KR20210110185A - Concave chest diagnosis algorithm and system - Google Patents

Abstract

Disclosed are an algorithm and system for diagnosing a concave chest. The algorithm for diagnosing the concave chest of the present invention comprises: (a) a step of measuring a chest part using a concave chest diagnosis device; (b) a step of obtaining the raw data for diagnosing the concave chest from the data measured in step (a); (c) a step of pre-processing the raw data obtained in step (b); and (d) a step of diagnosing the concave chest by analyzing the pre-processed data, thereby enabling accurate diagnosis because of being measured using a depth camera that photographs based on the RGB and IR-based depth information.

Description

Concave chest diagnosis algorithm and system {CONCAVE CHEST DIAGNOSIS ALGORITHM AND SYSTEM}

The present invention relates to an apparatus for diagnosing pectoralis concave, and more particularly, to an apparatus for diagnosing pectoralis concave, and more particularly, to an apparatus and method for diagnosing pectoralis concave by scanning the shape of the chest.

Concave breast is a congenital anomaly in which the pronotum is excessively depressed.

The disease puts pressure on the heart's organs and can cause a variety of symptoms, including palpitations (a feeling of a rapid heartbeat), cardiac arrhythmias, and difficulty breathing.

Methods for measuring the concave chest include a measurement method using a physical examination, a measurement method using computed tomography, and a diagnostic method using an optical markerless 3D measurement system.

First, the measurement methods using a physical examination are typically chest cyrtometry and anthropometric index. Chest cyrtometry is a method to directly measure the chest circumference through a metric tape. Anthropometric index is the longest diameter at 1/3 point near the outside of the sternum It is a method of calculating the ratio of the deepest depth at this point to the Haller Index and Anthropometric index, which are concave chest measurement indexes.

As an index for measuring concave chest using computed tomography (CT), the chest wall deformation was performed using the Haller Index (A/C), which is the ratio of the lateral length of the chest in the frontal direction (A) and the length between the sternum in the spine (C). As one of the standard methods used to evaluate the degree of chest wall deformity, it has the advantage of being able to objectively and accurately evaluate chest wall anomalies.

The diagnostic method through the optical markerless 3D measurement system is to take pictures in a limited space and use 4 cameras to scan the upper part of the body. when compared to a similar Haller index the value of I and I _{3ds 3ds} on the CT of the chest using the same, are known tendency is shown similar results.

Among these conventional measurement methods, the first measurement method using a physical examination has a problem in that it is difficult to make an accurate diagnosis depending on the operator because it is diagnosed by the operator's empirical judgment. There is a problem that there is a risk of exposure to radiation, and the diagnostic method using an optical markerless 3D measurement system has a disadvantage in that it takes time and money for diagnosis and installation because four cameras must be used in a limited space.

KR Registered Patent Publication No. 10-1090375 (2011.11.30)

An object of the present invention to solve this problem is to provide a concave chest diagnosis algorithm and system using a depth camera that takes pictures based on RGB and IR-based depth information.

Another object of the present invention is to provide an algorithm and system for diagnosing concave chest that require a shorter imaging time (about 30 seconds) compared to a computed tomography method and enable accurate diagnosis compared to a physical examination.

Another object of the present invention is to provide an algorithm and a system for diagnosing pterygoid by scanning the shape of the chest with a single camera and observing convex chest depression.

The concave chest diagnosis algorithm of the present invention for solving these problems includes (a) measuring the chest using a concave chest diagnosis device, and (b) raw data for diagnosing concave chest from the data measured in step (a). It can be made by configuring it to include the steps of obtaining a septum, (c) pre-processing the raw data obtained in step (b), and (d) analyzing the pre-processed data to diagnose concave chest.

In addition, in step (b), the measured raw data for the chest is a chest (PE.stl) file edited with raw data by removing the parts unnecessary for analysis except for the chest among the captured data such as arms, tops, etc. made to include

In addition, step (c) rotates the PE.stl file so that the cross section of the chest and the ground are perpendicular for analysis in the transvers plane, and finds the angle at which the area of the orthographic projection in a specific plane is maximized. It may include the steps of rotating at a corresponding angle and obtaining a chest offset value based on the bottom surface of the chest file rotated at the maximum rotation angle.

In addition, in step (d), the remeshed.stl data is acquired and applied to an analysis algorithm to analyze and determine whether there is a concave chest.

