릎이 벌어져 양 발목 사이가 멀어지는 것에서 차이가 나기 때문에 표준화하기 위해서이다. 앞서 서술했듯이 파킨슨 증후군 환자와 일반인과의 보행에서 다른 점으로는 무릎을 벌리고 걷는 특징이 있다. 따라서 무릎이 벌리고 걸어, 양 발목의 사이도 멀어진다. 보행자(p)의 양 발목 사이의 거리를 측정하기 위해서는 보행자(p)의 측면에서 촬영하는 카메라(c)를 기준으로 원근을 나타낼 수 있는 z좌표를 이용해야 한다.The detection factor 2 is the average deviation/pelvic length of the difference between the z-coordinates of each ankle of the pedestrian. The expression for calculating detection factor 2 is
Figure PCTKR2022005088-appb-img-000017
to be. It is similar to detection element 1, but with the difference that detection element 1 uses the z-coordinate of the pedestrian (p)'s knee, and detection element 2 uses the z-coordinate of the pedestrian's (p) ankle. of the above formula
Figure PCTKR2022005088-appb-img-000018
at
Figure PCTKR2022005088-appb-img-000019
is the z-coordinate of the left ankle
Figure PCTKR2022005088-appb-img-000020
is the z-coordinate of the right ankle.
Figure PCTKR2022005088-appb-img-000021
denotes the time the pedestrian walked the preset distance. of the above formula
Figure PCTKR2022005088-appb-img-000022
is the pelvic length of the pedestrian (p) being photographed. In the equation for calculating detection factor 2,
Figure PCTKR2022005088-appb-img-000023
as
Figure PCTKR2022005088-appb-img-000024
The reason for dividing is the same as the reason for dividing the average deviation of the z-coordinate of each knee of the walking pedestrian (p) by the pelvis length of the walking pedestrian (p) in the equation for calculating the detection factor 1. This is to standardize because there is a difference, and the difference is in the fact that the knee is basically widened and the distance between the ankles is farther due to the wide pelvis. As mentioned earlier, the difference between walking with patients with Parkinson's syndrome and the general public is the characteristic of walking with the knees apart. As a result, the knees are spread apart, and the distance between the ankles is also farther away. In order to measure the distance between the ankles of the pedestrian (p), the z-coordinate that can represent perspective should be used based on the camera (c) photographed from the side of the pedestrian (p).
검출요소 3은 보행속도/다리길이이다. 검출요소 3을 연산하는 식은
Figure PCTKR2022005088-appb-img-000025
이다.
Figure PCTKR2022005088-appb-img-000026
에서
Figure PCTKR2022005088-appb-img-000027
는 보행자(p)의 허리의 x 좌표를 의미하고,
Figure PCTKR2022005088-appb-img-000028
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000029
는 보행자(p)의 다리 길이를 의미한다. 앞서 서술했듯이, x좌표는 보행자가 앞 뒤로 이동할 때 변화하는 값으로 보행자가 시작지점에서 끝지점까지 보행한 거리를 나타내는 좌표가 되는 것이다. 따라서 검출요소 3을 연산하는 식
Figure PCTKR2022005088-appb-img-000030
에서
Figure PCTKR2022005088-appb-img-000031
는 보행자가 촬영 중 보행한 총 거리를 보행한 시간으로 나눈것이므로 보행자가 보행한 속도를 의미한다.
Figure PCTKR2022005088-appb-img-000032
를
Figure PCTKR2022005088-appb-img-000033
로 나누는 이유는 다리 길이에 따라서 보폭에 크기가 달라져 같은 거리를 걷더라도 속도가 달라지기 때문에 표준화하기 위해서이다. The detection factor 3 is walking speed/leg length. The expression for calculating detection element 3 is
Figure PCTKR2022005088-appb-img-000025
to be.
Figure PCTKR2022005088-appb-img-000026
at
Figure PCTKR2022005088-appb-img-000027
is the x-coordinate of the waist of the pedestrian (p),
Figure PCTKR2022005088-appb-img-000028
denotes the time the pedestrian walked the preset distance. In addition,
Figure PCTKR2022005088-appb-img-000029
is the leg length of the pedestrian (p). As described above, the x-coordinate is a value that changes when the pedestrian moves back and forth, and becomes a coordinate indicating the distance the pedestrian walked from the starting point to the ending point. Therefore, the expression for calculating the detection factor 3
Figure PCTKR2022005088-appb-img-000030
at
Figure PCTKR2022005088-appb-img-000031
is the total distance traveled by the pedestrian during the recording divided by the time walked, so it means the speed at which the pedestrian walked.
Figure PCTKR2022005088-appb-img-000032
cast
Figure PCTKR2022005088-appb-img-000033
The reason for dividing by ' is to standardize because the size of the stride varies according to the length of the leg and the speed changes even when walking the same distance.
검출요소 4는 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발길이이다. 검출요소 4를 연산하는 식은
Figure PCTKR2022005088-appb-img-000034
이다. 검출요소 4를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000035
는 왼발의 x좌표이고
Figure PCTKR2022005088-appb-img-000036
는 왼 발목의 x좌표를 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000037
는 오른발의 x좌표를 의미하고
Figure PCTKR2022005088-appb-img-000038
는 오른발목의 x좌표를 의미한다. 또한, 상기 검출요소 4를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000039
는 보행자의 발길이를 의미하고,
Figure PCTKR2022005088-appb-img-000040
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다. 정상적인 일반인이 걸었을 때 측면에서 바라본 발길이는 본인의 발길이와 동일하지만 파킨슨 증후군 환자는 무릎을 벌리고 걷기 때문에 발목이 밖을 향하여 걷는다 따라서 본인의 발길이보다 측면에서 바라본 발길이가 더 짧을 수 있다. 검출 요소 4를 구하는 식에서
Figure PCTKR2022005088-appb-img-000041
는, 측면에서 바라본 양 발의 발길이를 구하여 2로 나눈 것이다. 구체적으로, 측면에서 바라본 왼발의 발길이는
Figure PCTKR2022005088-appb-img-000042
로 연산 할 수 있다. 측면에서 바라본 보행중인 보행자의 왼발의 앞 지점(
Figure PCTKR2022005088-appb-img-000043
)에서 왼발의 발목 지점(
Figure PCTKR2022005088-appb-img-000044
)을 빼면 측면에서 바라본 왼발의 발길이가 나온다. 또한, 측면에서 바라본 오른발의 발길이는
Figure PCTKR2022005088-appb-img-000045
로 연산할 수 있다. 측면에서 바라본 보행중인 보행자의 오른발의 앞 지점(
Figure PCTKR2022005088-appb-img-000046
)에서 오른발의 발목 지점(
Figure PCTKR2022005088-appb-img-000047
)을 빼면 측면에서 바라본 오른발의 발길이가 나온다.The detection factor 4 is the average foot length / foot length of both feet of a walking pedestrian when viewed from the side. The expression for calculating detection factor 4 is
