KR20240033329A - Waist load measuring method and terminal equipped with waist load prediction application - Google Patents

Abstract

A method and device for measuring waist load are disclosed. The waist load measurement method for measuring the load generated on the waist between bending movements of the present invention in real time uses upper body movement data (kinematic data) to measure lumbar vertebrae 5 and sacrum 1 (L5/S1) during waist bending. ) By measuring the lumbar load using the compressive force generated between the upper body movement data (kinematic data), the compressive force applied between the 5th lumbar vertebra and the 1st sacral vertebra (L5/S1) ) can be measured, which has the effect of minimizing back injuries by allowing workers to recognize the load on their backs in real time while working.

Description

Waist load measuring method and device {WAIST LOAD MEASURING METHOD AND TERMINAL EQUIPPED WITH WAIST LOAD PREDICTION APPLICATION}

The present invention relates to a prediction algorithm for lumbar load occurring between movements, and more specifically, a method for measuring lumbar load occurring between movements that can measure in real time the load occurring between bending the waist, which is one of the causes of low back pain. and technology related to the device.

Musculoskeletal diseases are closely related to worker characteristics, the type of work performed by the worker, and the work environment. As technology advances and many parts of the work are transformed into mechanized and automated production systems, workers at production and assembly sites spend a lot of time performing simple and repetitive tasks. The problem with continuously performing simple and repetitive tasks is that workers feel bored psychologically, and physically it can act as a cause of work-related musculoskeletal disorders.

Work-related musculoskeletal diseases require a lot of money and time to treat, and involve a risk of recurrence even after returning to work. Therefore, it is necessary to prevent musculoskeletal diseases in advance rather than treat them.

Work rotation is a representative prevention method to prevent musculoskeletal disorders, but it is difficult to accurately measure worker workload, and there is no way to determine an effective work rotation sequence to prevent musculoskeletal disorders.

A representative method for measuring a worker's workload in the field is evaluation based on working posture.

However, directly evaluating all working postures consumes a lot of time and manpower, so the reality is that workload is generally measured by selecting representative working postures.

Collecting and evaluating workers' working posture in real time has become possible through the use of motion capture technology. Therefore, it is possible to calculate the workload for all movements of the worker, which was previously impossible, and the worker workload can be determined more accurately.

In particular, research is active to attempt to lower the probability of low back pain by measuring the load generated on the lower back in real time during repetitive bending, which is one of the causes of low back pain.

Existing equipment for measuring waist motion is the Lumbar Motion Monitor (LMM) developed by Ohio State University in the United States. The lumbar motion measuring instrument, also known as LMM, developed in the United States, expresses the state of the spine while a worker performs work in an industrial site. This is a measuring device for accurately expressing dynamic waist motion in three dimensions, but it expresses the estimated value by differentiating the measured motion angle using four potentiometers.

In addition, in order to use a lumbar motion measuring device, existing measurement data related to the compressive force applied to the lumbar spine must be input into the equipment for analysis, and it has the disadvantage of being able to analyze data on the worker's motion in real time. It had the disadvantage of only displaying the final result.

Meanwhile, motion capture refers to the process of capturing the movement of the human body and recording it in digital form by attaching a sensor to the body or using an infrared camera. Motion capture data obtained through such motion capture is digital content and is used in a variety of ways, such as load measurement, animation, movie games, motion analysis, and rehabilitation.

Motion capture systems can be classified into optical, mechanical, and sensor types depending on the method of extracting data. The optical method is a method of extracting location data by attaching a marker to a certain area to measure the movement of an object and then photographing it through multiple cameras. It has the advantage of being able to shoot at high speeds and has few restrictions on the subject's movement, but it has the disadvantage of being very expensive and causing data loss if the marker is obscured. The mechanical method captures the movement of each joint of the human body by attaching a mechanical device, and the data provided is rotation data for each joint extracted in real time, so very accurate data can be provided. However, by attaching a heavy mechanical device, the subject's natural movement is restricted, which limits the extraction of motion data. The sensor method attaches a sensor that generates a magnetic field to each joint of the object and extracts location data by measuring changes in the magnetic field according to the object's movement. There are problems such as limitations in the movement of objects due to cables, but these problems are being solved with the recent development of MEMS (Microelectromechanical systems) technology and wireless communication technology.

In the case of a sensor-type motion capture system, an acceleration sensor is essentially included to measure the acceleration signal used in position movement calculation. By integrating the acceleration signal, the moving distance and moving speed of the object are calculated.

