TWI652039B - Simple detection method and system for sarcopenia - Google Patents

Abstract

The invention discloses a simple detection method and system for sarcopenia, which uses an image detecting sensor device to capture a walking image of a subject walking. The image detection sensing device computing wafer converts the walking image into a 3D walking depth image. The action posture recognition unit counts the time required for the walking depth image to move to the preset distance to obtain the walking speed value of the subject. The image detecting sensor sensor picks up the image of the subject while holding the object. The subject generates a grip force sensing message by the hand pressing operation grip sensing means, and the action posture recognizing unit reads the grip force sensing message to generate a corresponding grip strength value. When the walking speed of the subject is lower than the preset walking speed value and the grip strength value is lower than the preset grip strength value, it is determined that the subject is a suspected myasthenia patient, and the purpose of automatically detecting the muscular dystrophy can be achieved.

Description

Simple detection method and system for sarcopenia

The invention relates to a simple detection method and system for muscular dystrophy, in particular to a simple detection technology capable of realizing automatic detection of sarcopenia.

According to the fact that as Taiwan and the world gradually enter the advanced society, the diseases caused by the age increase are getting more and more attention. The role of skeletal muscle in the human body, in addition to the action, is an important part of the body's protein storage. In particular, when the body is under stress or starvation, the body converts the protein of the skeletal muscle into energy to provide the nutrients needed for the body to continue to function. It can be seen that to maintain the normal living function of the elderly, the health of the musculoskeletal system is indispensable. Therefore, with the loss of muscles caused by old age, and thus debilitating, it has already become one of the key research directions of the current medical system. .

In general, diseases that are responsible for the loss of skeletal muscle mass and function, called sarcopenia; elderly people diagnosed with sarcopenia can easily cause debilitation if they have decreased activity and weight. In recent years, both sarcopenia and weakness have been considered as symptoms of senile illness. The mechanism of sarcopenia includes multiple factors such as age-related factors, endocrine changes, nutrition, disuse and immobility, and neurodegeneration. It is a product of the interaction between disease and age, and has multiple causes, which leads to the mobility of the elderly. Obstacles, increased chances of falling, disability, and even death.

Furthermore, according to the current medical system, the definition of sarcopenia can be from the following three parties. To discuss:

1. Muscle quality Clinically, the commonly used measurement methods are Dual energy X-ray absorptionmetry (DXA) or Bio-impedance analysis (BIA), computed tomography and nuclear magnetic resonance. Although the image is more accurate, considering the cost and radiation, it is currently based on research purposes. The body muscle mass is generally assessed by the appendicular skeletal muscle mass index. The algorithm is the limb skeletal muscle mass divided by the square of the height (appendicular skeletal muscle mass/squared height, ASM/ht2). According to the National Institutes of Health's integration of data from National Taiwan University, Chengdu University, Zhongshan University, China Medical University and the National Health Center itself, the results of the latest research are as follows: if ASM/ht2 is lower than the average standard deviation of the young ethnic group or the lowest of the research population 20 The distribution of % defines the cut-off point of the muscle mass of the muscle deficiency. The former has a cut-off point of 6.76 kg/m2 for men and 5.28 kg/m2 for women, and the cut-off point for the latter is 7.09 kg/m2 for men and 5.70 kg/m2 for women.

2. Muscle strength (muscle strength), the most commonly used method in clinical practice is to measure the handgrip strength with a grip; in addition, the knee flexion/extension or maximum expiratory flow rate can be measured. (peak expiratory flow). In the same study as the National Institutes of Health mentioned above, the hand grip strength is cut to the lowest 20% of the study population.

3. Ability to act, according to the recommendations of the EU Minisis Working Group, calculate the normal gait speed and use the short physical performance battery (SPPB), which can be used for clinical practice or research purposes. If the elderly over the age of 65 walks at a speed of less than 0.8 meters per second, the possibility of sarcopenia needs to be further examined. Other measurement methods include the 6-min walk test and the stair climb power test.

In addition, the 2013 Asian Working Group for Sarcopenia (AWGS) Consensus Conference proposed a set of criteria for Asians to diagnose sarcopenia based on the EWGSOP's definition of muscle strength. The amount of muscle is measured at the male point, whether measured by DEXA or BIA. 7.0 kg / m ² , but the female DEXA and BIA measurement cut points are 5.4 kg / m 2 and 5.7 kg / m ² . The standard value of the gripper used for body-measuring grip strength is <26 kg for men and <18 kg for women, and the average walking speed is <0.8 m/s.