Figure PCTKR2022005088-appb-img-000034
to be. In the equation for calculating detection factor 4,
Figure PCTKR2022005088-appb-img-000035
is the x-coordinate of the left foot
Figure PCTKR2022005088-appb-img-000036
is the x-coordinate of the left ankle. In addition,
Figure PCTKR2022005088-appb-img-000037
is the x-coordinate of the right foot
Figure PCTKR2022005088-appb-img-000038
is the x-coordinate of the right ankle. In addition, in the formula for calculating the detection element 4,
Figure PCTKR2022005088-appb-img-000039
means the footsteps of pedestrians,
Figure PCTKR2022005088-appb-img-000040
denotes the time the pedestrian walked the preset distance. When a normal person walks, the foot length viewed from the side is the same as the person's foot length, but the foot length viewed from the side may be shorter than the foot length of the person with Parkinson's syndrome. In the equation for finding the detection factor 4,
Figure PCTKR2022005088-appb-img-000041
, is the foot length of both feet viewed from the side and divided by 2. Specifically, the foot length of the left foot viewed from the side is
Figure PCTKR2022005088-appb-img-000042
can be calculated with The front point of the pedestrian's left foot viewed from the side (
Figure PCTKR2022005088-appb-img-000043
) at the ankle point of the left foot (
Figure PCTKR2022005088-appb-img-000044
) to get the foot length of the left foot viewed from the side. In addition, the foot length of the right foot viewed from the side is
Figure PCTKR2022005088-appb-img-000045
can be calculated with The front point of the pedestrian's right foot viewed from the side (
Figure PCTKR2022005088-appb-img-000046
) at the ankle point of the right foot (
Figure PCTKR2022005088-appb-img-000047
) to get the foot length of the right foot viewed from the side.
검출요소 5는 보행중인 양 무릎의 굽힘 각 평균이다. 검출요소 5를 연산하는 식은
Figure PCTKR2022005088-appb-img-000048
이다. 상기 검출요소 5를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000049
는 왼쪽 엉덩이, 왼 무릎, 왼 발목의 x좌표와 y좌표로 삼각형 세변의 길이를 계산 후 cos법칙을 이용하여 연산 한 왼 다리의 무릎의 굽힘 각을 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000050
는 오른쪽 엉덩이, 오른 무릎, 오른 발목의 x좌표와 y좌표로 삼각형 세변의 길이를 계산 후 cos법칙을 이용하여 연산한 오른다리의 무릎의 굽힘 각을 의미한다.
Figure PCTKR2022005088-appb-img-000051
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다.The detection element 5 is the average flexion angle of both knees during walking. The expression for calculating detection element 5 is
Figure PCTKR2022005088-appb-img-000048
to be. In the formula for calculating the detection element 5,
Figure PCTKR2022005088-appb-img-000049
is the bending angle of the knee of the left leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the left hip, left knee, and left ankle. In addition,
Figure PCTKR2022005088-appb-img-000050
is the bending angle of the knee of the right leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the right hip, right knee, and right ankle.
Figure PCTKR2022005088-appb-img-000051
denotes the time the pedestrian walked the preset distance.
파킨슨 증후군 환자들은 중증 단계가 심해질수록 무릎을 더 많이 벌려 걷게 된다. 무릎을 더 많이 벌리면 벌릴수록, 발목도 더 많이 밖을 향하게 되고, 무릎도 많이 굽히게 된다. 보행속도도 많이 느려 진다. 앞서 설명한 검출요소 5가지는 이러한 특징들을 이용하여 파킨슨 증후군 환자들의 중증단계를 판별하는 지표로 이용된다.People with Parkinson's syndrome tend to walk with more spread out of their knees as the severity increases. The more you spread your knees, the more your ankles point out and the more your knees bend. Walking speed is also very slow. The five detection elements described above are used as indicators to determine the severity of Parkinson's syndrome patients using these characteristics.
이하에서는 도8에 기초하여 보행데이터 기반 파킨슨 중증 단계 판별 방법에 대해서 구체적으로 설명하기로 한다. 도 8은 본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법에 대한 순서도이다.Hereinafter, a detailed description will be given of a method for determining the severity of Parkinson's disease based on gait data based on FIG. 8 . 8 is a flowchart of a method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention.
본 발명의 일실시예에 따른 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법은 파킨슨 판별 데이터를 딥러닝하는 제1단계(s100), 보행자가 전방을 향해 보행할 때, 보행자의 측면을 촬영하는 제2단계(s200), 촬영된 사진데이터로부터 보행자의 스켈레톤 데이터를 추출하고, 상기 스켈레톤 데이터로부터 상기 파킨슨 판별데이터로 연산하는 제 3단계(s300), 딥러닝 된 상기 파킨슨 판별 데이터에 기초하여 기 설정된 기준보다 높은 지 판별하여 보행자의 파킨슨 증후군 중증단계를 판별하는 제 4단계(s400)를 포함 할 수 있다.The method for determining the severity of Parkinson's syndrome based on gait data according to an embodiment of the present invention includes a first step (s100) of deep learning the Parkinson's determination data, a second step of photographing the side of the pedestrian when the pedestrian walks forward (s200), a third step (s300) of extracting skeleton data of a pedestrian from the photographed photo data, and calculating the parkinson's discrimination data from the skeleton data (s300), based on the deep learning Parkinson's discrimination data higher than a preset standard It may include a fourth step (s400) of determining the severity of the pedestrian's Parkinson's syndrome by determining the