JP Patent Publication No. 6968351 (2021.10.29)

To solve this problem, the present invention uses a motion capture system to develop an algorithm to measure the compressive force that occurs between the 5th lumbar vertebra and the 1st sacrum (L5/S1) when bending the waist, and provides it to the app. The purpose is to provide a method and device for measuring lumbar load that can be mounted.

In addition, the present invention provides a method and device for measuring lumbar load that can measure the compressive force applied between lumbar 5 and sacral 1 (L5/S1) using upper body movement data (kinematic data). provided for a different purpose.

Another purpose of the present invention is to provide a method and device for measuring waist load that can predict the pressure force applied to the waist in real time through an IMU sensor worn on the body.

To solve this problem, the waist load measurement method for measuring the load generated on the waist between bending movements in real time of the present invention uses upper body movement data (kinematic data) to measure the 5th lumbar vertebra and sacrum between waist bending. This can be achieved by configuring the waist load to be measured using the compressive force that occurs between number 1 (L5/S1).

This upper body movement data (kinematic data) is composed by collecting upper body position data using a motion capture system (VICON) and data on acceleration and angular velocity using an IMU, and the measured pressure force is collected from the U.S. National Institute of Safety and Health Based on the (NIOSH) back injury prevention guidelines (pressure force of 3,400N or less), if the measured pressure force is less than 3,400N, it is judged as a safe movement, and if the pressure force exceeds 3,400N, it is judged as a dangerous movement.

For convenience of explanation, the present invention will describe workers by distinguishing between care recipients and caregivers.

The waist load includes (a) collecting location data of the caregiver and the caregiver using a motion capture system and IMU while performing the task, and (b) step (a) above. (c) measuring the waist load using the upper body position data collected in and (c) creating a waist load standard classification model using the data collected by the IMU.

Therefore, according to the lumbar load prediction method of the present invention and the terminal equipped with the application, the compressive force applied between the 5th lumbar vertebra and the 1st sacrum (L5/S1) is calculated using upper body movement data (kinematic data). can be measured, which has the effect of minimizing back injuries by allowing workers to recognize the load on their backs in real time while working.

In addition, according to the lumbar load prediction method of the present invention and the terminal equipped with the application, an algorithm was developed to measure the compressive force that occurs between the 5th lumbar vertebra and the 1st sacrum (L5/S1) during waist bending. Because it can be installed in the app, workers can recognize the load on their backs in real time while working, which has the effect of minimizing back injuries.

And according to the terminal equipped with the waist load prediction method and application of the present invention, the pressure force applied to the waist is predicted in real time through an IMU sensor worn on the body, and back injury prevention guidelines (pressure force of 3,400N or less) are predicted. Based on this, it is effective in preventing danger in advance by judging dangerous/safe movements and notifying workers.

1 is a main configuration diagram of a waist load measuring device according to an embodiment of the present invention,
Figure 2 is a flowchart illustrating a method for measuring waist load according to an embodiment of the present invention;
Figure 3 is a flowchart illustrating a method of displaying waist load measurement results using the application of the present invention;
Figure 4 is a diagram illustrating markers attached to the caregiver and the care recipient;
Figure 5 is a diagram showing an example of calculating the distance between the skin to which the marker is attached and the disc;
and
Figure 6 is a diagram illustrating a formula calculated based on the static balance formula.

The terms or words used in this specification and claims are not to be construed limited to their ordinary or dictionary meanings, and the inventor can appropriately define the concept of terms in order to explain his or her invention in the best way. It must be interpreted based on the meaning and concept consistent with the technical idea of the present invention.

Throughout the specification, when a part "includes" a certain element, this means that it may further include other elements rather than excluding other elements, unless specifically stated to the contrary. In addition, terms such as "...unit", "...unit", "module", and "device" used in the specification refer to a unit that processes at least one function or operation, which is implemented by a combination of hardware and/or software. It can be.

Throughout the specification, the term “and/or” should be understood to include all possible combinations from one or more related items. For example, “the first item, the second item and/or the third item” means not only the first, second or third item, but also the meaning that may be derived from two or more of the first, second or third items. It means a combination of all possible items.