The Asian Hysteria Working Group recommends that both grip strength and general walking speed be used as the primary screening conditions for sarcopenia. When one of the two meets the constitutive requirements, it is necessary to further measure whether the muscle mass is too low to determine whether it is consistent. Diagnosis of sarcopenia, if there is no hand pressure sensing component in the clinic, you can ask the senior to do the "timed up walking test", stand up from the chair, walk 3 meters and then turn, then walk 3 meters, then sit down. If there are more than 20 seconds, there may be myasthenia.

Most medical institutions are equipped with relevant medical personnel and testing equipment to detect whether elderly patients suffer from sarcopenia, although large hospitals can generally determine whether elderly patients suffer from sarcopenia; The establishment of medical personnel in medical institutions in the region is generally inadequate, resulting in medical institutions becoming sweatshops that severely exploit human resources. This has seriously affected the medical quality of medical institutions. Therefore, how to develop a set can save medical care. The automatic detection of the detection of muscle dysfunction in personnel configuration has become a technical issue that the medical and medical industry is eager to challenge and solve.

Up to now, there has not been a patent or paper that combines 3D immersive image and image recognition technology to detect sarcopenia, and based on the urgent needs of related industries, the inventors have been continuously researching and developing. Finally, a set of inventions different from the above-mentioned prior art has been developed.

The main object of the present invention is to provide a simple detection method and system for muscular dystrophy The system mainly combines 3D immersive image and image recognition technology to detect the presence of sarcopenia, thus enabling automated detection of sarcopenia to reduce the cost of medical staffing. The technical means adopted for achieving the main object of the present invention is to capture the walking image of the subject by using the image detecting sensor device sensor. The image detection sensing device computing wafer converts the walking image into a 3D walking depth image. The action posture recognition unit counts the time required for the walking depth image to move to the preset distance to obtain the walking speed value of the subject. The image detecting sensor sensor picks up the image of the subject while holding the object. The subject generates a grip force sensing message by the hand pressing operation grip sensing means, and the action posture recognizing unit reads the grip force sensing message to generate a corresponding grip strength value. When the walking speed of the subject is lower than the preset walking speed value and the grip strength value is lower than the preset grip strength value, the subject is determined to be a suspected muscular dystrophy patient.

10‧‧‧Image detection sensor

11‧‧‧Image Detection Sensor Sensor

12‧‧‧Image detection sensor computing chip

110‧‧‧RGB color camera

111‧‧‧3D structured light depth sensor

20‧‧‧Action posture recognition unit

21‧‧‧Image database

30‧‧‧Signal transmission module

40‧‧‧ grip force sensing means

40a‧‧‧ gripper

P0‧‧‧ Full body depth image

P1‧‧‧ walking depth image

P2‧‧‧ Holding depth image

P3‧‧‧ torso depth image

P4, P6‧‧‧ leg depth image

P5‧‧‧Standing depth image

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic illustration of the implementation of a particular operation of the present invention

2 is a schematic view showing the implementation of walking speed in the walking depth image of the present invention.

Fig. 3 is a schematic view showing the first action of detecting the grip strength of the present invention.

Fig. 4 is a schematic view showing the second action of detecting the grip strength of the present invention.

Fig. 5 is a schematic view showing another embodiment of the second action of the present invention for detecting the grip strength.

Fig. 6 is a schematic view showing the implementation of detecting the shaking angle according to the torso depth image of the present invention.

Fig. 7 is a schematic view showing the implementation of the height raising detection of the present invention.

FIG. 8 is a schematic diagram showing the step implementation of performing the leg raising angle detection according to the present invention.

FIG. 9 is a schematic view showing the relative coordinates and angles of the leg angle detection performed by the present invention.

Figure 10 is a functional block diagram of the circuit architecture of the present invention.

In order to allow the reviewing committee to further understand the technical features of the present invention and the achievement of the object of the present invention, the specific embodiments and the drawings will be described in detail as follows: please refer to FIGS. 1~3 and FIG. In order to achieve a specific embodiment of the main object of the present invention, the following steps are included:

(a) preparation step: providing an image detecting and sensing device 10, a motion gesture recognition unit 20 (such as a combination of a computer and a motion recognition software), a signal transmission module 30, and a grip force sensing means 40, and the like. The image detecting and sensing device 10 includes an image detecting sensor device 11 and an image detecting device computing chip 12, and the image detecting sensor device 11 and the image detecting sensor device 12 are processed for image detection. The device computing chip 12 and the motion posture recognition unit 20 are connected by a signal transmission module 30 (such as a Bluetooth communication module, an RS232 transmission interface or a USB transmission interface). The action posture recognition unit 20 is electrically connected to the grip force 40.