파킨슨 판별 데이터를 딥러닝하는 제1단계(s100)는 중증 단계를 판별할 파킨슨 증후군 환자의 보행을 촬영하기 전, 다른 파킨슨 증후군 환자의 파킨슨 판별 데이터를 미리 딥러닝 할 수 있는 단계이다. 예를 들어, 미리 다른 파킨슨 증후군 환자들의 파킨슨 판별데이터가 딥러닝이 되어있다. 중증 단계를 판별할 파킨슨 증후군 환자는 본 발명의 기 설정된 거리를 보행하고, 이를 측면에서 카메라로 파킨슨 증후군 환자를 촬영한다. 촬영한 데이터와 미리 딥러닝된 파킨슨 판별데이터로부터 파킨슨 증후군 환자의 파킨슨 중증단계를 더욱 빠르게 판별할 수 있다.The first step (s100) of deep learning the Parkinson's identification data is a step in which the Parkinson's identification data of other Parkinson's syndrome patients can be deep-learned before photographing the gait of the Parkinson's syndrome patient to determine the severe stage. For example, the Parkinson's discrimination data of other Parkinson's syndrome patients is deep learning. The Parkinson's syndrome patient to determine the severe stage walks the preset distance of the present invention, and photographs the Parkinson's syndrome patient with a camera from the side. It is possible to more quickly identify the Parkinson's severe stage of Parkinson's syndrome patients from the captured data and the deep-learned Parkinson's identification data.
보행자가 전방을 향해 보행할 때, 보행자의 측면을 촬영하는 제2단계(s200)은 보행자가 기 설정된 거리를 보행하면, ToF모듈을 포함하고 있는 카메라로 보행자의 측면을 촬영할 수 있는 단계이다.When the pedestrian walks forward, the second step (s200) of photographing the side of the pedestrian is a step in which the side of the pedestrian can be photographed with a camera including a ToF module when the pedestrian walks a preset distance.
촬영된 사진데이터로부터 보행자의 스켈레톤 데이터를 추출하고, 상기 스켈레톤 데이터로부터 파킨슨 판별 데이터로 연산하는 제3단계(s300)는 촬영된 사진데이터 중 보행자의 스켈레톤 데이터를 추출하는 단계(s310)를 포함 할 수 있다. 또한, 촬영된 사진데이터로부터 보행자의 스켈레톤 데이터를 추출하고, 상기 스켈레톤 데이터로부터 파킨슨 판별 데이터로 연산하는 제3단계(s300)는 검출요소 1을 연산하는 단계(s321), 검출요소 2를 연산하는 단계(s322), 검출요소 3을 연산하는 단계(s323), 검출요소 4를 연산하는 단계(s324) 및 검출요소 5를 연산하는 단계(s325)를 포함 할 수 있다. The third step (s300) of extracting the pedestrian's skeleton data from the photographed photo data and calculating the parkinson's determination data from the skeleton data may include the step (s310) of extracting the pedestrian's skeleton data from the photographed photograph data. have. In addition, the third step (s300) of extracting the skeleton data of the pedestrian from the photographed photo data and calculating the Parkinson's discrimination data from the skeleton data is the step of calculating the detection element 1 (s321), the steps of calculating the detection element 2 (s322), calculating the detection element 3 (s323), calculating the detection element 4 (s324), and calculating the detection element 5 (s325) may include.
검출요소 1을 연산하는 단계(s321)는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000052
를 기준으로 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이를 연산하는 단계이다. 검출요소 1을 연산하는 식에서
Figure PCTKR2022005088-appb-img-000053
는 왼 무릎의 z좌표를 의미하고
Figure PCTKR2022005088-appb-img-000054
는 오른 무릎의 z좌표를 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000055
는 보행자의 골반길이를 의미한다. 보행자의 골반길이로 보행중인 보행자의 각 무릎의 z좌표의 차의 평균편차를 나누는 이유는 촬영되는 보행자마다 키가 커서 보행 속도의 차이가 나기도 하고, 골반이 넓은 것에 의해 기본적으로 무릎이 벌어지는 것에 차이가 나기 때문에 표준화하기 위해서이다. 파킨슨 증후군 환자는 무릎 및 허리 등의 통증으로 인하여 보행할 때, 무릎을 벌리고 걷는다. 일반인들은 보행할 때 무릎이 벌려 걷지 않기 때문에 보행하는 일반인의 양 무릎의 z좌표를 측정하여 검출요소 1을 연산하면 양 무릎의 차가 일정할 것이다. 하지만 파킨슨 증후군 환자의 보행을 촬영하면 무릎을 계속 벌리고 걷고, 통증 때문에 걸음걸이가 일정하지 않는다. 따라서 보행중인 보행자의 무릎의 z좌표를 이용하여 연산된 검출요소1에서 보행자의 골반길이 대비 보행자의 무릎의 벌림 정도를 알 수 있다.The step of calculating the detection element 1 (s321) is from the skeleton data,
Figure PCTKR2022005088-appb-img-000052
This is a step of calculating the average deviation/pelvic length of the difference between the z-coordinates of each knee of the pedestrian based on . In the equation for calculating detection element 1,
Figure PCTKR2022005088-appb-img-000053
is the z-coordinate of the left knee,
Figure PCTKR2022005088-appb-img-000054
is the z-coordinate of the right knee. In addition,
Figure PCTKR2022005088-appb-img-000055
is the length of the pedestrian's pelvis. The reason for dividing the average deviation of the difference in the z-coordinate of each knee of a walking pedestrian by the pedestrian's pelvis length is that there is a difference in walking speed due to the height of each pedestrian being photographed. This is to standardize because Parkinson's syndrome patients walk with their knees apart when walking due to pain in the knee and back. Since ordinary people do not walk with their knees spread apart when walking, if the detection factor 1 is calculated by measuring the z-coordinates of both knees of a walking ordinary person, the difference between the two knees will be constant. However, when photographing the gait of a patient with Parkinson's syndrome, he walks with his knee wide open, and the gait is not consistent due to pain. Therefore, it is possible to know the degree of divergence of the pedestrian's knee compared to the pedestrian's pelvis length in the detection element 1 calculated using the z-coordinate of the walking pedestrian's knee.
후술할 본 발명의 보행자의 파킨슨 증후군 중증 단계를 판별하는 제4단계(s400)는 검출요소 1을 연산하는 단계(s321)에서 연산된 검출요소 1을 이용하여 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있다. 일반인의 보행과는 다르게 파킨슨 증후군 환자의 보행은 무릎을 벌리고 걷는 특징을 가지고 있다. 보행중인 보행자의 무릎의 z좌표를 이용하여 연산된 검출 요소 1은 골반길이 대비 무릎 사이의 벌림 정도를 내포하고 있다. 따라서, 검출요소 1에 내포되어 있는 골반길이 대비 무릎 사이의 벌림 정도를 가지고 보행자의 파킨슨 증후군 중증단계(s400)에서 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있는 것이다.In the fourth step (s400) of determining the severe stage of Parkinson's syndrome of the pedestrian of the present invention to be described later, the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 1 calculated in the step (s321) of calculating the detection element 1. have. Unlike the gait of the general public, the gait of patients with Parkinson's syndrome has the characteristic of walking with the knees spread apart. The detection element 1 calculated using the z-coordinate of the walking pedestrian's knee contains the degree of divergence between the knees compared to the pelvic length. Therefore, it is possible to determine the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the degree of divergence between the knees compared to the pelvic length contained in the detection element 1.
