Gait analysis device using ai technology

Abstract

Disclosed is a gait analysis device using artificial intelligence (AI) technology, according to some embodiments of the present invention. The gait analysis device may comprise: a camera unit for capturing an image of the gait of a target user in response to a user interaction through a display unit; a gait analysis unit for generating a gait analysis result obtained by analyzing the gait of the target user on the basis of the captured gait image, wherein the gait analysis result includes at least one of a stride length, a gait speed, a swing angle of both arms, an upper body angle, a hip joint angle of both legs, or a knee angle of the target user; and the display unit for outputting the gait analysis result.

Description

Gait analysis device using AI technology

[0001] The embodiments relate to a gait analysis device.

[0002] Due to the rapid aging of the population, age-related and degenerative diseases such as dementia, Parkinson's disease, stroke, and sarcopenia are increasing at a very fast rate.

[0003] Unlike other diseases, the genetics, pathology, and mechanisms of degenerative brain diseases have only recently been discovered, and there are many aspects that remain unknown.

[0004] Degenerative diseases can include neurodegenerative diseases or degenerative brain diseases. Neurodegenerative diseases are closely related to aging. Unlike the normal aging process, they involve rapid, abnormal neuronal cell death in parts of the nervous system or throughout the brain, resulting in loss of brain and spinal cord function, leading to declines in cognitive abilities, walking, and motor skills.

[0005] Degenerative diseases have a time gap between the onset of symptoms and the occurrence of pathology, making it difficult to achieve fundamental treatment even if an accurate diagnosis is made after the onset of symptoms.

[0006] Therefore, it is important to diagnose and prevent degenerative diseases early.

[0007] The present disclosure is conceived in response to the aforementioned background technology and aims to provide a gait analysis device.

[0008] The technical problems of the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0009] According to one embodiment of the present disclosure for solving the aforementioned problem, a gait analysis device utilizing AI (Artificial Intelligence) technology is disclosed. The gait analysis device may include: a camera unit for photographing a target user's gait according to a user interaction through a display unit; a gait analysis unit for generating a gait analysis result analyzing the target user's gait based on the photographed gait image, wherein the gait analysis result includes at least one of the target user's stride length, gait speed, arm swing angles, upper body angles, hip joint angles, and knee angles; and a display unit for outputting the gait analysis result.

[0010] In addition, the apparatus may further include a muscle strength measurement unit that measures the muscle strength of the target user; a questionnaire acquisition unit that obtains questionnaire information from the target user based on user interaction through the display unit; and a processor that selects a degenerative disease based on at least one of the gait analysis results, the questionnaire information, and the muscle strength information.

[0011] In addition, the processor may obtain a round-trip image captured from the side of the target user performing a round-trip walk at a predetermined distance from the gait analysis device through the camera unit, determine a first point and a second point where the user stopped in the round-trip image, determine a third point located at the center of the first point and the second point, determine a depth value of the user at each of the first point, the second point, and the third point, and correct the first depth value associated with the first point and the third depth value associated with the third point based on the second depth value associated with the second point, thereby correcting distortion of the image captured through the camera unit.

[0012] In addition, the gait analysis unit can generate a skeleton model including a plurality of nodes related to joints of the target user based on the gait image, obtain a plurality of coordinates corresponding to each of the plurality of nodes based on the skeleton model, and generate the gait analysis result including walking speed and stride length based on the obtained plurality of coordinates.

[0013] In addition, the gait analysis unit may determine at least two nodes that exist at a horizontal position within a predetermined angle among the plurality of nodes, extract a plurality of time-series coordinates for each of the at least two nodes, determine a plurality of average values ​​according to the time series from the plurality of time-series coordinates, perform filtering on the plurality of average values, and determine the walking speed based on a result of numerical differentiation on the plurality of filtered average values.

[0014] In addition, the gait analysis unit can determine a corrected gait speed by correcting the gait speed through a spline interpolation method that uses a preset first correction equation when the gait speed is greater than a predetermined value and uses a preset second correction equation when the gait speed is less than a predetermined value.

[0015] In addition, the gait analysis unit determines a first heel node and a second heel node related to the user's heel among the plurality of nodes, extracts a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the first heel node, extracts a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the second heel node, and calculates a mathematical expression The stride is determined based on the above, and is the i-th X-axis coordinate for the first heel node, and is the i-th X-axis coordinate for the second heel node, and is the i-th Y-axis coordinate for the first heel node, and is the i-th Y-axis coordinate for the second heel node, wherein i is a natural number, and j and k may be constants.

[0016] In addition, the gait analysis unit can determine a corrected stride length by correcting the stride length through a spline interpolation method using a preset third correction equation when the stride length is greater than a predetermined value, and using a preset fourth correction equation when the stride length is less than a predetermined value.

[0017] In addition, the method may further include a processor that determines a first node related to the user's nose among the plurality of nodes and de-identifies an area related to the first node.

[0018] The technical solutions obtainable in the present disclosure are not limited to the solutions mentioned above, and other solutions not mentioned will be clearly understood by a person having ordinary skill in the art to which the present disclosure pertains from the description below.

[0019] According to some embodiments of the present disclosure, a gait analysis device capable of early diagnosis and prevention of degenerative diseases can be provided.

[0020] The effects that can be obtained from the present disclosure are not limited to the effects mentioned above, and other effects that are not mentioned will be clearly understood by a person having ordinary skill in the art to which the present disclosure pertains from the description below.

[0021] Various aspects are now described with reference to the drawings, wherein like reference numerals are used to refer to similar elements generally. In the following examples, for purposes of explanation, numerous specific details are set forth to provide a thorough understanding of one or more aspects. However, it will be apparent that such aspects may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form to facilitate the description of one or more aspects.

[0022] FIG. 1 is a block diagram illustrating an example of a gait analysis device according to some embodiments of the present disclosure.

[0023] FIG. 2 is a flowchart illustrating an example of a method for analyzing the gait of a target user by a gait analysis device according to some embodiments of the present disclosure.

[0024] FIG. 3 is a flowchart illustrating an example of a method for analyzing the gait of a target user by a gait analysis unit according to some embodiments of the present disclosure.

[0025] FIG. 4 is a drawing illustrating an example of a skeleton model according to some embodiments of the present disclosure.

[0026] FIG. 5 is a diagram illustrating an example of a method for performing calibration of a gait analysis device according to some embodiments of the present disclosure.

[0027] FIG. 6 is a flowchart illustrating an example of a method for determining a walking speed by a gait analysis device according to some embodiments of the present disclosure.

[0028] FIG. 7 is a graph illustrating an example of a method for determining a walking speed by a gait analysis device according to some embodiments of the present disclosure.

[0029] FIG. 8 is another graph illustrating an example of a method for determining a walking speed by a gait analysis device according to some embodiments of the present disclosure.