Throughout the specification, identification codes (e.g., a, b, c, ...) for each step are used for convenience of explanation. The identification codes do not limit the order of each step, and each step Unless a specific order is clearly stated in the context, it may occur in a different order than the specified order. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the opposite order.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

Figure 1 is a main configuration diagram of a waist load measuring device according to an embodiment of the present invention, and Figure 2 is a flow chart for explaining a method of measuring waist load according to an embodiment of the present invention. As shown, the present invention The lumbar load measurement method and device uses upper body movement data (kinematic data) to measure the load generated on the waist between bending movements in real time, measuring the 5th lumbar vertebrae and 1st sacrum (L5/S1) between bending the waist. ) One feature is that the waist load is measured by the compressive force generated between the two.

In other words, the compressive force applied to the waist is predicted in real time through the motion capture system (VICON) and IMU sensor attached to the body, and the measured pressure force is measured by the U.S. National Institute of Safety and Health. This is about a technology that determines dangerous and safe movements and informs workers of them, compared to the (NIOSH) back injury prevention guidelines (pressure force of 3,400N or less).

In addition, the motion capture system (hereinafter referred to as VICON) uses VICON to measure movement data of the upper body, and at the same time collects information on acceleration, speed, and angular velocity using an IMU sensor. It allows features to be extracted from speed and angular velocity information.

In other words, the present invention collects upper body position data using VICON as kinematic information, and acceleration, velocity, and angular velocity data using IMU.

Referring to Figure 1, the device for measuring the lumbar load in real time by observing the user's movements according to an embodiment of the present invention is first attached to both sides or one side of the user's major joints and measures the bending angle and twist of the major joints. , and a marker 130 is attached to the user's body to measure location data, and an IMU sensor is attached to the user's body to measure acceleration and angular velocity.

In the present invention, the Vicon motion capture system is used to obtain location information of each joint part.

IMU (Inertial Measurement Unit) is an inertial measurement device, and the final value to be obtained by measuring inertia is used to accurately measure the tilt angle of an object, using an acceleration sensor and an angular velocity sensor (gyroscope). It consists of (6 axes) and some may include a geomagnetic sensor (Magnetometer).

An acceleration sensor instantly detects changes in acceleration (unit: m/s^2) due to changes in movement when dynamic forces such as acceleration force, vibration force, and impact force acting on an object occur.

By integrating the acceleration, which is the output value obtained from the sensor, the speed in the direction of movement of the object can be calculated, and by processing this, the position of the object can be found.

An angular velocity sensor (gyroscope) is a representative inertial sensor that measures angular velocity (unit: rad/s), which is the change in rotation of an object.

Gyroscopes use gravity to detect the Coriolis force. The change in vibration speed when gravity is applied can be calculated by calculating the angular speed and detecting the value by measuring the mass and vibration speed.

A geomagnetic sensor is a sensor that measures the strength (magnetic flux) and direction (angle offset relative to magnetic north) of the Earth's magnetic field.

When the direction an object is facing is aligned with the north direction and coincides with the Earth's magnetic field lines, the sensor's measured values have maximum and minimum values.

Therefore, caution must be taken with geomagnetic sensors as measured values may vary depending on the presence or absence of surrounding electric or magnetic fields.

These IMU sensors will be explained on the basis that they are attached to both arms, forearms, thighs, shins, and waist.

That is, the IMU includes the first sensor 141 attached to the forearm, the second sensor 142 attached to both arms, the third sensor 143 attached to the waist, the fourth sensor 144 attached to the thigh, and the second sensor 142 attached to the both arms. It operates to extract acceleration, velocity, angular velocity, etc. using the attached fifth sensor 145.

Additionally, in the present invention, since location information of each joint is required to know the body's center of gravity, the Vicon motion capture system is used to obtain location information.

The motion capture system shoots infrared rays at a marker attached to the body through an infrared camera, and the infrared rays reflected by the marker provide coordinates of the x, y, and z axes according to time in three dimensions.

In the present invention, location information of joint parts of the upper body is acquired from 17 markers attached to the caregiver (see FIG. 4) and 6 markers attached to the caregiver (see FIG. 4).

Figure 4 is a diagram illustrating markers attached to the caregiver and the care recipient, illustrating the locations where the markers are attached to the body parts of the caregiver and the caregiver.

For example, the caregiver attaches 17 markers at 2 locations on each left and right wrist, 2 locations each on the left and right elbows, 1 location each on the left and right shoulders, 3 locations on the pelvis, and 4 locations on the waist, and the caregiver attaches 17 markers on the front shoulder area at 3 locations. Markers are attached to three places on the back and shoulder area.