(b) Walking speed detecting step: the subject is allowed to walk in a region range (for example, the indoor space is in a range region that the image detecting sensor device 11 can sense), and the image detecting sensor device 11 is taken The tester continuously moves the image while walking, and outputs the depth information of each walking image. The image detecting and sensing device computing chip 12 converts each walking image into a realistic continuous dynamic 3D walking depth image P1 according to the depth information. As shown in FIG. 2, the motion posture recognizing unit 20 calculates a continuous dynamic complex walking depth image. The time required for P1 to move from point A to the preset distance B (about 8 meters) is used to determine the walking speed value of the subject.

(c) grip strength detecting step: causing the subject to generate the grip force sensing message by pressing the grip force sensing means 40 by hand, and the grip force sensing means 40 may be a kind of sensing force of the subject as shown in FIG. The grip force 40a of the message is measured, and the action gesture recognition unit 20 reads the grip force sensing signal output by the gripper 40a through a signal capture module to generate a corresponding grip strength value; the specific grip force detection operation is The tester presses the gripper 40a several times (about 5-10 times) with a dominant hand (such as a left hand or a right hand), and after being interpreted by the action posture recognition unit 20, the number of presses can be sequentially displayed by the display screen connected thereto. Grip force sensing information such as maximum grip strength, average grip strength, and instant grip strength, and finally the average grip strength value is taken as the grip strength value. In addition, this step may also allow the subject to hold the grip force sensing means 40 with the hand. The grip force sensing means 40 may be the grip force 40a as shown in FIGS. 4-5; or the elastic sphere , but not limited thereto; then, the image detecting sensor device 11 captures the image of the subject holding the grip 40a and outputs the depth information of the image; the image detecting sensor The computing chip 12 converts the holding image into a realistic 3D holding depth image P2 according to the depth information of the holding image; and creates an image database 21 storing the plurality of sample images in the action posture recognizing unit 20, each of which Each of the sample images has image characteristics of the deformation range of the gripper 40 caused by different grip strengths, and each sample image defines a grip strength value; the motion gesture recognition unit 20 can compare the grip depth maps in the image database 21 A sample image like the P2 feature, where the grip strength of the sample image can be read.

(d) Body size input step: The input parameters such as the name, identification code, height, torso, limbs, and fingers of the subject can be sequentially input through the input interface (such as the remote controller, keyboard, or gesture recognition unit 20). In the set parameter database of the image detecting and sensing device computing chip 12, in addition to being used as a conversion, the holding depth image P2 and the walking depth image can be established. The size of P1 is based on reference and can be easily converted into a human skeleton system with multiple movable joints. The specific structure is shown in Figure 1.

(e) Judgment step: when the subject's walking speed is lower than the preset walking speed value (preferably 0.8 m/s) and the grip strength value is lower than the preset grip strength value (the male preset grip strength value is between 25 and 27) When the kilogram and the female preset grip strength value are between 14 and 19 kg, the action posture recognition unit 20 determines that the subject is a suspected muscular dystrophy patient.

In the first application embodiment of the present invention, as shown in FIG. 1 and FIG. 6, when the subject is determined to be a suspected myasthenia patient, the body shaking detecting step is performed to cause the subject to stand and The image detecting sensor sensor 11 captures a continuous dynamic standing image containing the subject standing and a depth message of each standing image. The image detecting sensor computing chip 12 converts each standing image into a realistic continuous dynamic 3D torso depth image P3 including the head according to the depth information of the standing image. The action posture recognizing unit 20 calculates a continuous dynamic complex torso depth image P3 swaying trajectory before and after the trajectory of the continuous dynamic plural torso depth image P3; or the left and right yaw amplitude is higher than a predetermined angle (about 5 to 30 degrees) At this time, the action posture recognizing unit 20 determines that the subject is a confirmed patient with sarcopenia.