[0030] FIG. 9 is a flowchart illustrating an example of a method for determining a stride length by a gait analysis device according to some embodiments of the present disclosure.

[0031] FIG. 10 is a graph illustrating an example of a method for determining a stride length by a gait analysis device according to some embodiments of the present disclosure.

[0032] FIG. 11 is another graph illustrating an example of a method for determining stride length by a gait analysis device according to some embodiments of the present disclosure.

[0033] FIG. 12 is a diagram illustrating an example of a method for a gait analysis device according to some embodiments of the present disclosure to obtain questionnaire information.

[0034] FIG. 13 is a diagram illustrating an example of a method for screening for a degenerative disease using a gait analysis device according to some embodiments of the present disclosure.

[0035] The present invention is susceptible to various modifications and embodiments. Specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, but rather to encompass all modifications, equivalents, and alternatives falling within the spirit and technical scope of the present invention. Throughout the description of each drawing, similar reference numerals have been used to designate similar components.

[0036] Terms such as "first," "second," "A," and "B" may be used to describe various components, but the components should not be limited by these terms. These terms are used solely to distinguish one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the "second component," and similarly, the second component may also be referred to as the "first component." The term "and / or" includes a combination of a plurality of related items described herein or any of a plurality of related items described herein.

[0037] When a component is referred to as being "connected" or "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but that there may be other components intervening. Conversely, when a component is referred to as being "directly connected" or "connected" to another component, it should be understood that there are no other components intervening.

[0038] The terminology used in this application is only used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprise" or "have" indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0039] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0040] In the present disclosure, the gait analysis device can perform gait-related analysis, such as the target user's stride length, gait speed, arm swing angles, upper body angles, hip joint angles, and knee angles. Furthermore, the gait analysis device can screen the target user for a degenerative disease based on the gait analysis results. Degenerative diseases are closely related to aging, and may include diseases in which, unlike the normal aging process, abnormal nerve cell death occurs rapidly in a part of the nervous system or the entire brain, resulting in loss of brain and spinal cord functions and a decline in cognitive ability, gait ability, or motor ability. Degenerative diseases may include Alzheimer's disease, Lewy body dementia, vascular dementia, dementia-induced diseases, Parkinson's disease, tremor, cerebellar degenerative diseases, osteoarthritis, rheumatoid arthritis, osteoporosis, compression fractures, myasthenia gravis, myotonic dystrophy, amyotrophic lateral sclerosis, or sarcopenia. The gait analysis device can screen for degenerative diseases in a target user based on gait images captured by a camera, muscle strength information obtained by a muscle strength measurement unit, and questionnaire information. Hereinafter, the gait analysis device according to the present disclosure will be described with reference to FIGS. 1 to 13.

[0041] FIG. 1 is a block diagram illustrating an example of a gait analysis device according to some embodiments of the present disclosure.

[0042] Referring to FIG. 1, the gait analysis device (100) may include a processor (110), a camera unit (120), a muscle strength measurement unit (130), a questionnaire acquisition unit (140), a gait analysis unit (150), a display unit (160), and a storage unit (170). However, the above-described components are not essential for implementing the gait analysis device (100), and thus the gait analysis device (100) may have more or fewer components than the components listed above.

[0043] The gait analysis device (100) may include any type of computer system or computer device, such as, for example, a microprocessor, a mainframe computer, a digital processor, a portable device, or a device controller. The gait analysis device (100) may be implemented in a personal computer (PC), a data server, a kiosk, or a portable device.

[0044] The gait analysis device (100) may achieve desired system performance by using a combination of typical computer hardware (e.g., devices that may include computer processors, memory, storage, input devices and output devices, and other components of conventional computing devices; electronic communication devices such as routers, switches, etc.; electronic information storage systems such as network-attached storage (NAS) and storage area networks (SAN)) and computer software (i.e., instructions that cause the computing device to function in a specific manner).

[0045] The processor (110) can typically process the overall operation of the gait analysis device (100). The processor (110) can process signals, data, information, etc. input or output through components of the gait analysis device (100) or run application programs stored in the storage unit (170), thereby providing or processing appropriate information or functions to the user.

[0046] The processor (110) may be composed of one or more cores and may include a processor for data analysis, such as a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), or a tensor processing unit (TPU).

[0047] In the present disclosure, the processor (110) can screen for a degenerative disease in a target user. Specifically, the processor (110) can obtain a gait analysis result by analyzing a gait video of the target user through the gait analysis unit (150). The processor (110) can obtain muscle strength information by measuring the target user's muscle strength through the muscle strength measurement unit (130). The processor (110) can obtain questionnaire information related to the target user from the questionnaire acquisition unit (140). The processor (110) can screen for a degenerative disease based on the obtained gait analysis result, questionnaire information, and muscle strength information. Hereinafter, an example of a method for the processor (110) to screen for a degenerative disease will be described with reference to FIG. 2.

[0048] The camera unit (120) can process image frames, such as still images or videos, obtained by an image sensor. The processed image frames can be displayed on a display unit (not shown) or stored in a storage unit (170). The processed image frames can also be transmitted to the gait analysis unit (150). The camera unit (120) can capture the target user's gait. The camera unit (120) can capture the target user's gait from the side.

[0049] In the present disclosure, the camera unit (120) may be a 2D camera. The 2D camera may be a camera that extracts image information by projecting a subject in 3D space onto a 2D plane. The camera unit (120) may be a 2D RGB wide-angle camera. The 2D RGB wide-angle camera may be a camera that uses a wavelength of 400 to 700 nm, which is the same spectrum as that recognized by the human eye. The 2D RGB wide-angle camera may be a camera that generates less distortion when shooting.

[0050] According to one embodiment, the processor (110) can correct distortion of an image captured through the camera unit (120).

[0051] Specifically, the processor (110) can obtain a reciprocating image of a user performing a back-and-forth walk at a predetermined distance from the gait analysis device (100) through the camera unit (120). For example, a mat can be installed at a location at least 2 m away from the front of the gait analysis device (100). The mat can be a 4 m, 4.5 m, 5 m, or 6 m mat. The user can walk back and forth on the installed mat. The camera unit (120) can capture the user walking back and forth from the side. The processor (110) can determine a first point and a second point where the user stopped from the captured reciprocating image. Since the user stops at a point where the back-and-forth movement starts and a point where the user turns during the back and forth, the processor (110) can determine the first point and the second point where the user stopped. For example, the first point can be the left end point of the mat, and the second point can be the right end point of the mat. The processor (110) can determine a third point located at the center of the first point and the second point. The processor (110) can determine the depth values ​​of the user at each of the first point, the second point, and the third point. If distortion occurs in the walking image captured by the camera unit (120), the depth values ​​of the user at each of the first point, the second point, and the third point may be different. Therefore, the processor (110) can determine the depth values ​​of the user at each of the first point, the second point, and the third point. The processor (110) can correct the distortion of the image captured by the camera unit (120) by correcting the first depth value associated with the first point and the third depth value associated with the third point based on the second depth value associated with the second point. The second point may be a point located in front of the walking analysis device (100). Accordingly, distortion may not occur at the second point.The processor (110) can correct distortion of an image captured through the camera unit (120) by correcting the first depth value and the third depth value based on the second depth value related to the second point where no distortion occurs.