검출요소 2를 연산하는 단계(s322) 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000056
를 기준으로 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이를 연산하는 단계이다. 검출요소 2를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000057
는 왼 발목의 z좌표를 의미하고
Figure PCTKR2022005088-appb-img-000058
는 오른 발목의 z좌표를 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000059
는 보행자의 골반길이를 의미한다.
Figure PCTKR2022005088-appb-img-000060
로
Figure PCTKR2022005088-appb-img-000061
를 나누는 이유도 검출요소 1를 연산하는 식에서 보행중인 보행자의 각 무릎의 z좌표의 평균편차를 보행자의 골반 길이로 나누는 이유와 동일하게 촬영되는 보행자마다 키가 커서 보행 속도의 차이가 나고, 골반이 넓은 것에 의해 기본적으로 무릎이 벌어져 양 발목 사이가 멀어지는 것에서 차이가 나기 때문에 표준화하기 위해서이다. 앞서 서술했듯이 파킨슨 증후군 환자와 일반인과의 보행에서 다른 점으로는 무릎을 벌리고 걷는 특징이 있다. 따라서 무릎이 벌리고 걸어, 양 발목의 사이도 멀어진다. 따라서 보행중인 보행자의 발목의 z좌표를 이용하여 연산된 검출요소2에서 보행자의 골반길이 대비 보행자의 발목 사이의 벌림 정도를 알 수 있다.Calculating the detection element 2 (s322) From the skeleton data,
Figure PCTKR2022005088-appb-img-000056
This is a step of calculating the average deviation/pelvic length of the difference between the z-coordinates of each ankle of the pedestrian based on . In the equation for calculating detection factor 2,
Figure PCTKR2022005088-appb-img-000057
is the z-coordinate of the left ankle
Figure PCTKR2022005088-appb-img-000058
is the z-coordinate of the right ankle. In addition,
Figure PCTKR2022005088-appb-img-000059
is the length of the pedestrian's pelvis.
Figure PCTKR2022005088-appb-img-000060
as
Figure PCTKR2022005088-appb-img-000061
The reason for dividing is the same as the reason for dividing the average deviation of the z-coordinate of each knee of a walking pedestrian by the length of the pedestrian's pelvis in the equation for calculating the detection factor 1. This is to standardize because there is a difference in the distance between the ankles because the knee is basically widened by the wide one. As mentioned earlier, the difference between walking with patients with Parkinson's syndrome and the general public is the characteristic of walking with the knees apart. As a result, the knees are spread apart, and the distance between the ankles is also farther away. Therefore, in the detection element 2 calculated using the z-coordinate of the walking pedestrian's ankle, the degree of divergence between the pedestrian's ankles compared to the pedestrian's pelvis length can be known.
후술할 본 발명의 보행자의 파킨슨 증후군 중증 단계를 판별하는 제4단계(s400)는 검출요소 2를 연산하는 단계(s322)에서 연산된 검출요소 2를 이용하여 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있다. 일반인의 보행과는 다르게 파킨슨 증후군 환자의 보행은 무릎을 벌리고 걷는 특징이 있다. 무릎을 벌리고 걷기 때문에 발목 사이의 거리도 멀어진다. 따라서 보행중인 보행자의 발목의 z좌표를 이용하여 연산된 검출 요소 2는 골반길이 대비 발목 사이의 벌림 정도를 내포하고 있다. 따라서, 검출요소 2에 내포되어 있는 골반길이 대비 발목 사이의 벌림 정도를 가지고 보행자의 파킨슨 증후군 중증단계(s400)에서 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있는 것이다.In the fourth step (s400) of determining the severe stage of Parkinson's syndrome of the pedestrian of the present invention to be described later, the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 2 calculated in the step (s322) of calculating the detection element 2. have. Unlike the walking of the general public, patients with Parkinson's syndrome walk with their knees apart. As you walk with your knees apart, the distance between your ankles also increases. Therefore, the detection element 2 calculated using the z-coordinate of the ankle of the walking pedestrian contains the degree of divergence between the ankles compared to the pelvic length. Therefore, it is possible to discriminate the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the degree of divergence between the ankles compared to the pelvic length contained in the detection element 2.
검출요소 3을 연산하는 단계(s323)는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000062
를 기준으로 보행속도/다리길이를 연산하는 단계이다.
Figure PCTKR2022005088-appb-img-000063
에서
Figure PCTKR2022005088-appb-img-000064
는 보행자(p)의 허리의 x 좌표를 의미하고,
Figure PCTKR2022005088-appb-img-000065
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000066
는 보행자(p)의 다리 길이를 의미한다.
Figure PCTKR2022005088-appb-img-000067
를
Figure PCTKR2022005088-appb-img-000068
로 나누는 이유는 다리 길이에 따라서 보폭에 크기가 달라져 같은 거리를 걷더라도 속도가 달라지기 때문에 표준화하기 위해서이다. 앞서 서술했듯이, 파킨슨 증후군 환자는 허리, 무릎 등 통증으로 인하여 무릎을 벌리고 걷는 특징을 가지고 있다. 하지만 무릎을 벌리고 걷는 특징 외에도 파킨슨 증후군 환자는 통증으로 인하여 빠르게 걸을 수가 없다. 그렇기 때문에 보폭이 큰 파킨슨 증후군 환자여도 같은 보폭을 가지고 있는 일반인과는 걷는 속도에서 차이가 날 수 있다.The step of calculating the detection element 3 (s323) is from the skeleton data,
Figure PCTKR2022005088-appb-img-000062
It is a step of calculating walking speed/leg length based on .
Figure PCTKR2022005088-appb-img-000063
at
Figure PCTKR2022005088-appb-img-000064
is the x-coordinate of the waist of the pedestrian (p),
Figure PCTKR2022005088-appb-img-000065
denotes the time the pedestrian walked the preset distance. In addition,
Figure PCTKR2022005088-appb-img-000066
is the leg length of the pedestrian (p).
Figure PCTKR2022005088-appb-img-000067
cast
Figure PCTKR2022005088-appb-img-000068
The reason for dividing by ' is to standardize because the size of the stride varies according to the length of the leg and the speed changes even when walking the same distance. As mentioned earlier, patients with Parkinson's syndrome have the characteristic of walking with their knees apart due to back and knee pain. However, in addition to the characteristic of walking with the knees apart, patients with Parkinson's syndrome cannot walk quickly due to pain. Therefore, even a Parkinson's syndrome patient with a large stride may have a different walking speed than an ordinary person with the same stride.