The control device 100 is configured to measure waist load using information collected from the marker.

To this end, the control device 100 includes a photographing unit 120 for photographing the user's movements, a device for calculating the amount of waist load in real time based on data obtained from the marker and the user's movements photographed by the photographing portion 120. It includes a control unit 110 and a lighting module 154 for controlling lighting for the user.

The photographing unit 120 consists of a camera module for photographing the user's movements, and may include a gimbal to maintain the camera module horizontally.

The control unit 120 includes an image analysis module for analyzing the image captured by the photographing unit 120, a camera and lighting control module for controlling the operations of the photographing unit 120 and the lighting unit 150, and an image analysis module for analyzing the image captured by the sensor. It may include a real-time lumbar load calculation module for calculating the user's lumbar load in real time based on the results and image analysis results.

Additionally, the control device 100 includes a marker 130 and a communication unit 160 for communicating with the external terminal 200.

Additionally, the control device 100 may store the measured waist load by event in the storage unit 170 for each measurer and transmit the event details to the administrator terminal 200.

In addition, the terminal 200 executes the application to receive and display event details occurring from the control device 100 or display content about the measured waist load to confirm whether riding is dangerous or safe through voice or vision. let it be

Even in this case, it is possible to prevent safety accidents in advance by drawing attention by outputting a safe movement or a dangerous movement through the terminal's speaker.

These applications are stored in the storage of the terminal, and the related apps are downloaded and stored through the App Store, etc., or the applications are downloaded and installed by connecting to a cloud server (not shown) that operates as a web server through a mobile communication network. .

Preferably, during the installation process of the app, you can register as a member and go through the server's authentication process to install the application.

Of course, you can simply sign up with your unique member ID, such as your e-mail address or password.

These apps can be downloaded and installed by connecting to a server that operates as a website, but you can also use the method of distributing the app appropriate for each OS by uploading it to the app store, regardless of whether it is IOS or Android.

The downloading and installation process of these apps, as well as the authentication steps, etc. are general, so detailed explanations are omitted.

Meanwhile, the terminal 200 is equipped with a short-range communication unit and can be paired when approaching the control device 100 and automatically run the application to check the waist load measurement details.

In this case, it is natural that the communication unit 160 of the control device must also be equipped with a short-distance communication function.

That is, when a terminal user equipped with a short-distance wireless communication function approaches the communication unit 160 of the control device, the paired application stored in the storage unit 170 and the application of the manager terminal 500 are activated, and event details or It can measure waist load in real time and display the results.

In this case, using beacons allows operation with a small amount of packet transmission, does not require pairing to connect two devices, and communicates at low power, so location can be recognized at a low cost compared to other short-range wireless communication technologies.

Beacons are non-contact and support long-distance communication up to 50m.

In addition, the location of the device can be determined within an error range of 5cm, and since both one-to-many and many-to-many services are possible, a variety of active services can be provided.

The beacon itself serves as a reference point for indicating the location, and the actual information is transmitted based on short-range communication technologies such as Bluetooth and infrared. It is preferable to use a Bluetooth beacon that combines Bluetooth.

Hereinafter, with reference to FIG. 2, the operation of a device for measuring lumbar load in real time by observing the user's movements according to an embodiment of the present invention will be described.

In other words, the lumbar load measurement method uses upper body movement data (kinematic data) to measure the compressive force that occurs between lumbar vertebrae 5 and sac 1 (L5/S1) during waist bending, and is added to the measured value. Depending on the movement, it can be judged as a safe movement or a dangerous movement and caution can be drawn to prevent safety accidents in advance.

The waist load measurement method collects location data of the caregiver and care recipient during the task (S110), measures the waist load using the upper body location data collected in step S110 (S120), and uses the IMU in step S130. Using the collected data, a classification model based on back load is created.

The task in step S110 is to transfer from the bed to the wheelchair.

The waist load is measured using the upper body position data collected in step S110 (S120).

Specifically, step S120 applies the position data collected in step S110 to the center of gravity ratio of the upper body segment to obtain the position of the lumbar disc between the 5th lumbar vertebra and the 1st sacral vertebra (L5/S1) (S121). , a step of calculating muscle pulling force and waist load using a static equilibrium equation (S122), and a step of classifying the waist load calculated in step S122 into dangerous movements and safe movements according to NIOSH standards ( S123).