[0052] According to one embodiment, the processor (110) may correct distortion of an image based on physical quantities at each of the first point, the second point, and the third point.

[0053] For example, the processor (110) can determine the sizes of objects relative to the user at the first point, the second point, and the third point. The processor (110) can correct distortion of an image captured through the camera unit (120) by correcting the first size of the object at the first point and the third size of the object at the third point based on the second size of the object at the second point.

[0054] The muscle strength measurement unit (130) can measure the muscle strength of the target user. The muscle strength measurement unit (130) may be a grip strength tester or a digital grip strength tester. The muscle strength measurement unit (130) may be connected to the gait analysis device (100) via wired or wireless means. Alternatively, the muscle strength measurement unit (130) may exist independently from the gait analysis device (100). In this case, the target user may input the muscle strength value measured by the muscle strength measurement unit (130) into the gait analysis device (100).

[0055] The questionnaire acquisition unit (140) can acquire questionnaire information from the target user. The questionnaire acquisition unit (140) can provide a pre-stored questionnaire, etc. to the target user through the display unit. The questionnaire may be based on the Sarcopenia Self-Diagnosis Questionnaire (SARC-F). Table 1 below may show an example of the questionnaire.

[0056] Item Question Score Muscle Strength How difficult is it to lift and carry a 4.5 kg box of 9 pears? □ Not at all 0 □ Somewhat difficult 1 □ Very difficult / I can't 2 Walking Assistance How difficult is it to walk from one end to the other? □ Not at all 0 □ Somewhat difficult 1 □ Very difficult / I can / I can't with the help of an assistive device (e.g., a cane) 2 Getting up from a chair How difficult is it to get up from a chair (wheelchair) to a bed (bed), or from a bed (bed) to a chair (wheelchair)? □ Not at all 0 □ Somewhat difficult 1 □ Very difficult / I can't without help 2 Climbing stairs How difficult is it to climb 10 stairs without stopping? □ Not at all 0 □ Somewhat difficult 1 □ Very difficult / I can't 2 Falls How many times have you fallen in the past year? □ Not at all 0 □ 1-3 times 1 □ 4 or more times 2

[0057] The gait analysis unit (150) can analyze the gait of the target user based on the gait image captured by the camera unit (120). The gait analysis unit (150) can generate a skeleton model including a plurality of nodes related to the joints of the user based on the gait image. The gait analysis unit (150) can obtain a plurality of coordinates corresponding to each of the plurality of nodes based on the generated skeleton model. The plurality of coordinates can be determined based on a coordinate system within the image. For example, if the gait image is an image with a resolution of 1280 x 720, the upper left corner point can have a coordinate of 0,0. If the gait image is an image with a resolution of 1280 x 720, the lower right corner point can have a coordinate of 1279,719. The gait analysis unit (150) can generate a gait analysis result including gait speed and stride length based on the obtained plurality of coordinates. The processor (110) can screen for degenerative diseases based on the gait analysis results generated from the gait analysis unit (150). Hereinafter, an example of a method by which the gait analysis unit (150) generates the gait analysis results will be described with reference to FIGS. 6 and 9.

[0058] The display unit (160) displays (outputs) information processed in the gait analysis device (100). For example, the display unit (151) can display execution screen information of an application program driven by the gait analysis device (100), or UI (User Interface) or GUI (Graphical User Interface) information according to such execution screen information.

[0059] The display unit (151) can be formed as a touch screen by forming a mutual layer structure with the touch sensor or by forming an integral structure. This touch screen can function as a user input unit that provides an input interface between the gait analysis device (100) and the user, and at the same time, can provide an output interface between the gait analysis device (100) and the user.

[0060] The storage unit (170) may include memory and / or a permanent storage medium. The memory may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. The storage unit (170) may store a questionnaire and a walking video captured by the camera unit (120).

[0061] According to one embodiment, the gait analysis device (100) may be calibrated prior to capturing a gait video. Specifically, the gait analysis device (100) may be calibrated to convert coordinate values ​​into units of cm. An example of a method for the gait analysis device (100) to perform calibration is described below with reference to FIG. 5, and an example of a method for the gait analysis device (100) to screen for a degenerative disease in a target user is described below.

[0062] FIG. 2 is a flowchart illustrating an example of a method for analyzing the gait of a target user by a gait analysis device according to some embodiments of the present disclosure.

[0063] Referring to FIG. 2, the camera unit (120) of the gait analysis device (100) can capture the gait of the target user according to user interaction through the display unit (S110).

[0064] For example, a mat on which a target user can walk back and forth may be installed at a location at least 2 m away from the front of the gait analysis device (100). A mat having a length of 4 m, 4.5 m, 5 m, or 6 m may be installed in front of the gait analysis device (100). The camera unit (120) may capture the side view of the target user performing the back and forth walk, depending on a user interaction to start filming or a user interaction to obtain the gait analysis results.

[0065] According to one embodiment, the processor (110) can correct a walking image captured through the camera unit (120) and store it in the storage unit (170).

[0066] Specifically, the processor (110) can determine the nose of the target user as the first node through the gait analysis unit (150). The processor (110) can de-identify the facial area of ​​the target user by enlarging the first node. The processor (110) can store the gait image with the de-identified facial area in the storage unit (170). Alternatively, the processor (110) can recognize the user's facial area from the gait image. The processor (110) can de-identify the recognized facial area. The processor (110) can store the gait image with the de-identified facial area in the storage unit (170).

[0067] The gait analysis unit (150) can generate gait analysis results by analyzing the target user's gait based on the captured gait video (S120). The gait analysis results may include at least one of the target user's stride length, walking speed, arm swing angle, upper body angle, hip joint angle of both feet, and knee angle.

[0068] Specifically, the gait analysis unit (150) can generate a skeleton model including multiple nodes related to the joints of the target user based on the gait video. The skeleton model may be a structured model in which the joints of the target user are nodes and the joints are connected to each other by edges. The gait analysis unit (150) can analyze the gait of the target user based on the generated skeleton model. For convenience of explanation, reference may be made to FIGS. 3 and 4.

[0069] FIG. 3 is a flowchart illustrating an example of a method for analyzing the gait of a target user by a gait analysis unit according to some embodiments of the present disclosure. FIG. 4 is a diagram illustrating an example of a skeleton model according to some embodiments of the present disclosure.