후술할 본 발명의 보행자의 파킨슨 증후군 중증 단계를 판별하는 제4단계(s400)는 검출요소 3을 연산하는 단계(s323)에서 연산된 검출요소 3을 이용하여 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있다. 일반인의 보행과는 다르게 파킨슨 증후군 환자의 보행은 속도가 느리다는 특징이 있다. 따라서, 검출요소 3에 내포되어 있는 다리길이 대비 보행속도를 가지고 보행자의 파킨슨 증후군 중증단계(s400)에서 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있는 것이다.In the fourth step (s400) of determining the severe stage of Parkinson's syndrome of the pedestrian of the present invention to be described later, the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 3 calculated in the step (s323) of calculating the detection element 3. have. Unlike the gait of the general public, the gait of patients with Parkinson's syndrome is characterized by a slow pace. Therefore, it is possible to discriminate the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the walking speed compared to the leg length contained in the detection element 3 .
검출요소 4를 연산하는 단계(s324)는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000069
를 기준으로 측면에서 바라본 보행중인 보행자의 발길이/보행자의 발 길이를 연산하는 단계이다. 검출요소 4를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000070
는 왼발의 x좌표이고
Figure PCTKR2022005088-appb-img-000071
는 왼 발목의 x좌표를 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000072
는 오른발의 x좌표를 의미하고
Figure PCTKR2022005088-appb-img-000073
는 오른발목의 x좌표를 의미한다. 또한, 상기 검출요소 4를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000074
는 보행자의 발길이를 의미하고,
Figure PCTKR2022005088-appb-img-000075
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다. 정상적인 일반인이 걸었을 때 측면에서 바라본 발길이는 본인의 발길이와 동일하지만 파킨슨 증후군 환자는 무릎을 벌리고 걷기 때문에 발목이 밖을 향하여 걷는다 따라서 본인의 발길이보다 측면에서 바라본 발길이가 더 짧을 수 있다.The step of calculating the detection element 4 (s324) is from the skeleton data,
Figure PCTKR2022005088-appb-img-000069
This is a step of calculating the foot length of the pedestrian / the foot length of the pedestrian while walking as viewed from the side. In the equation for calculating detection factor 4,
Figure PCTKR2022005088-appb-img-000070
is the x-coordinate of the left foot
Figure PCTKR2022005088-appb-img-000071
is the x-coordinate of the left ankle. In addition,
Figure PCTKR2022005088-appb-img-000072
is the x-coordinate of the right foot
Figure PCTKR2022005088-appb-img-000073
is the x-coordinate of the right ankle. In addition, in the formula for calculating the detection element 4,
Figure PCTKR2022005088-appb-img-000074
means the footsteps of pedestrians,
Figure PCTKR2022005088-appb-img-000075
denotes the time the pedestrian walked the preset distance. When a normal person walks, the foot length viewed from the side is the same as the person's foot length, but the foot length viewed from the side may be shorter than the foot length of the person with Parkinson's syndrome.
후술할 본 발명의 보행자의 파킨슨 증후군 중증 단계를 판별하는 제4단계(s400)는 검출요소 4를 연산하는 단계(s324)에서 연산된 검출요소 4를 이용하여 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있다. 일반인의 보행에서 측면에서 바라본 발길이와 일반인의 발길이는 같다. 하지만 파킨슨 증후군 환자의 보행에서 측면에서 바라본 발길이와 파킨슨 증후군 환자의 발길이는 다르다. 이러한 특징을 이용하여 보행할 때 측면에서 바라본 발길이와 보행자의 발길이를 이용하여 검출요소 4를 연산한다. 따라서, 검출요소 4에 내포되어 있는 발길이 대비 보행할 때 측면에서 바라본 발길이를 가지고 보행자의 파킨슨 증후군 중증단계(s400)에서 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있는 것이다.In the fourth step (s400) of determining the severe stage of Parkinson's syndrome of the pedestrian of the present invention to be described later, the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 4 calculated in the step (s324) of calculating the detection element 4. have. The foot length seen from the side in the walking of ordinary people is the same as that of ordinary people. However, the foot length viewed from the side in the gait of the Parkinson's syndrome patient is different from the foot length of the Parkinson's syndrome patient. When walking using these features, the detection element 4 is calculated using the foot length viewed from the side and the foot length of the pedestrian. Therefore, it is possible to determine the pedestrian's Parkinson's syndrome severe stage in the pedestrian's Parkinson's syndrome severe stage (s400) with the foot length viewed from the side when walking compared to the foot length contained in the detection element 4 .
검출요소 5를 연산하는 단계(s325)는 상기 스켈레톤 데이터로부터,
Figure PCTKR2022005088-appb-img-000076
를 기준으로 보행중인 양 무릎 굽힘 각 평균을 연산하는 단계이다. 상기 검출요소 5를 연산하는 식에서
Figure PCTKR2022005088-appb-img-000077
는 왼쪽 엉덩이, 왼 무릎, 왼 발목의 x좌표와 y좌표로 삼각형 세변의 길이를 계산 후 cos법칙을 이용하여 연산 한 왼 다리의 무릎의 굽힘 각을 의미한다. 또한,
Figure PCTKR2022005088-appb-img-000078
는 오른쪽 엉덩이, 오른 무릎, 오른 발목의 x좌표와 y좌표로 삼각형 세변의 길이를 계산 후 cos법칙을 이용하여 연산한 오른다리의 무릎의 굽힘 각을 의미한다.
Figure PCTKR2022005088-appb-img-000079
는 보행자가 기 설정된 거리를 보행한 시간을 의미한다. 앞서 서술했듯이 파킨슨 증후군 환자와 일반인과의 보행에서 다른 점으로는 무릎을 벌리고 걷는 특징이 있다. 따라서 파킨슨 증후군 환자는 중증단계가 심해질수록 무릎을 양쪽으로 더 많이 벌리면서, 무릎도 더 많이 굽히게 된다. 따라서 무릎의 굽힘 각 평균을 이용하여 연산된 검출요소5에서 보행할 때 보행자가 무릎을 굽힌 정도를 알 수 있다. The step of calculating the detection element 5 (s325) is from the skeleton data,
Figure PCTKR2022005088-appb-img-000076
This is a step of calculating the average of each knee bend while walking. In the formula for calculating the detection element 5,
Figure PCTKR2022005088-appb-img-000077
is the bending angle of the knee of the left leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the left hip, left knee, and left ankle. In addition,
Figure PCTKR2022005088-appb-img-000078