In step S121, location information of each joint is required to know the center of gravity of the upper body, and the Vicon motion capture system is used to obtain the location information.

In addition, the motion capture system shoots infrared rays to a marker attached to the body through an infrared camera, and provides x, y, and z coordinates over time in three dimensions with the infrared rays reflected by the marker.

First, in order to know the center of gravity of the upper body, the center of gravity of the torso at the distance from the shoulder joint to the center of the hip joint, the center of gravity of both upper arms, and the center of gravity of both lower arms must be calculated, which is suggested by the above upper body position information and prior research. It is calculated by incorporating the center of gravity ratio (the ratio of the center of gravity of the torso, the distance from the shoulder joint to the center of the hip joint: 0.66; the ratio of the center of gravity of the upper arm: 0.436; the ratio of the center of gravity of the lower arm: 0.682).

The calculation formula is as follows:

CTRX = HX+0.66*(SX-HX)

CTRY = HY+0.66*(SY-HY)

CTRZ = HZ+0.66*(SZ-HZ)

CTRX is the X coordinate of the center of gravity of the torso, CTRY is the Y coordinate of the center of gravity of the torso, CTRZ is the Z coordinate of the center of gravity of the torso, HX is the means the Z coordinate of the center of the shoulder joint, SX means the X coordinate of the center of the shoulder joint, SY means the Y coordinate of the center of the shoulder joint, and SZ means the Z coordinate of the center of the shoulder joint.

For all data, the same equation is repeated three times to calculate the X, Y, and Z axes as above, so if you omit

CUAL = LSHOA+0.436*((LELEC+LEMEC/2)-LSHOA)

CUAR = RSHOA+0.436*((RELEC+REMEC/2)-RSHOA)

CFAL =

(LELEC+LEMEC/2)+0.682*((LWLEC+LWMEC/2)-(LELEC+LEMEC/2))

CFAR =

(RELEC+REMEC/2)+0.682*((RWLEC+RWMEC/2)-(RELEC+REMEC/2

Here, CUAL is the center of gravity of the left humerus, CUAR is the center of gravity of the right humerus, CFAL is the center of gravity of the left lower arm, CFAR is the center of gravity of the right lower arm, and LSHOA is the marker in front of the left shoulder blade. is the anterior right shoulder blade marker, LELEC is the left elbow lateral prominence marker, LEMEC is the left elbow medial prominence marker, RELEC is the right elbow lateral prominence marker, REMEC is the right elbow medial prominence marker, and LWLEC is the left wrist marker. LWMEC refers to the left wrist marker, RWLEC refers to the right wrist marker, and RWMEC refers to the right wrist marker.

Lastly, the center of gravity of the upper body is calculated by considering the weight ratio to the center of gravity of each joint, and is as follows (torso weight ratio of the distance from the shoulder joint to the center of the hip joint: 0.578; upper arm weight ratio: 0.011; lower arm weight ratio: 0.014)

COG

= ((0.022*(CFAR+CFAL))+(0.028*(CUAR+CUAL))+(0.578*CTR))/(0.628COG)

COG refers to the center of gravity of the upper body.

The center of gravity of the person being cared for (patient) is as follows.

PCOG = PRSHOA+PSTRN+PLSHOA+PLSHOP+PC7+PRSHOP/6

Here, PCOG is the patient's center of gravity, PRSHOA is a marker in front of the patient's right shoulder blade, PRSHOP is a marker in the back of the patient's right shoulder blade, PSTRN is a marker in front of the patient's sternum, and PLSHOA is a marker in front of the patient's left shoulder blade. And PLSHOP refers to the marker behind the patient's left shoulder blade, and PC7 refers to the marker of the patient's 7th cervical vertebra.

Additionally, in order to know the exact location (in other words, X, Y, Z axis coordinates) of the lumbar disc between the 5th lumbar vertebra and the 1st sacral vertebra (L5/S1), the relative position must be calculated based on the marker attached to the pelvis. . Because the marker provides information about the front and back of the pelvis, the relative position can be calculated based on the anatomical location.

Therefore, the distance between the skin where the marker is attached and the disc is calculated as shown in FIG. 5.

Figure 5 is a diagram showing an example of calculating the distance between the skin to which a marker is attached and the disc, and shows the distance from skin to disc center, spinal canal depth, and intervertebral disc length. (intervertebral disc length) is illustrated.