[0070] Referring to FIG. 3, the gait analysis unit (150) can generate a skeleton model including a plurality of nodes related to the joints of the target user based on the gait image (S121).

[0071] Specifically, the gait analysis unit (150) can determine multiple joints of the target user within the gait video. The gait analysis unit (150) can create a skeleton model by defining multiple joints as nodes and determining edges connecting the joints.

[0072] For example, the gait analysis unit (150) may determine the nose of the target user as the first node. The gait analysis unit (150) may determine the left shoulder of the target user as the second node, and the right shoulder of the target user as the third node. The gait analysis unit (150) may create an edge connecting the second node and the third node. The gait analysis unit (150) may determine the left elbow of the target user as the fourth node, and the right elbow of the target user as the fifth node. The gait analysis unit (150) may create an edge connecting the second node and the third node. The gait analysis unit (150) may create an edge connecting the third node and the fifth node. The gait analysis unit (150) may determine the left wrist of the target user as the sixth node, and the right wrist of the target user as the seventh node. The gait analysis unit (150) may create an edge connecting the fourth node and the sixth node. The gait analysis unit (150) can create an edge connecting the fifth node and the seventh node. The gait analysis unit (150) can determine the left hip of the target user as the eighth node and the right hip of the target user as the ninth node. The gait analysis unit (150) can create an edge connecting the eighth node and the ninth node. The gait analysis unit (150) can create an edge connecting the second node and the eighth node. The gait analysis unit (150) can create an edge connecting the third node and the ninth node. The gait analysis unit (150) can determine the left knee of the target user as the tenth node and the right knee of the target user as the eleventh node. The gait analysis unit (150) can create an edge connecting the eighth node and the tenth node. The gait analysis unit (150) can create an edge connecting the ninth node and the eleventh node. The gait analysis unit (150) can determine the target user's left ankle as the 12th node and the target user's right ankle as the 13th node.The gait analysis unit (150) can create an edge connecting the 10th node and the 12th node. The gait analysis unit (150) can create an edge connecting the 11th node and the 13th node. The gait analysis unit (150) can determine the left heel of the target user as the 14th node, and the right heel of the target user as the 15th node. The gait analysis unit (150) can create an edge connecting the 12th node and the 14th node. The gait analysis unit (150) can create an edge connecting the 13th node and the 15th node. The gait analysis unit (150) can determine the left heel of the target user as the 16th node, and the right heel of the target user as the 17th node. The gait analysis unit (150) can create an edge connecting the 16th node and the 12th node. The gait analysis unit (150) can create an edge connecting the 16th node and the 14th node. The gait analysis unit (150) can create an edge connecting the 17th node and the 13th node. The gait analysis unit (150) can create an edge connecting the 17th node and the 15th node.

[0073] The walking analysis unit (150) can obtain multiple coordinates corresponding to each of multiple nodes based on the generated skeleton model (S122).

[0074] Multiple coordinates can be determined based on the coordinate system within the image. For example, if the gait video is an image with a resolution of 1280 x 720, the upper left corner point may have the coordinate 0,0. If the gait video is an image with a resolution of 1280 x 720, the lower right corner point may have the coordinate 1279,719.

[0075] The gait analysis unit (150) can generate gait analysis results including gait speed and stride length based on the acquired multiple coordinates (S123).

[0076] Specifically, the gait analysis unit (150) can determine at least two nodes that exist at a horizontal position within a predetermined angle among a plurality of nodes. For example, the gait analysis unit (150) can determine the eighth node and the ninth node as nodes that exist at a horizontal position within a predetermined angle. However, the at least two nodes determined by the gait analysis unit (150) are not limited to the eighth node and the ninth node. The gait analysis unit (150) can extract a plurality of time series coordinates for the eighth node and the ninth node. The time series coordinates can be understood as coordinates recorded over time. The gait analysis unit (150) can determine the walking speed of the target user from the plurality of time series coordinates. Hereinafter, an example of a method by which the gait analysis unit (150) determines the walking speed of the target user will be described with reference to FIG. 6.

[0077] Additionally, the gait analysis unit (150) may determine a first heel node and a second heel node related to the target user's heel among a plurality of nodes. The first heel node may be the 14th node, and the second heel node may be the 15th node. The gait analysis unit (150) may determine the target user's stride length based on the first heel node and the second heel node. An example of a method by which the gait analysis unit (150) determines the target user's stride length is described below with reference to FIG. 9.

[0078] The gait analysis unit (150) can determine the third node, the fifth node, and the seventh node related to the right arm of the target user among the plurality of nodes. The gait analysis unit (150) can determine the right arm swing angle based on the third node, the fifth node, and the seventh node. For example, the gait analysis unit (150) can determine the right arm swing angle with the torso of the target user as the normal axis. As another example, the gait analysis unit (150) can determine the right arm swing angle with the ground as the horizontal axis. The gait analysis unit (150) can determine the second node, the fourth node, and the sixth node related to the left arm of the target user among the plurality of nodes. The gait analysis unit (150) can determine the left arm swing angle based on the second node, the fourth node, and the sixth node. For example, the gait analysis unit (150) can determine the left arm swing angle with the torso of the target user as the normal axis. As another example, the gait analysis unit (150) can determine the left arm swing angle with the ground as the horizontal axis.

[0079] The gait analysis unit (150) can determine the second and third nodes related to the target user's upper body among a plurality of nodes. Based on the second and third nodes, the gait analysis unit (150) can determine the target user's upper body angle. For example, the gait analysis unit (150) can determine the upper body angle with the ground as the horizontal axis.

[0080] The gait analysis unit (150) can determine the eighth node and the ninth node related to the hip joint of the target user among the plurality of nodes. The gait analysis unit (150) can determine the hip joint angle of the target user's two feet based on the eighth node and the ninth node. The gait analysis unit (150) can determine the difference in the angle of the hip joint of the two feet based on the eighth node and the ninth node. For example, the gait analysis unit (150) can determine the difference in the angle of the hip joint of the two feet using the torso of the target user as the normal axis.

[0081] The gait analysis unit (150) can determine the 10th node and the 11th node related to the target user's knee among a plurality of nodes. The gait analysis unit (150) can determine the target user's knee angle based on the 10th node and the 11th node. The gait analysis unit (150) can determine the knee angle difference based on the 10th node and the 11th node. For example, the gait analysis unit (150) can determine the knee angle difference using the target user's torso as a normal axis.

[0082] Referring back to FIG. 2, the display unit (160) can output gait analysis results (S130). For example, the display unit (160) can output gait analysis results including at least one of the target user's stride length, walking speed, arm swing angle, upper body angle, hip joint angle, and knee angle.

[0083] Meanwhile, the processor (110) can screen for degenerative diseases based on gait analysis results, questionnaire information, and muscle strength information.