is the bending angle of the knee of the right leg calculated using the cos law after calculating the length of three sides of a triangle with the x-coordinate and y-coordinate of the right hip, right knee, and right ankle.
Figure PCTKR2022005088-appb-img-000079
denotes the time the pedestrian walked the preset distance. As mentioned earlier, the difference between walking with patients with Parkinson's syndrome and the general public is the characteristic of walking with the knees apart. Therefore, the more severe the Parkinson's syndrome, the more the knee is spread to both sides and the more the knee is bent. Therefore, it is possible to know the degree to which the pedestrian bends the knee when walking in the detection element 5 calculated using the average of the bending angle of the knee.
후술할 본 발명의 보행자의 파킨슨 증후군 중증 단계를 판별하는 제4단계(s400)는 검출요소 5를 연산하는 단계(s325)에서 연산된 검출요소 5를 이용하여 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있다. 일반인이 보행할 때 무릎을 아예 펴고 걷지 않는다. 하지만, 파킨슨 증후군 환자와 비교했을 때 무릎을 굽힌 각도에서 차이가 난다. 따라서, 검출요소 5에 내포되어 있는 보행할 때 파킨슨 증후군 환자의 엉덩이와 무릎 및 발목이 이루고 있는 각도를 가지고 보행자의 파킨슨 증후군 중증단계(s400)에서 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있는 것이다.In the fourth step (s400) of determining the severe stage of Parkinson's syndrome of the pedestrian of the present invention to be described later, the severe stage of the pedestrian's Parkinson's syndrome can be determined using the detection element 5 calculated in the step (s325) of calculating the detection element 5. have. When a normal person walks, they do not walk with their knees straight. However, compared to patients with Parkinson's syndrome, the knee flexion angle is different. Therefore, with the angle formed by the hip, knee, and ankle of the Parkinson's syndrome patient when walking contained in the detection element 5, the pedestrian's Parkinson's syndrome severe stage can be determined in the pedestrian's Parkinson's syndrome severe stage (s400).
보행자의 파킨슨 증후군 중증단계를 판별하는 제4단계(s400)는 촬영된 사진데이터로부터 보행자의 스켈레톤 데이터를 추출하고, 상기 스켈레톤 데이터로부터 파킨슨 판별 데이터로 연산하는 제 3단계(s300)에서 연산 된 파킨슨 판별 데이터에 기초하여 기 설정된 기준보다 높은 지 판별하여 보행자의 파킨슨 증후군 중증 단계를 판별할 수 있다. 앞서 서술한, 5개의 검출요소는 보행자의 파킨슨 증후군 중증단계를 판별하는 제4단계(s400)에서 보행자의 파킨슨 증후군 중증단계를 판별하는 각각의 지표로도 쓰인다. 또한, 보행자의 파킨슨 증후군 중증단계를 판별하는 제4단계(s400)에서 5개의 검출요소 전부를 이용하여 보행자의 파킨슨 증후군 중증단계를 판별할 수 있다.The fourth step (s400) of determining the severe stage of Parkinson's syndrome of a pedestrian extracts the pedestrian's skeleton data from the photographed photo data, and calculates the parkinson's discrimination data from the skeleton data as the Parkinson's discrimination data calculated in the third step (s300) By determining whether it is higher than a preset criterion based on the data, it is possible to determine the severity of Parkinson's syndrome in the pedestrian. The above-described five detection elements are also used as respective indicators for determining the pedestrian's Parkinson's syndrome severe stage in the fourth step (s400) of determining the pedestrian's Parkinson's syndrome severe stage. In addition, in the fourth step (s400) of determining the pedestrian's Parkinson's syndrome severe stage, all five detection elements may be used to determine the pedestrian's Parkinson's syndrome severe stage.
예를 들어, 파킨슨 증후군 환자(p)가 있다. 파킨슨 증후군 환자(p)는 기 설정된 거리를 전방을 향해 걷는다. 이때, 파킨슨 증후군 환자(p)의 측면을 촬영한다. 이 촬영된 사진데이터에서 파킨슨 증후군 환자(p)의 스켈레톤 데이터를 추출해 내는데, 이때 추출되는 스켈레톤 특성은 하반신에 위치한 9개의 특성만 사용한다. 파킨슨 증후군 환자들은 중증 정도가 심할수록 무릎을 많이 벌리고 걷는다. 무릎을 많이 벌리면 벌릴수록 발목이 밖을 향하게 되고, 무릎을 더 많이 굽히게 되며, 걸음 속도도 느려지게 된다. 촬영된 사진데이터에서 추출한 스켈레톤 데이터에서 무릎과 발목, 속도 등의 특징을 알아볼 수 있는 검출요소 5가지를 지정하여 이를 연산한다. 파킨슨 증후군 환자(p)의 연산 된 검출요소 5가지는 다른 파킨슨 증후군 환자들의 검출요소들과 비교하여 기 설정된 기준 이상보다 높은 지 판별하여 파킨슨 증후군 환자(p)의 중증 단계를 판별한다.For example, there is a patient with Parkinson's syndrome (p). The Parkinson's syndrome patient (p) walks a predetermined distance forward. At this time, the side of the Parkinson's syndrome patient (p) is photographed. Skeleton data of a Parkinson's syndrome patient (p) is extracted from the photographed photo data, and only nine characteristics located in the lower body are used for the extracted skeleton characteristics. Patients with Parkinson's syndrome walk with their knees wide apart as the severity increases. The more you spread your knees, the more your ankles point outward, the more you bend your knees, and the slower you walk. From the skeleton data extracted from the photographed photo data, five detection elements that can recognize characteristics such as knee, ankle, and speed are designated and calculated. The five calculated detection elements of the Parkinson's syndrome patient (p) are compared with the detection elements of other Parkinson's syndrome patients to determine whether they are higher than a preset standard or not, thereby determining the severity of the Parkinson's syndrome patient (p).
전술한 본 발명의 설명은 예시를 위한 것이며, 본 발명이 속하는 기술분야의 통상의 지식을 가진 자는 본 발명의 기술적 사상이나 필수적인 특징을 변경하지 않고서 다른 구체적인 형태로 쉽게 변형이 가능하다는 것을 이해할 수 있을 것이다. 그러므로 이상에서 기술한 실시예들은 모든 면에서 예시적인 것이며 한정적이 아닌 것으로 이해해야만 한다. 예를 들어 단일형으로 설명되어 있는 각 구성 요소는 분산되어 실시될 수도 있으며, 마찬가지로 분산된 것으로 설명되어 있는 구성요소들도 결합된 형태로 실시될 수 있다. The foregoing description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and likewise components described as distributed may be implemented in a combined form.
본 발명의 범위는 상기 상세한 설명보다는 후술하는 특허청구범위에 의하여 나타내어지며, 특허청구범위의 의미 및 범위 그리고 그 균등 개념으로부터 도출되는 모든 변경 또는 변형된 형태가 본 발명의 범위에 포함되는 것으로 해석되어야 한다. The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (11)