Referring to FIG. 1 and FIG. 6 , in the second application embodiment of the present invention, when the subject is determined to be a suspected myasthenia patient, the body shaking detection step is performed to enable the image detecting sensor to operate. The wafer 12 converts each of the walking images into a realistic continuous dynamic 3D torso depth image P3 including the head according to the depth information. The action posture recognition unit 20 recognizes the trajectory of the complex torso depth image P3, when the continuous dynamic plural torso depth image P3 oscillates the trajectory before or after; or the left and right yaw amplitude is higher than the preset angle (about 5 to 30 degrees), the action posture The identification unit 20 determines that the subject is a confirmed patient with sarcopenia.

Referring to FIG. 1 and FIG. 7 , in the third application embodiment of the present invention, when the subject is determined to be a suspected muscular dystrophy patient, a leg height detecting step is performed, which is performed by the subject. The leg raising action is performed, and the image detecting sensor device 11 is used to extract the depth information of the leg image and the leg image including the leg movement of the subject. The image detecting and sensing device computing chip 12 converts the leg image into a realistic 3D leg depth image P4 according to the depth information of the leg image; and calculates the height of the leg depth image P4 by the action posture recognizing unit 20 Whether the position is higher than a preset height, and when the determination result is lower than the preset height (for example, the heel is about 20 to 35 cm from the ground), it is determined that the subject is a confirmed patient with sarcopenia.

Referring to FIG. 1 , FIG. 8 and FIG. 9 , in the fourth application embodiment of the present invention, when the subject is determined to be a suspected muscular dystrophy patient, a step of detecting the leg angle is performed, and the method is The tester performs the actions of standing and lifting the leg; that is, the action of walking in place, and taking the image detecting sensor device 11 to capture the standing image and the leg image including the standing and lifting legs of the subject, and The depth information of the standing image and the lifted leg image. The image detecting and sensing device computing chip 12 sequentially converts the standing image and the lifting leg image into a realistic continuous dynamic 3D standing depth image P5 and a leg raising depth image P6 according to the depth information; the motion posture recognizing unit 20 can be stood The depth image P5 and the leg-lifting depth image P6 calculate the leg-lifting angle of the subject, and then determine whether the leg-lifting angle is higher than a predetermined angle. When the determination result is lower than the preset angle, the name is determined to be The tester is a confirmed patient with sarcopenia. Of course, the subject can also make continuous in-situ stepping action, and then take the average value of the leg raising angle, and then determine whether the average leg raising angle is higher than the above-mentioned preset angle, as the basis for determining the patient diagnosed with muscle deficiency. . Specifically, the above-described action posture recognizing unit 20 takes the hip joint of the standing depth image P5 as a turning point, and the angle at which the thigh of the standing depth image P6 rotates relative to the turning point is taken as the lifting leg angle θ, and the specific expression is As shown in FIG. 8 , the preset angle is less than 30 to 75 degrees; however, it is not limited thereto. In addition, the distance can be measured by the front side of the image detecting and sensing device 10, thereby obtaining the X, Y, Z three-dimensional space coordinates, wherein Z represents the depth distance, so the calculation of the lifting leg angle only needs Y, Z and the like, and the relationship can be Expressed as, , , the leg angle is .

Specifically, in the walking speed detecting step, the gripping force detecting step, and the body shaking detecting step, the present invention can provide two to six subjects to simultaneously perform tracking detection of sarcopenia. In addition, preferably, in the walking speed detecting step, the subject is allowed to travel a distance of 8 meters, and the 3D walking depth image of the imaginary continuous dynamic can be displayed through the display screen of the action posture recognizing unit 20. P1, in turn, calculates the walking speed of the subject. Secondly, in the leg raising angle detecting step, the subject is asked to make 15 left and right feet; or 15 or more in-situ walking movements, and the display screen of the action posture recognizing unit 20 can be displayed as imaginary continuous The dynamic standing depth image P5 and the leg-lifting depth image P6 are used to calculate the average leg raising angle of the subject's left and right feet 15 times.