[0084] Specifically, the processor (110) can obtain questionnaire information through the questionnaire acquisition unit (140). The questionnaire information may be information input from the target user based on the Sarcopenia Self-Diagnosis Questionnaire (SARC-F). The questionnaire information may further include the target user's height, age, weight, and gender. The processor (110) can obtain muscle strength information by measuring the target user's muscle strength through the muscle strength measurement unit (130). The processor (110) can screen for degenerative diseases based on the gait analysis results, questionnaire information, and muscle strength information.

[0085] For example, the processor (110) can screen for a degenerative disease based on the target user's walking speed, stride length, questionnaire information, and muscle strength measurement values ​​included in the body information included in the gait analysis results.

[0086] According to one embodiment, the processor (110) can screen for a degenerative disease in a target user through artificial intelligence (AI).

[0087] Specifically, the processor (110) can obtain a plurality of time series coordinates from the gait analysis unit (150). The processor (110) can preprocess the plurality of time series coordinates into individual indices. For example, the processor (110) can preprocess stride length, walking speed, right arm swing angle, left arm swing angle, upper body angle, hip joint angle of both feet, gait asymmetry, pelvic angle difference, knee angle difference, knee angular velocity difference, upper body bending, left / right pelvic angle, left / right knee angle, upper 5% amplitude, lower 5% amplitude, etc. as individual indices. Performing preprocessing can also be understood to mean that the processor (110) extracts individual indices from the plurality of time series coordinates. The processor (110) can determine a weighted score by mapping questionnaire information and muscle strength values ​​with preset scores. The processor (110) can preprocess the determined weighted score as an individual indice. The processor (110) can determine the similarity between the gait of a normal pedestrian and that of an abnormal pedestrian by inputting the determined individual indicators into a pre-trained artificial intelligence model. Based on the similarity output from the artificial intelligence model, the processor (110) can select a target user for a degenerative disease.

[0088] Artificial intelligence can refer to computer systems with capabilities such as learning, reasoning, or judgment. AI can be implemented using neural networks.

[0089] Neural networks (or artificial neural networks) can encompass statistical learning algorithms that mimic biological neurons in machine learning and cognitive science. A neural network can refer to a model in which artificial neurons (nodes) form a network through synaptic connections, which change the strength of the synaptic connections through learning, thereby achieving problem-solving capabilities.

[0090] Neurons in a neural network can contain a combination of weights or biases. A neural network can include one or more layers, each consisting of one or more neurons or nodes. A neural network can infer a desired outcome from any input by changing the weights of its neurons through learning.

[0091] Neural networks may include deep neural networks. Neural networks include CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), perceptron, multilayer perceptron, FF (Feed Forward), RBF (Radial Basis Network), DFF (Deep Feed Forward), LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit), AE (Auto Encoder), VAE (Variational Auto) Encoder), DAE (Denoising Auto Encoder), SAE (Sparse Auto Encoder), MC (Markov Chain), HN (Hopfield Network), BM (Boltzmann Machine), RBM (Restricted Boltzmann Machine), DBN (Depp Belief Network), DCN (Deep Convolutional Network), DN (Deconvolutional Network), DCIGN (Deep Convolutional Inverse Graphics Network), Generative Adversarial Network (GAN), Liquid State Machine (LSM), Extreme Learning Machine (ELM), It may include ESN (Echo State Network), DRN (Deep Residual Network), DNC (Differentiable Neural Computer), NTM (Neural Turning Machine), CN (Capsule Network), KN (Kohonen Network), transformer, and AN (Attention Network).

[0092] According to the above-described configuration, the gait analysis device (100) can output gait analysis results including stride length, gait speed, arm swing angles, upper body angles, hip joint angles and knee angles of both feet, etc. In the past, an expert had to analyze a user's gait using motion graphs generated from gait videos. However, the gait analysis device (100) according to the present disclosure can analyze the gait of a target user using AI technology, thereby enabling the target user to obtain gait analysis results without the assistance of others. Furthermore, the gait analysis device (100) can screen for degenerative diseases of the target user based on the gait analysis results, muscle strength information, and questionnaire information. In the past, the walking speed of the target user was not measured, or even if the walking speed was measured, only an inaccurate walking speed was measured using an IMU (Inertial Measurement Unit) sensor, a smartwatch, or a smart insole. On the other hand, the walking analysis device (100) according to the present disclosure can determine an accurate walking speed by measuring the walking speed according to the user's set protocol of round-trip walking. In addition, the walking analysis device (100) can screen for a degenerative disease of the target user by combining the determined walking speed with muscle strength values ​​and questionnaire information.

[0093] Meanwhile, according to some embodiments of the present disclosure, the gait analysis device (100) may perform calibration before acquiring a gait image of the target user. Here, calibration may be understood as an operation performed to precisely adjust coordinate values ​​determined from the gait image. An example of a method by which the gait analysis device (100) performs calibration will be described below with reference to FIG. 5 .

[0094] FIG. 5 is a diagram illustrating an example of a method for performing calibration of a gait analysis device according to some embodiments of the present disclosure.

[0095] Referring to Fig. 5, the gait analysis device (100) can output a screen for setting guidelines through the display unit (160). In Fig. 5, it is assumed that the gait analysis device (100) exists in a fixed environment.

[0096] A mat may be installed at a location at least 2 m away from the front of the gait analysis device (100). For example, a mat having a length of 4 m, 4.5 m, 5 m, or 6 m may be installed in front of the gait analysis device (100).

[0097] The gait analysis device (100) can output a guideline setting screen for performing calibration through the display unit (160). The user can input each item output through the display unit (160). The user may be an administrator of the gait analysis device (100). Alternatively, the user may be a target user who wishes to obtain gait analysis results. For example, the user may input the length of the installed mat as the actual walking path length. As another example, the user may input the size of the guideline so that it corresponds to the size of the mat. The gait analysis device (100) can calculate a value of cm per pixel based on user interaction. Accordingly, the gait analysis results analyzed by the gait analysis unit (150) can be made more precise.

[0098] Meanwhile, the gait analysis device (100) may be located in a non-stationary environment. In this case, the gait analysis device (100) can calculate a centimeter-per-pixel value based on the user's height. The user's height can be obtained based on user interaction via the display unit (160).

[0099] For example, the gait analysis device (100) can determine a node that estimates the user's crown of the head from an image for performing calibration. The gait analysis device (100) can determine a node that estimates the center of the heel from an image for performing calibration. The gait analysis device (100) can determine the lengths of two nodes. The gait analysis device (100) can calculate a cm value per pixel based on the user's height and the lengths of the two nodes. For example, the gait analysis device (100) can calculate a cm value per pixel by multiplying a value obtained by a preset parameter by the user's height and dividing the value by the lengths of the two nodes.