The calculation formula is as follows:

DSTD = SCD + (SDL)/2 + (IVDL)/2

Where DSTD = distance from skin to disc center; SCD = distance from skin to spinal canal; SDL = spinal canal depth; IVDL = intervertebral disc length.

Therefore, the x, y, and z axis position information of the disc center is calculated as follows.

DISCX = L5X + (DSTD/(L5ANTX-L5X))*(L5ANTX-L5X);

DISCY = L5Y + (DSTD/(L5ANTY-L5Y))*(L5ANTY-L5Y);

DISCZ = L5Z + (DSTD/(L5ANTZ-L5Z))*(L5ANTZ-L5Z);

Here, DISCX refers to the Hereinafter, for convenience of explanation, X, Y, and Z will be omitted since they represent X, Y, and Z coordinates, respectively. L5ANT refers to the parallel line ventral marker of lumbar vertebra 5, and L5 refers to the marker 5 to lumbar vertebra.

In step S122, muscle pulling force and waist load are calculated using a static equilibrium equation.

STATIC EQUILIBRIUM EQUATION refers to a formula calculated based on the theory that the sum of all forces and moments in a stationary state is 0. In other words, when a person stands, he or she can remain standing still because the strength of muscles, which are active structures, and the tendons and ligaments, which are passive structures, act to counteract the force of gravity that pulls towards the center of the Earth and cancel out gravity. Here, moment is a ROTATIONAL FORCE and is a force that rotates an object.

Figure 6 is a formula calculated based on the static balance formula.

In Figure 6, Fm = muscle force; d1 = distance from skin to disc center = DSTD; d2 = distance from disc center to COG; d3 = distance from disc center to patient's COG; wb = weight of caregiver's upper body; wp = means weight of patient.

Referring to FIG. 6, the static balance formula in step S122 is expressed as the equation below.

ΣM = 0;Fm*d1 = Wb*d2+Wp*d3

ΣF = 0;-Fm + Fc - Wbcos θ

Fm is the pulling force of the muscle, Fc is the back load, Wb is the weight of the caregiver's upper body, Wp is the weight of the patient, d1 is the distance between Fc and Fm, d2 is the distance between the center of gravity of the caregiver and the center of the disc, and d3 is the distance between the center of gravity of the caregiver and the center of the disc. The distance between the disc center and the caregiver's center of gravity, and θ refers to the caregiver's waist flexion angle.

Step S123 classifies the pressure force (ΣF) calculated in step S122 into dangerous movement and safe movement according to NIOSH standards.

In other words, based on the back injury prevention guidelines (pressure force of 3,400 N or less), if the measured pressure force is less than 3,400 N, it is judged as a safe movement, and if the pressure force exceeds 3,400 N, it is judged as a dangerous movement.

The determined result is stored in the storage unit 170 or transmitted to the terminal 200.

Step S130 creates a back load-based classification model using data collected by the IMU.

Step S130 includes extracting features from the acceleration and angular velocity information of the IMU collected in step S110 (S131) and generating a quadratic classification model using an RBF kernel support vector machine (radial basis function support vector machine) (S132). ) is specified.

In step S131, the IMU is attached to both arms, forearms, thighs, shins, and waist and extracts features using measured data.

In step S131, the feature extraction (windows size and lead time) is set to 0.1 to 0.5 seconds before the maximum back load, the data length used (windows size) is set to 0.1 to 0.5 seconds, and the extracted features are Extracts acceleration average, acceleration variance, velocity average, velocity variance, angular velocity average, and angular velocity variance as features.

These conditions can be applied in various ways, but appropriate conditions should be set from the investigator's perspective. For example, you can set the lead time at 0.3 seconds and the window size at 0.3 seconds, but depending on how you set it, the information and size of the data provided will vary, resulting in different accuracy of the model.

The first reason to limit the numbers as above is because there is no prior research on the timing of maximum back load, so it was applied based on information provided by fall detection research, and the lead time and necessary data to know the fall before the fall. The window size, which is the length of , is provided as above. The second reason is that if it is not limited as above, the data will become countless, so it is applied in a limited manner.

The extracted acceleration average, acceleration variance, velocity average, velocity variance, angular velocity average, and angular velocity variance are used as input data (input variables) for machine learning.

The output data (result variable) is two types: a dangerous version that exceeds 3,400 N and a safe version that does not exceed 3,400 N. The input data and output data are used as training and testing data, and through several loops, the model maximizes accuracy. The height sets the standard (parameter of the RBF kernel).