On the other hand, the image detecting and sensing device 10 of the present invention can adopt the Kinect technology architecture developed by Microsoft Corporation; or other image detecting and sensing devices, but not limited thereto. The Kinect technology architecture is a peripheral device developed by Microsoft Corporation for use on the Xbox 360 and Xbox One consoles. In addition to detecting gestures, it can even detect the user's muscle activity and heartbeat. The Kinect technology architecture analyzes and captures the user's hand, finger movements and body movements, no matter whether it is close or long distance. In addition, the change of the light will not affect the tracking function, because the Kinect sensor (the image detection sensor sensor 11 in this case) is really working, it is not a camera that needs to rely on light, optical code. (Light Coding) is to generate speckle by emitting infrared rays, record the depth information through the Kinect sensor, and then transfer it to the Kinect computing chip (ie The image detection sensing device of the present invention operates the wafer 12) to calculate the desired depth image.

The Kinect technology architecture obtains only basic image data through known Light Coding technology. The key point is to identify the image to convert the image into the desired motion command, body motion or gesture recognition result. . In the Kinect technology architecture, the detected 3D depth image can be converted to the skeleton tracking system, as shown in Figure 1. In this skeleton tracking system, up to 6 people can be detected at the same time, including the action of recognizing 2 people at the same time. Each person can record 20 sets of details, including the scope of the torso, limbs and fingers, etc. It can achieve the purpose of identification of physical operation. In order to understand the action of the subject, the present invention can utilize machine learning to create a huge image database, and thus has the ability to recognize wisdom to understand the subject's limb or trunk movement as much as possible. What is the meaning of the representative, such as identifying the action meanings of the walking depth image P1, the depth image P2, and the torso depth image P3 of the present invention? Here, the identification purpose of the lower limbs, the hand and the trunk of the body of the present invention can be achieved. The optimal distance detected by the image detecting sensor sensor 11 is between 1.2 meters and 3.5 meters, and the horizontal field of view is 57 degrees.

Specifically, the image detecting sensor device 11 is a device similar to a network camera. As shown in FIG. 1 and FIG. 10, the image detecting sensor device 11 includes an RGB color camera 110 located in the middle, and The 3D structured light depth sensor 111 consists of two infrared emitters on the two sides; in addition, the Kinect technology architecture is also equipped with a tracking technology, and the base motor will follow the movement of the focusing object. In addition, the Kinect technology architecture has an array microphone built in, which is composed of multiple sets of microphones simultaneously, which can eliminate noise after comparison. The latest image detection sensor sensor 11 can track the display of 6 complete skeletons and 25 joints of each complete skeleton, thereby making the tracking of the human body part more accurate and stable, and The tracking range is also wider. Through higher depth immersive capabilities and significantly improved noise layers, the image detection sensor sensor 11 provides better 3D virtual capabilities, better viewing of smaller objects and a clearer view of all objects, and Can improve the stability of human body tracking.

In addition, it must be noted that the Kinect technology architecture is indeed a well-known application technology architecture, and can provide general users with services for body tracking projects, in addition to providing developer tools (such as Unity Pro Support, including drivers, tools, APIs, device interfaces and code examples to assist in the development of commercial deployment applications (apps) that support the Kinect architecture.

The above is only a possible embodiment of the present invention, and is not intended to limit the scope of the patents of the present invention, and the equivalent implementations of other changes according to the contents, features and spirits of the following claims should be It is included in the patent of the present invention. The invention is specifically defined in the structural features of the request item, is not found in the same kind of articles, and has practicality and progress, has met the requirements of the invention patent, and has filed an application according to law, and invites the bureau to approve the patent according to law to maintain the present invention. The legal rights of the applicant.

Claims (10)