[0100] For another example, the gait analysis device (100) can determine a node for estimating the user's nose from an image for performing calibration. The gait analysis device (100) can determine a node for estimating the center of the heel from an image for performing calibration. The gait analysis device (100) can determine the lengths of two nodes. The gait analysis device (100) can calculate a cm value per pixel based on the user's height and the lengths of the two nodes. For example, the gait analysis device (100) can calculate a cm value per pixel by multiplying the user's height by a preset parameter and dividing the value by the lengths of the two nodes. According to the results of repeated experiments, it has been shown that the parameter is determined to be 0.94 to produce the optimal result, but the present invention is not limited thereto.

[0101] According to the above-described configuration, the gait analysis device (100) can be calibrated before acquiring a gait image of the target user. Accordingly, the accuracy of the gait analysis results determined by the gait analysis device (100) can be improved.

[0102] Meanwhile, according to some embodiments of the present disclosure, the gait analysis device (100) can determine gait analysis results including walking speed. Hereinafter, an example of a method for the gait analysis device (100) according to the present disclosure to determine walking speed will be described with reference to FIGS. 6 to 8.

[0103] FIG. 6 is a flowchart illustrating an example of a method for determining a walking speed by a gait analysis device according to some embodiments of the present disclosure.

[0104] Referring to FIG. 6, the gait analysis unit (150) of the gait analysis device (100) can determine at least two nodes that exist at a horizontal position within a predetermined angle among a plurality of nodes (S1231). Existing at a horizontal position within a predetermined angle can also be understood to mean that the angle of the edge connecting at least two nodes is approximately horizontal. For example, the gait analysis unit (150) can determine that at least two nodes exist at a horizontal position when it is determined that the angle of the edge connecting at least two nodes is less than 20 degrees with respect to the x-axis. Alternatively, the gait analysis unit (150) can determine at least two nodes that exist at a horizontal position within a predetermined angle based on the X-axis coordinate of each of the plurality of nodes.

[0105] For example, the gait analysis unit (150) may determine the eighth node and the ninth node as nodes existing in a horizontal position within a predetermined angle. However, the at least two nodes determined by the gait analysis unit (150) are not limited to the eighth node and the ninth node. The gait analysis unit (150) may determine the second node and the third node as nodes existing in a horizontal position within a predetermined angle. Alternatively, the gait analysis unit (150) may determine the second node, the third node, the eighth node, and the ninth node as nodes existing in a horizontal position within a predetermined angle.

[0106] The gait analysis unit (150) can extract multiple time-series coordinates for each of at least two nodes (S1232). Time-series coordinates can be understood as coordinates recorded over time.

[0107] For example, a gait video may be a video having a speed of 30 frames per second. If the gait video is 6 seconds long, the gait video may be formed of 180 images. The gait analysis unit (150) may extract the first time-series coordinates of each of at least two nodes in the first frame. The gait analysis unit (150) may extract the second time-series coordinates of each of at least two nodes in the second frame. The gait analysis unit (150) may extract the 180th time-series coordinates of each of at least two nodes in the 180th frame.

[0108] In one embodiment, the multiple time series coordinates may be represented as a coordinate system (e.g., 0,0) or may be represented in units of cm depending on how the calibration is performed.

[0109] The gait analysis unit (150) can determine multiple average values ​​according to the time series from multiple time series coordinates (S1233).

[0110] For example, the gait analysis unit (150) can determine the average value of the first time series coordinates of each of at least two nodes. The gait analysis unit (150) can determine the average value of the second time series coordinates of each of at least two nodes. The gait analysis unit (150) can determine the average value of the third time series coordinates of each of at least two nodes.

[0111] The gait analysis unit (150) can perform filtering on multiple average values ​​(S1234). The gait analysis unit (150) can perform filtering on multiple average values ​​by extracting multiple average values ​​at preset data intervals.

[0112] For example, the gait analysis unit (150) may perform filtering on the plurality of average values ​​by extracting the average value of the first time series coordinates of at least two nodes, the average value of the 11th time series coordinates of at least two nodes, the average value of the 21st time series coordinates of at least two nodes, the average value of the 31st time series coordinates of at least two nodes, and the average value of the 171st time series coordinates of at least two nodes among the plurality of average values. However, the filtering performing method performed by the gait analysis unit (150) is not limited thereto, and filtering on the plurality of average values ​​may also be performed using a moving average filter method, etc.

[0113] The gait analysis unit (150) can determine the walking speed based on the numerical differentiation results for the filtered multiple average values ​​(S1235).

[0114] Specifically, the gait analysis unit (150) can perform numerical differentiation for multiple average values ​​based on the mathematical formula below.

[0115] [Mathematical Formula 1]

[0116]

[0117] V can be a velocity. can be the velocity at point i. i can be a natural number. X can be the mean. can be the average value at point i. may be the average value at time i+1. t may be a data time interval. If the gait video is a video of 30 frames per second and the preset data interval is 10 units, t may be 1 / 3 second. The gait analysis unit (150) may perform numerical differentiation on multiple average values ​​to determine multiple speed values.

[0118] When multiple speed values ​​are determined, the gait analysis unit (150) can determine an appropriate maximum value related to the multiple speed values. For example, the gait analysis unit (150) can determine the Nth largest value among the multiple speed values ​​as the appropriate maximum value. N may be a preset value. Alternatively, the gait analysis unit (150) can determine at least two speed values ​​having a size in the top 5% among the multiple speed values. The gait analysis unit (150) can determine an average value of the at least two determined speed values ​​as the appropriate maximum value.

[0119] The gait analysis unit (150) can determine the target user's walking speed by correcting the determined appropriate maximum value. The gait analysis unit (150) can determine the target user's walking speed by correcting the appropriate maximum value through linear regression analysis.

[0120] Specifically, the gait analysis unit (150) can determine the walking speed based on the mathematical formula below.

[0121] [Equation 2]

[0122]

[0123] V may be the target user's walking speed. may be the appropriate maximum value. and may be a preset parameter.

[0124] According to one embodiment, the gait analysis unit (150) can determine a corrected gait speed by correcting the gait speed through a spline interpolation method that uses a preset first correction equation when the gait speed is greater than a predetermined value and uses a preset second correction equation when the gait speed is less than a predetermined value.

[0125] Specifically, the gait analysis unit (150) can determine the corrected gait speed based on the mathematical formula below.

[0126] [Equation 3]

[0127]

[0128]

[0129] may be the target user's walking speed. may be a corrected walking speed. may be the appropriate maximum value. , , and may be a preset parameter.

[0130] According to the above-described configuration, the gait analysis device (100) can determine the target user's walking speed from the target user's walking video. The gait analysis device (100) can perform corrections to the determined walking speed. Accordingly, the accuracy of the result value for the target user's walking speed determined by the gait analysis device (100) can be improved.

[0131] FIG. 7 is a graph illustrating an example of a method for determining a walking speed by a gait analysis device according to some embodiments of the present disclosure.