In addition, when new data (information with acceleration average, acceleration variance, velocity average, velocity variance, angular velocity average, and angular velocity variance) comes in, the model created based on the data (RBF kernel parameters) meets the model's standards. It will be possible to classify it accordingly and let you know whether it is a dangerous or safe world.

The lumbar load applied to the user's lumbar spine, calculated in real time as described above, not only measures the lumbar load when riding and informs the level of risk, but also estimates the labor intensity of the worker in the factory where the work is performed, and determines the estimated worker's It can be used as a guideline for developing work protocols for workers so that labor intensity does not exceed a certain level.

Although the present invention has been described in detail with respect to the described embodiments above, it is clear to those skilled in the art that various changes and modifications are possible within the technical scope of the present invention, and it is natural that such changes and modifications fall within the scope of the appended patent claims.

100: control device 110: control unit
120: recording unit 130: marker
140: IMU 150: Lighting unit
160: communication unit 170: storage unit
200: terminal

Claims (20)

In the waist load measurement method for measuring in real time the load generated on the waist between bending movements,
A method of measuring lumbar load, characterized in that the lumbar load is measured by the compressive force generated between the 5th lumbar vertebra and the 1st lumbar vertebrae (L5/S1) during back bending, using upper body movement data (kinematic data). .
In claim 1,
The upper body movement data (kinematic data) is
A method of measuring waist load, characterized by collecting upper body position data using a motion capture system (VICON) and data on acceleration, speed, and angular velocity using an IMU.
In claim 2,
The pressure force applied to the waist is measured in real time through an IMU sensor worn on the body, and the measured pressure force is based on the back injury prevention guidelines (pressure force of 3,400N or less) of the U.S. National Institute of Safety and Health (NIOSH). A method of measuring waist load that determines a safe movement if the measured pressure force is less than 3,400N and a dangerous movement if the measured pressure force exceeds 3,400N.
In claim 2,
(a) Collecting location data of the caregiver and the caregiver using a motion capture system and IMU while performing the task;
(b) measuring the waist load using the upper body position data collected in step (a); and
(c) creating a lumbar load-based classification model using data collected by the IMU;
A method of measuring lumbar load including.
In claim 4,
Step (a) above is
The Vicon motion capture system is used to obtain location information, and it is characterized by acquiring location information on the joints of the upper body from 17 markers attached to the caregiver and 6 markers attached to the caregiver. How to measure waist load.
In claim 4,
In step (a) above, the Task is
A method of measuring waist load, characterized by transferring from a bed to a wheelchair.
In claim 4,
Step (b) above is
(b-1) Applying the positional data collected in step (a) to the center of gravity ratio of the upper body segment to obtain the position of the lumbar disc between the 5th lumbar vertebra and the 1st sacral vertebra (L5/S1) ;
(b-2) calculating muscle pulling force and waist load using the static equilibrium equation; and
(b-3) classifying the waist load calculated in step (b-2) into dangerous movements and safe movements according to NIOSH standards;
A method of measuring lumbar load including. In claim 7,
The center of gravity ratio in step (b-1) above is 0.682, 0.436, and 0.66 for the forearm, upper arm, and trunk,
A method of measuring waist load in which the X, Y, Z coordinates of the body's center of gravity (CTR) are expressed in the equation below.
CTRX = HX+0.66*(SX-HX)
CTRY = HY+0.66*(SY-HY)
CTRZ = HZ+0.66*(SZ-HZ)
Here, CTRX is the X coordinate of the body's center of gravity, CTRY is the Y coordinate of the body's center of gravity, CTRZ is the Z coordinate of the body's center of gravity, HX is the It means the Z coordinate of the center, SX means the X coordinate of the center of the shoulder joint, SY means the Y coordinate of the center of the shoulder joint, and SZ means the Z coordinate of the center of the shoulder joint.
In claim 8,
The center of gravity of the upper body reflects the weight ratio in the center of gravity of each joint, and the torso weight ratio of the distance from the shoulder joint to the center of the hip joint is set to 0.578, the upper arm weight ratio: 0.011, and the lower arm weight ratio: 0.014. And the center of gravity (COG) of the upper body is a method of measuring waist load expressed by the equation below.