  1. 보행자가 전방을 향해 보행할 때, 보행자의 측면을 촬영하는 촬영부; 및When the pedestrian walks toward the front, the photographing unit for photographing the side of the pedestrian; and
    상기 촬영부에서 촬영된 사진 데이터를 전달받아 중증단계를 판별하는 서버;a server for receiving the photo data taken by the photographing unit and determining a severe stage;
    상기 서버는 촬영된 사진 데이터로부터 보행자의 스켈레톤 데이터를 추출하는 추출부;The server includes an extraction unit for extracting the skeleton data of the pedestrian from the photographed photo data;
    Figure PCTKR2022005088-appb-img-000080
    를 기준으로 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이를 연산하고,
    Figure PCTKR2022005088-appb-img-000081
    에 기준으로 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이를 연산하는 연산부; 및
    Figure PCTKR2022005088-appb-img-000080
    Calculate the average deviation / pelvic length of the difference between the z-coordinates of each knee of the pedestrian based on
    Figure PCTKR2022005088-appb-img-000081
    a calculation unit that calculates the average deviation/pelvic length of the difference between the z-coordinates of each ankle of the pedestrian based on ; and
    상기 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이 및 상기 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이로부터 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하는 판별부;를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템A determination unit that determines whether the Parkinson's syndrome state of the pedestrian is included within a preset range from the average deviation/pelvic length of the difference in the z-coordinate of each knee of the pedestrian and the average deviation/pelvic length of the difference in the z-coordinate of each ankle of the pedestrian Parkinson's syndrome severity stage determination system based on gait data including ;
  2. 제1항에 있어서,According to claim 1,
    상기 연산부는,The calculation unit,
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000082
    를 기준으로 보행속도/다리길이를 연산하고, 상기 판별부는 상기 보행속도/다리길이로부터 보행자의 파킨슨 증후군 상태 상태가 기 설정된 범위내에 포함되는지 판별하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템    From the skeleton data,
    Figure PCTKR2022005088-appb-img-000082
    A gait data-based Parkinson's syndrome severe stage determination system that calculates walking speed/leg length based on the
  3. 제2항에 있어서,3. The method of claim 2,
    상기 연산부는,The calculation unit,
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000083
    를 기준으로 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발길이를 연산하고, 상기 판별부는 상기 측면에서 바라본 보행중인 보행자의 양 발의 평균 발길이/보행자의 발길이로부터 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템    From the skeleton data,
    Figure PCTKR2022005088-appb-img-000083
    Calculates the average foot length of both feet of the walking pedestrian / foot length of the pedestrian viewed from the side based on A gait data-based Parkinson's syndrome severity stage determination system that determines whether
  4. 제3항에 있어서,4. The method of claim 3,
    상기 연산부는,The calculation unit,
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000084
    에 기준으로 보행중인 양 무릎의 굽힘 각 평균을 연산하고, 상기 판별부는 상기 보행중인 양 무릎의 굽힘 각 평균으로부터 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템    From the skeleton data,
    Figure PCTKR2022005088-appb-img-000084
    Calculates the average of the bending angles of both knees while walking, and the determination unit determines whether the Parkinson's syndrome state of the pedestrian is included within a preset range from the average of the bending angles of both knees while walking. system
  5. 상기 판별부는, The determining unit,
    상기 연산부에서 연산된 상기 보행속도/다리길이, 상기 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이, 상기 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이, 상기 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발길이 및 상기 보행중인 양 무릎의 굽힘 각 평균의 값인 파킨슨 판별 데이터를 이용하여 보행자의 파킨슨 증후군 상태가 기 설정된 범위내에 포함되는지 판별하고, 상기 서버는 상기 파킨슨 판별 데이터를 딥러닝 하는 학습부;를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 시스템The walking speed/leg length calculated by the calculation unit, the average deviation/pelvic length of the difference between the z-coordinates of each knee of the pedestrian, the average deviation/pelvic length of the difference between the z-coordinates of each ankle of the pedestrian, as viewed from the side It is determined whether the parkinson's syndrome state of the pedestrian is within a preset range using the average foot length of both feet of the walking pedestrian / the foot length of the pedestrian and the Parkinson's determination data, which is the value of each average of the bending of both knees while walking, and the server is A gait data-based Parkinson's syndrome severe stage determination system including a learning unit that deep-learns the Parkinson's identification data
  6. 파킨슨 판별 데이터를 딥러닝하는 제1단계; 보행자가 전방을 향해 보행할 때, 보행자의 측면을 촬영하는 제2단계; 촬영된 사진데이터로부터 보행자의 스켈레톤 데이터를 추출하고, 상기 스켈레톤 데이터로부터 상기 파킨슨 판별 데이터로 연산하는 제3단계; 및 딥러닝 된 상기 파킨슨 판별데이터에 기초하여 기 설정된 기준보다 높은 지 판별하여 보행자의 파킨슨 증후군을 판별하는 제4단계; 를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법A first step of deep learning the Parkinson's discrimination data; A second step of photographing the side of the pedestrian when the pedestrian walks forward; A third step of extracting the skeleton data of the pedestrian from the photographed photo data, and calculating the Parkinson's determination data from the skeleton data; and a fourth step of determining whether a pedestrian's Parkinson's syndrome is higher than a preset standard based on the deep-learning Parkinson's determination data; A method for determining the severity of Parkinson's syndrome based on gait data comprising
  7. 제6항에 있어서,7. The method of claim 6,
    상기 제3단계는,The third step is
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000085
    를 기준으로 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이를 연산하는 단계를 포함하고,     From the skeleton data,
    Figure PCTKR2022005088-appb-img-000085
    Comprising the step of calculating the average deviation / pelvic length of the difference of the z-coordinate of each knee of the pedestrian based on
    상기 제 4단계는 상기 보행자의 각 무릎의 z좌표의 차이의 평균편차/골반길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법The fourth step is a method for determining the severity of Parkinson's syndrome based on gait data, comprising determining whether the pedestrian is included within a preset standard from the average deviation/pelvic length of the difference in the z-coordinate of each knee of the pedestrian
  8. 제7항에 있어서,8. The method of claim 7,
    상기 제3단계는,The third step is
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000086
    를 기준으로 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이를 연산하는 단계를 더 포함하고, 상기 제 4단계는 상기 보행자의 각 발목의 z좌표의 차이의 평균편차/골반길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법    From the skeleton data,
    Figure PCTKR2022005088-appb-img-000086
    Further comprising the step of calculating the average deviation / pelvic length of the difference of the z-coordinate of each ankle of the pedestrian based on Gait data-based Parkinson's syndrome severe stage determination method further comprising the step of determining whether it is included within the set criteria
  9. 제8항에 있어서,9. The method of claim 8,
    상기 제3단계는,The third step is
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000087
    를 기준으로 보행속도/다리길이를 연산하는 단계를 더 포함하고, 상기 제 4단계는 상기 보행속도/다리길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법    From the skeleton data,
    Figure PCTKR2022005088-appb-img-000087
    The step of calculating the walking speed / leg length based on the gait data further comprising the step of determining whether the fourth step is included within a preset standard from the walking speed / leg length Determination of the severe stage of Parkinson's syndrome Way
  10. 제9항에 있어서,10. The method of claim 9,
    상기 제3단계는,The third step is
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000088
    를 기준으로 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발 길이를 연산하는 단계를 더 포함하고, 상기 제 4단계는 상기 측면에서 바라본 보행중인 보행자 양 발의 평균 발길이/보행자의 발 길이로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법    From the skeleton data,
    Figure PCTKR2022005088-appb-img-000088
    Further comprising the step of calculating the average foot length of both feet of the walking pedestrian / foot length of the pedestrian viewed from the side based on A gait data-based Parkinson's syndrome severe stage determination method further comprising the step of determining whether it is included within a preset criterion from
  11. 제10항에 있어서,11. The method of claim 10,
    상기 제3단계는,The third step is
    상기 스켈레톤 데이터로부터,
    Figure PCTKR2022005088-appb-img-000089
    를 기준으로 보행중인 양 무릎의 굽힘 각 평균을 연산하는 단계를 더 포함하고, 상기 제 4단계는 상기 보행중인 양 무릎의 굽힘 각 평균으로부터 기 설정된 기준내에 포함되는지 판별하는 단계를 더 포함하는 보행데이터 기반 파킨슨 증후군 중증 단계 판별 방법    From the skeleton data,
    Figure PCTKR2022005088-appb-img-000089
    Further comprising the step of calculating the average of the bending angle of both knees while walking based on Based on Parkinson's Syndrome Severe Stage Determination Method
PCT/KR2022/005088 2021-04-09 2022-04-08 System and method for determining severe stage of parkinson's syndrome on basis of gait data WO2022216102A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2021-0046442 2021-04-09
KR1020210046442A KR102545358B1 (en) 2021-04-09 2021-04-09 Pedestrian data-based parkinson's syndrome severe phase identification system and method