[0132] The gait analysis device (100) can extract a plurality of time series coordinates. The gait analysis device (100) can determine a plurality of average values ​​according to the time series from the plurality of time series coordinates to generate a first graph (210). The gait analysis device (100) can perform filtering on the plurality of average values ​​to generate a second graph (220). The gait analysis device (100) can perform numerical differentiation on the filtered plurality of average values. Meanwhile, the third graph (230) can represent the result of the gait analysis device (100) performing numerical differentiation without performing filtering on the plurality of average values. Referring to the third graph (230), it can be confirmed that the third graph (230) has a large amount of noise because the gait analysis device (100) does not perform filtering on the plurality of average values.

[0133] The gait analysis device (100) can perform filtering on a plurality of average values ​​by extracting the plurality of average values ​​at preset data intervals. The gait analysis device (100) can generate a fourth graph (240) by performing numerical differentiation on the plurality of filtered average values.

[0134] The gait analysis device (100) can perform filtering on a plurality of average values ​​using a moving average filter method. The gait analysis device (100) can generate a fifth graph (250) by performing numerical differentiation on the plurality of filtered average values.

[0135] The gait analysis device (100) can compare the noise present in the fourth graph (240) and the fifth graph (250). The gait analysis device (100) can determine the walking speed using the graph in which it is determined that less noise has occurred.

[0136] For example, referring to the fourth graph (240) and the fifth graph (250) in FIG. 7, it can be confirmed that the fifth graph (250) has significantly more noise than the fourth graph (240). The gait analysis device (100) can determine the walking speed using the fourth graph (240).

[0137] FIG. 8 is another graph illustrating an example of a method for determining a walking speed by a gait analysis device according to some embodiments of the present disclosure.

[0138] The gait analysis unit (150) of the gait analysis device (100) can determine the target user's walking speed by correcting the determined appropriate maximum value. The gait analysis unit (150) can determine the target user's walking speed by correcting the appropriate maximum value through linear regression analysis.

[0139] Referring to the sixth graph (310) of Fig. 8, the gait analysis unit (150) can determine the corrected walking speed of the target user by correcting the appropriate maximum value through linear regression analysis.

[0140] According to one embodiment, the gait analysis unit (150) can determine a corrected gait speed by correcting the gait speed through a spline interpolation method using a preset first correction equation when the gait speed is greater than a predetermined value, and using a preset second correction equation when the gait speed is less than the predetermined value.

[0141] For example, referring to the first point (3211) of the first line (thick line) (321) of the seventh graph (320), the gait analysis unit (150) can determine the corrected gait speed through the second correction equation when the target user's walking speed is less than the preset speed. The gait analysis unit (150) can determine the corrected gait speed through the first correction equation when the target user's walking speed is greater than the preset speed.

[0142] Meanwhile, according to some embodiments of the present disclosure, the gait analysis device (100) can determine gait analysis results including stride length. Hereinafter, an example of a method for the gait analysis device (100) according to the present disclosure to determine walking speed will be described with reference to FIG. 9.

[0143] FIG. 9 is a flowchart illustrating an example of a method for determining a stride length by a gait analysis device according to some embodiments of the present disclosure.

[0144] Referring to FIG. 9, the gait analysis unit (150) can determine a first heel node and a second heel node related to the user's heel among a plurality of nodes (S210).

[0145] For example, according to the example described above through FIG. 4, the gait analysis unit (150) may determine the 14th node associated with the left heel as the first heel node. The gait analysis unit (150) may determine the 15th node associated with the right heel as the second heel node.

[0146] For another example, the gait analysis unit (150) may determine the 16th node associated with the left heel as the first heel node. The gait analysis unit (150) may determine the 17th node associated with the right heel as the second heel node.

[0147] The gait analysis unit (150) can extract multiple X-axis coordinates and multiple Y-axis coordinates for the first heel node, and can extract multiple X-axis coordinates and multiple Y-axis coordinates for the second heel node (S210).

[0148] For example, the gait analysis unit (150) can extract a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the first heel node according to the time series. The gait analysis unit (150) can extract a plurality of X-axis time series coordinates and a plurality of Y-axis time series coordinates for the first heel node. The gait analysis unit (150) can extract a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the second heel node according to the time series. The gait analysis unit (150) can extract a plurality of X-axis time series coordinates and a plurality of Y-axis time series coordinates for the second heel node.

[0149] The gait analysis unit (150) can determine a stride length based on a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the first heel node, and a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the second heel node.

[0150] Specifically, the gait analysis unit (150) can determine the stride length based on the mathematical formula below (S230).

[0151] [Equation 4]

[0152]

[0153] is the i-th X-axis coordinate for the first heel node, may be the i-th X-axis coordinate for the second heel node. is the i-th Y-axis coordinate for the first heel node, may be the ith Y-axis coordinate for the second heel node. i is a natural number, and j and k may be constants. For example, j may be 2, and k may be 0.5. However, the values ​​of j and k are not limited thereto.

[0154] According to one embodiment, the gait analysis unit (150) can determine a plurality of stride lengths through mathematical equation 4. When a plurality of stride lengths are determined, the gait analysis unit (150) can determine an appropriate maximum value related to the plurality of stride lengths. For example, the gait analysis unit (150) can determine the Nth largest value among the plurality of stride lengths as the appropriate maximum value. N can be a preset value. Alternatively, the gait analysis unit (150) can determine at least two stride lengths having a size in the top 5% among the plurality of stride lengths. The gait analysis unit (150) can determine an average value of the at least two determined stride lengths as the appropriate maximum value.

[0155] According to one embodiment, the gait analysis unit (150) can determine the stride of the target user by correcting the determined appropriate maximum value. The gait analysis unit (150) can determine the stride of the target user by correcting the appropriate maximum value through linear regression analysis. The gait analysis unit (150) can determine the corrected stride through spline interpolation using a preset third correction equation when the appropriate maximum value is greater than a predetermined value, and using a preset fourth correction equation when the appropriate maximum value is less than the predetermined value.

[0156] Specifically, the gait analysis unit (150) can determine the corrected stride length based on the mathematical formula below.

[0157] [Equation 5]

[0158]

[0159]

[0160] may be an appropriate maximum value related to multiple strides. may be a correction step. , , and may be a preset parameter.

[0161] According to the above-described configuration, the gait analysis device (100) can determine the target user's stride length from the target user's gait video. The gait analysis device (100) can perform corrections to the determined stride length. Accordingly, the accuracy of the result value for the target user's stride length determined by the gait analysis device (100) can be improved.

[0162] FIG. 10 is a graph illustrating an example of a method for determining a stride length by a gait analysis device according to some embodiments of the present disclosure.