COG = ((0.022*(CFAR+CFAL))+(0.028*(CUAR+CUAL))+(0.578*CTR))/(0.628)
Here, COG is the center of gravity of the upper body, CUAL is the center of gravity of the left humerus, CUAR is the center of gravity of the right humerus, CFAL is the center of gravity of the left lower arm bone, and CFAR is the center of gravity of the right lower arm bone. .
In claim 9,
A method of measuring lumbar load in which the distance between the skin where the marker is attached and the disc is expressed by the equation below.
DSTD = SCD + (SDL)/2 + (IVDL)/2
Here, DSTD stands for distance from skin to disc center, SCD stands for distance from skin to spinal canal, SDL stands for spinal canal depth, and IVDL stands for intervertebral disc length.
A method of measuring waist load using DSTD, where the x, y, and z axis position information up to the disc center is expressed in the equation below.
DISCX = L5X + (DSTD/(L5ANTX-L5X))*(L5ANTX-L5X);
DISCY = L5Y + (DSTD/(L5ANTY-L5Y))*(L5ANTY-L5Y);
DISCZ = L5Z + (DSTD/(L5ANTZ-L5Z))*(L5ANTZ-L5Z);
In claim 7,
The static balance formula in step (b-2) above is
Waist load measurement method expressed by the equation below.
ΣM = 0;Fm*d1 = Wb*d2+Wp*d3
ΣF = 0;-Fm + Fc - Wbcos θ
Fm is the pulling force of the muscle, Fc is the back load, Wb is the weight of the caregiver's upper body, Wp is the weight of the patient, d1 is the distance between Fc and Fm, d2 is the distance between the center of gravity of the caregiver and the center of the disc, and d3 is the distance between the center of gravity of the caregiver and the center of the disc. The distance between the center of the disc and the center of gravity of the caregiver, and θ refers to the flexion angle of the caregiver's back.
In claim 4,
Step (c) above is
(c-1) extracting features from acceleration and angular velocity information collected by IMU;
(c-2) generating a quadratic classification model using an RBF kernel support vector machine (radial basis function support vector machine);
A method of measuring lumbar load including.
In claim 13,
In (c-1), the IMU is a method of measuring waist load, characterized in that it is attached to both arms, forearm, thighs, shins, and waist.
In claim 13,
In (c-1) above, feature extraction (windows size and lead time) is
The standard for the peak back load (lead time) is 0.1 to 0.5 seconds before, the data length used (windows size) is 0.1 to 0.5 seconds, and the extracted features are acceleration average, acceleration variance, velocity average, velocity variance, and angular velocity. A waist load measurement method characterized by extracting the average and angular velocity distribution as features.
In the lumbar load measuring device that measures the lumbar load using the compressive force generated between the 5th lumbar vertebra and the 1st sac (L5/S1) during lumbar bending using an IMU sensor and a motion detection system,
To obtain location information, the Vicon motion capture system is used, and the pressure force is calculated by acquiring location information on the joints of the upper body from 17 markers attached to the caregiver and 6 markers attached to the caregiver. A device for measuring waist load.
In claim 16,
A control unit that measures the compressive force occurring between lumbar 5 and sacral 1 (L5/S1) using information collected from the marker;
Including,
The control unit
By applying the collected positional data to the center of gravity ratio of the upper body segment, the position of the lumbar disc between lumbar vertebra 5 and sacral 1 (L5/S1) was obtained, and the static equilibrium equation was used to determine the position of the lumbar disc. A lumbar load measuring device that calculates muscle pulling force and lumbar load and then classifies lumbar load into dangerous and safe movements according to NIOSH standards.
In claim 17,
The caregiver attaches 17 markers at 2 locations on each left and right wrist, 2 locations each on the left and right elbows, 1 location each on the left and right shoulders, 3 locations on the pelvis, and 4 locations on the waist, and the caregiver attaches 17 markers on 3 locations on the front and back shoulder areas. A device for measuring waist load, characterized by attaching markers at three locations.
In claim 18,
The control unit
An image analysis module for analyzing images captured by the imaging unit, a camera and lighting control module for controlling the operation of the imaging unit and the lighting unit, and the user's lumbar spine in real time based on the analysis results from the sensor and the image analysis results. A lumbar load measurement device including a real-time lumbar load calculation module to calculate the compressive force occurring between L5 and L5/S1.
In claim 19,
The control unit
A device for measuring waist load, characterized in that it is configured to store the measured compressive force by event in a storage unit for each measurer and transmit the event details to the administrator terminal.

Images