Publications (1)

Publication Number Publication Date
WO2022216102A1 true WO2022216102A1 (en) 2022-10-13

Family

ID=83546505

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2022/005088 WO2022216102A1 (en) 2021-04-09 2022-04-08 System and method for determining severe stage of parkinson's syndrome on basis of gait data

Country Status (2)

Country Link
KR (1) KR102545358B1 (en)
WO (1) WO2022216102A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104598722A (en) * 2014-12-25 2015-05-06 中国科学院合肥物质科学研究院 Parkinson patient walking ability evaluation method based on gait time-space parameters and three-dimensional force characteristics
KR20160000339A (en) * 2014-06-24 2016-01-04 (주)이튜 Motion extraction-based ambulation posture correcting apparatus
KR101688016B1 (en) * 2015-12-11 2016-12-20 강신현 walk rehabilitation device with gait analysis function
KR20180058999A (en) * 2016-11-25 2018-06-04 알바이오텍 주식회사 System and method for gait analyzing and computer readable record medium thereof
KR20200120365A (en) * 2019-04-12 2020-10-21 울산대학교 산학협력단 Machine learning method and system for automatically grading severity from separated actions of a parkinsonian patient video

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20160000339A (en) * 2014-06-24 2016-01-04 (주)이튜 Motion extraction-based ambulation posture correcting apparatus
CN104598722A (en) * 2014-12-25 2015-05-06 中国科学院合肥物质科学研究院 Parkinson patient walking ability evaluation method based on gait time-space parameters and three-dimensional force characteristics
KR101688016B1 (en) * 2015-12-11 2016-12-20 강신현 walk rehabilitation device with gait analysis function
KR20180058999A (en) * 2016-11-25 2018-06-04 알바이오텍 주식회사 System and method for gait analyzing and computer readable record medium thereof
KR20200120365A (en) * 2019-04-12 2020-10-21 울산대학교 산학협력단 Machine learning method and system for automatically grading severity from separated actions of a parkinsonian patient video

Also Published As

Publication number Publication date
KR102545358B1 (en) 2023-06-21
KR20220140224A (en) 2022-10-18
KR102545358B9 (en) 2023-10-12

Similar Documents

Publication Publication Date Title
US9710920B2 (en) Motion information processing device
US20150294481A1 (en) Motion information processing apparatus and method
CN104274182B (en) Action message processing unit and method
JP6466139B2 (en) Robot measuring instrument that measures human movement
JP2015061579A (en) Motion information processing apparatus
Harsted et al. Concurrent validity of lower extremity kinematics and jump characteristics captured in pre-school children by a markerless 3D motion capture system
WO2022035110A1 (en) User terminal for providing augmented reality medical image and method for providing augmented reality medical image
CN107115102A (en) A kind of osteoarticular function appraisal procedure and device
CN114999646A (en) Newborn exercise development assessment system, method, device and storage medium
McGuirk et al. Feasibility of markerless motion capture for three-dimensional gait assessment in community settings
WO2022216102A1 (en) System and method for determining severe stage of parkinson's syndrome on basis of gait data
KR102556001B1 (en) Gait data-based parkinson's syndrome severity identification system and method
WO2021225249A1 (en) Gait analysis system and method
KR20230102297A (en) Gait data-based parkinson's syndrome severity identification system and method
KR20230102308A (en) system and method for determining sarcopenia based on gait data
KR102556002B1 (en) Skeleton data-based parkinson's syndrome severe phase identification system and method
Jamsrandorj et al. Deep Learning Networks for View-independent Knee and Elbow Joint Angle Estimation
KR20230102262A (en) SKELETON DATA-BASED sarcopenia SEVERE PHASE IDENTIFICATION SYSTEM AND METHOD
Grasso et al. Markerless Pose Estimation of DeepLabCut for Shoulder Motion Assessment in Patients with Cervical Spinal Cord Injury
Usami et al. Evaluation of lower extremity gait analysis using Kinect V2® tracking system
WO2022211244A1 (en) Single two-dimensional image-based quantitative gait analysis device and method
WO2022102863A1 (en) Extended-reality-based acupuncture system and method therefor
US20220296158A1 (en) System, method, and apparatus for temperature asymmetry measurement of body parts
US20240023833A1 (en) Sensor-based walking aid adjustment system
WO2021118084A2 (en) Breast cancer self-examination server and method therefor

Legal Events

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

Ref document number: 22785011

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22785011

Country of ref document: EP

Kind code of ref document: A1