A simple detection method for myasthenia gravis comprises the following steps: (a) preparing step: providing an image detecting and sensing device, a gripping force sensing device and an action posture recognition unit, wherein the image detection sensing device comprises an image Detecting a sensing device sensor and an image detecting sensing device computing chip, wherein the image detecting sensor computing chip and the motion posture recognition unit are connected by a signal transmission module signal, and the holding force sensing means and the action The posture recognition unit is electrically connected; (b) the walking speed detecting step: the image detecting sensor device is used to capture the continuous moving image of the subject while walking, and output the depth information of each walking image; The image detecting and sensing device computing chip converts each of the walking images into a realistic continuous dynamic 3D walking depth image according to the depth information; and the moving posture recognition unit calculates the continuous dynamic multiple walking depth image to move to the step The required time of a maximum distance to determine the walking speed value of the subject; (c) the grip strength detecting step: making the subject take the hand Pressing the grip force sensing means to generate a grip force sensing message, the action posture recognizing unit reading the grip force sensing message to generate a corresponding grip strength value; and (d) determining step: when the subject's walking speed is low When the preset walking speed value and the grip strength value are lower than a predetermined grip strength value, the motion posture recognizing unit determines that the subject is a suspected muscular dystrophy patient. The method for simple detection of sarcopenia according to claim 1, further comprising a body size input step of performing the body size input step, the name, the identification code, the height, the trunk, the limbs, and the finger of the subject At least two setting parameters are input to the image detecting sensor One of the set-up wafers is set in the parameter database, and the image-sensing device is used as the image-sensing device to calculate the size of the grip depth image and the walking depth image. The method for simple detection of myasthenia as described in claim 1, wherein when the subject is determined to be a suspected myasthenia patient, a body shake detecting step is performed to cause the subject to stand and use the image The detecting sensor sensor captures a continuous dynamic standing image including the standing position of the subject and a depth information of each of the standing images; the image detecting sensing device computing chip will each stand according to the depth information of the standing image The image conversion processing is a realistic continuous dynamic 3D torso depth image including a head; and the motion posture recognition unit calculates the continuous dynamic torso depth image swaying trajectory, when the swaying trajectory is higher than the left and right yaw amplitude At a predetermined angle, the action posture recognizing unit determines that the subject is a confirmed patient with sarcopenia. The method for the simple detection of myasthenia as described in claim 1, wherein when the subject is determined to be a suspected myasthenia patient, a body shake detecting step is performed to cause the image detecting sensor to operate the wafer according to the method. The depth message converts each of the walking images into a realistic continuous dynamic 3D torso depth image including a head; and the motion posture recognition unit recognizes the complex torso depth image swaying trajectory, when the swaying trajectory is anterior or left yaw When the amplitude is higher than a preset angle, the action posture recognition unit determines that the subject is a confirmed patient with dystrophy; the preset angle is between 5 and 30 degrees. The method for simple detection of myasthenia as described in claim 3 or 4, wherein when the subject is determined to be a suspected myasthenia patient, a leg raising angle detecting step is performed to cause the subject to stand And lifting the leg and the like, and using the image detecting sensor sensor to capture a standing image, a leg image, and the standing image and the leg image including the standing and lifting legs of the subject a depth information; the image detecting sensor computing chip sequentially converts the standing image and the leg image into a realistic 3D standing depth image and a leg depth image according to the depth message; the motion posture recognition unit Calculating the leg raising angle of the subject by the standing depth image and the leg depth image, and determining whether the leg raising angle is higher than a preset angle, when the determination result is lower than the preset angle, Then, the subject is determined to be a confirmed patient with sarcopenia. The simple detection method of the muscle dysfunction according to claim 5, wherein the motion posture recognition unit uses the hip joint of the standing depth image as a turning point, and the thigh of the standing depth image is rotated relative to the turning point. The angle of the leg is the angle of the leg, and the preset angle is less than 30 to 75 degrees. The method for simple detection of sarcopenia according to claim 1, wherein the preset walking speed value is between 0.7 and 0.9 meters/second. The method for the simple detection of muscle dysfunction according to claim 1, wherein the predetermined grip strength of the male subject is between 25 and 27 kg, and the predetermined grip strength of the female subject is between 14 and 19. kg. The method for simple detection of sarcopenia according to claim 1, wherein the maximum distance is between 2 and 6 meters. A simple detection system for myasthenia gravis includes: an image detecting and sensing device sensor for capturing a continuous dynamic walking image of at least one subject during walking, and outputting a depth information of each of the walking images; And the image detecting sensor sensor captures the image of the object holding the object and outputs the depth information of the image; and the image detecting device operates the wafer according to the depth The message converts each of the walking images into a realistic continuous dynamic 3D walking depth image; and converts the holding image into an immersive image The 3D holds the depth image; and an action gesture recognition unit is coupled to the image detection sensor computing chip by a signal transmission module signal for calculating the continuous dynamic movement of the complex walking depth image. And a required time to reach a maximum distance to obtain a walking speed value of the subject; and the motion posture identifying unit establishes an image database storing the plurality of sample images, each of the sample images having different grip strengths a deformation magnitude of the object, and each of the sample images defines a grip force value; the motion gesture recognition unit compares the sample image corresponding to the grip depth image in the image database to read a grip strength value of the sample image; when the walking speed of the subject is lower than a preset walking speed value and the grip strength value is lower than a preset grip strength value, the motion posture recognizing unit determines that the subject is Suspected patients with hypothyroidism.