[0163] The gait analysis unit (150) of the gait analysis device (100) can determine a first heel node and a second heel node related to the user's heel among a plurality of nodes. The gait analysis unit (150) can extract a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the first heel node, and can extract a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the second heel node. In addition, the gait analysis unit (150) can determine an eighth graph (410) in which a stride length is determined based on mathematical expression 4.

[0164] FIG. 11 is another graph illustrating an example of a method for determining stride length by a gait analysis device according to some embodiments of the present disclosure.

[0165] Referring to the ninth graph (510) of Fig. 11, the gait analysis unit (150) can determine the corrected stride of the target user by correcting the appropriate maximum value through linear regression analysis.

[0166] For example, referring to the second point (512) of the second line (511) of the ninth graph (510), the gait analysis unit (150) can determine the corrected stride length through the second correction equation if the appropriate maximum value related to the stride length of the target user is less than the preset value. The gait analysis unit (150) can determine the corrected stride length through the first correction equation if the appropriate maximum value related to the stride length of the target user is greater than the preset value.

[0167] FIG. 12 is a diagram illustrating an example of a method for a gait analysis device according to some embodiments of the present disclosure to obtain questionnaire information.

[0168] Referring to Fig. 12, the questionnaire acquisition unit (140) of the gait analysis device (100) can output questions related to sarcopenia through the display unit. The questionnaire acquisition unit (140) can acquire questionnaire information based on user interaction.

[0169] The processor (110) can output the questionnaire results based on the acquired questionnaire information.

[0170] FIG. 13 is a diagram illustrating an example of a method for screening for a degenerative disease using a gait analysis device according to some embodiments of the present disclosure.

[0171] Referring to FIG. 13, the gait analysis device (100) can output a start screen through the display unit (S310). The gait analysis device (100) can receive member information or non-member information from the target user.

[0172] The gait analysis device (100) can obtain height, age, gender, etc. from the target user (S320). The gait analysis device (100) can obtain the target user's height, age, weight, gender, etc. as physical information.

[0173] The gait analysis device (100) can receive user interaction for capturing gait images (S330). The gait analysis device (100) can provide functions such as starting recording after 20 seconds or starting recording after 30 seconds to enable the target user to smoothly perform a round-trip walk.

[0174] In some embodiments, the gait analysis device (100) may output a questionnaire through the display unit prior to step S330. Accordingly, the gait analysis device (100) can obtain questionnaire information from the target user.

[0175] The gait analysis device (100) can capture the round-trip gait of the target user (S340).

[0176] The gait analysis device (100) can analyze the target user's gait video and output a result sheet screening for degenerative diseases (S350). According to an embodiment, the gait analysis device (100) can screen for degenerative diseases and output a result sheet based on gait analysis results, questionnaire information, and muscle strength information.

[0177] According to an embodiment, the gait analysis device (100) may be connected to a printing device for outputting a result sheet. The gait analysis device (100) may output the result sheet via the connected printing device. Alternatively, the gait analysis device (100) may transmit the result sheet to a server or a user terminal of a target user, thereby causing the result sheet to be displayed on the server or the user terminal.

[0178] According to the configuration described above with reference to FIGS. 1 to 13, the gait analysis device (100) can easily perform a gait test on a target user within 1 minute. Furthermore, the gait analysis device (100) can rapidly determine whether or not a target user has sarcopenia within 5 minutes. Furthermore, since the gait analysis device (100) only needs to obtain questionnaire information and a gait video from the target user, the target user can conveniently perform the test without the assistance of others.

[0179] The description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the embodiments disclosed herein, but is to be construed in the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. As a gait analysis device utilizing AI (Artificial Intelligence) technology, A camera unit that captures the target user's walking according to user interaction through the display unit; A gait analysis unit that generates a gait analysis result by analyzing the gait of the target user based on the captured gait video, wherein the gait analysis result includes at least one of the target user's stride length, gait speed, arm swing angle, upper body angle, hip joint angle of both feet, and knee angle; and A display unit for outputting the above gait analysis results; including, Gait analysis device.

2. In paragraph 1, A muscle strength measurement unit for measuring the muscle strength of the target user; A questionnaire acquisition unit that obtains questionnaire information from the target user according to user interaction through the display unit; and A processor for screening for a degenerative disease based on at least one of the above gait analysis results, the questionnaire information, and the muscle strength information; including more, Gait analysis device.

3. In paragraph 2, The above processor, A reciprocating image of the target user performing a reciprocating walk from a predetermined distance away from the gait analysis device is obtained from the side through the camera unit, In the above round trip video, determine the first and second points where the user stopped. Determine a third point located at the center of the first point and the second point, Determining the depth value of the user at each of the first point, the second point, and the third point, By correcting the first depth value associated with the first point and the third depth value associated with the third point based on the second depth value associated with the second point, distortion of the image captured through the camera unit is corrected. Gait analysis device.

4. In paragraph 1, The above gait analysis unit, Based on the above walking image, a skeleton model including a plurality of nodes related to the joints of the target user is generated, Obtaining a plurality of coordinates corresponding to each of the plurality of nodes based on the above skeleton model, Generating the gait analysis results including walking speed and stride length based on the acquired plurality of coordinates, Gait analysis device.

5. In paragraph 4, The above gait analysis unit, Determine at least two nodes that are located at a horizontal position within a predetermined angle among the above plurality of nodes, Extracting multiple time series coordinates for each of the at least two nodes above, Determine multiple average values ​​according to the time series from the multiple time series coordinates above, Perform filtering on the above multiple average values, Determining the walking speed based on the numerical differentiation results for the filtered multiple average values, Gait analysis device.

6. In paragraph 5, The above gait analysis unit, When the walking speed is greater than a predetermined value, a preset first correction equation is used, and when the walking speed is less than a predetermined value, a preset second correction equation is used to determine a corrected walking speed by correcting the walking speed through a spline interpolation method. Gait analysis device.

7. In paragraph 4, The above gait analysis unit, Determine a first heel node and a second heel node related to the user's heel among the plurality of nodes, Extracting a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the first heel node, and extracting a plurality of X-axis coordinates and a plurality of Y-axis coordinates for the second heel node, Mathematical formula Determine the stride based on the above, Above is the i-th X-axis coordinate for the first heel node, and is the i-th X-axis coordinate for the second heel node, and is the i-th Y-axis coordinate for the first heel node, and is the i-th Y-axis coordinate for the second heel node, wherein i is a natural number, and j and k are constants. Gait analysis device.

8. In paragraph 7, The above gait analysis unit, If the stride is greater than a predetermined value, a preset third correction equation is used, and if the stride is less than a predetermined value, a preset fourth correction equation is used to determine a corrected stride by compensating the stride through a spline interpolation method. Gait analysis device.

9. In paragraph 4, A processor that determines a first node related to the user's nose among the plurality of nodes and de-identifies an area related to the first node; including more, Gait analysis device.

Images