On the other hand, the concave chest diagnosis apparatus of the present invention for solving this problem is the depth camera, the bed on which the subject is located, and the depth camera so that the camera can be photographed in a cross section of the chest of the subject from the upper part spaced a certain distance from the bed. It can be configured to include a track that moves in an oval shape, and a diagnostic unit that acquires image data captured by the depth camera and diagnoses the chest cavity by using a chest chest diagnosis algorithm.

In addition, the diagnosis unit acquires raw data for diagnosing concave chest from the chest measurement data photographed using the camera, pre-processes the obtained raw data, and then analyzes the pre-processed data to diagnose the presence of concave chest. .

In addition, the diagnostic unit If the ratio of chest width to height is 3.2 or more, it is judged that there is a concave chest and treated as a severe patient requiring surgery. to be judged as

Therefore, according to the concave chest diagnosis algorithm and system of the present invention, accurate diagnosis is possible because measurement is performed using a depth camera that takes pictures based on RGB and IR-based depth information.

In addition, according to the concave chest diagnosis algorithm and system of the present invention, a shorter imaging time (about 30 seconds) compared to a computed tomography method and an accurate diagnosis are possible compared to a physical examination.

In addition, according to the concave chest diagnosis algorithm and system of the present invention, there is an effect that can observe the concave chest depression by scanning the shape of the chest with one camera.

1 is a main configuration diagram of the present invention concave chest diagnosis apparatus
2 is a basic flowchart for explaining the method for diagnosing concave chest of the present invention;
3 is a reference view for explaining the operation of the device for measurement;
Figure 4 is a view for explaining the process of creating a chest file edited raw data,
5 is a view of a state in which a part other than the chest is removed from the captured image;
6 is a view for explaining the process of rotating the chest file for each axis;
7 is a reference view of a state in which the chest file is rotated for each axis;
8 is a reference view for explaining the process of processing the chest file based on the floor surface during the pre-processing process;
9 is a reference view of a process of obtaining a chest offset value;
10 is a view for explaining a process of reducing distortion through remeshing;
11 is a reference view for explaining the remeshing process;
12 is a flowchart for explaining a detailed data analysis process;
13 is a flowchart for explaining the process of forming point coordinates for observing the depth of depression;
14 is a reference view of the point coordinate formation process;
15 is a flowchart for explaining the process of dividing the horizontal and vertical quarters for division of the central region of the sternum;
16 is a flowchart for explaining the process of expressing the extracted point data as a graph;
17 is a reference diagram illustrating data interpolated from the extracted points as a graph;
18 is a reference view for explaining the error of the lowest point;
and
19 to 21 are reference views of an analysis process using an index.

The terms or words used in the present specification and claims are not to be construed as limited in their ordinary or dictionary meanings, and on the principle that the inventor can appropriately define the concept of the term in order to best describe his invention. It should be interpreted as meaning and concept consistent with the technical idea of the present invention.

Throughout the specification, when a part "includes" a certain element, it means that other elements may be further included, rather than excluding other elements, unless otherwise stated. In addition, terms such as “…unit”, “…group”, “module”, and “device” described in the specification mean a unit that processes at least one function or operation, which is implemented by a combination of hardware and/or software 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 meaning of "the first item, the second item and/or the third item" may be presented from the first, second, or third item as well as two or more of the first, second, or third items. A combination of all possible items.

In each step throughout the specification, identification symbols (eg, a, b, c, ...) are used for convenience of description, and identification codes do not limit the order of each step, and each step is Unless the context clearly dictates a specific order, the order may differ from the stated order. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

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

1 is a main configuration diagram of the apparatus for diagnosing concave chest according to the present invention, and one feature of the apparatus for measuring chest concave of the present invention is to take pictures while moving along a track based on the chest in order to accurately measure the chest of a patient.

To this end, the concave chest diagnosis apparatus 100 of the present invention includes a depth camera (hereinafter referred to as a camera) 150 for photographing based on RGB and IR-based depth information, a bed 110 on which the subject is located, and a bed An elliptical track 140 that moves the camera 150 in an elliptical shape so that the camera 150 can take a cross-section of the subject's chest from the upper part spaced from the 110, and the image data taken by the camera 150 and a diagnosis unit 160 for acquiring and diagnosing the scaly chest by using a dentate diagnosing algorithm.

The camera 150 is a camera capable of calculating the depth value of each pixel of an image, and unlike the 2D method, it uses two camera modules to calculate the depth of a pixel using various techniques to display a 3D image. am.

The camera 150 used in the present invention is based on an RGB-IR sensor, and this RGB-IR sensor uses R (Red, Red), G (Green, Green), B (Blue, Blue) pixels in the visible light region. It has a structure of acquiring a color image of , and acquiring an infrared image through an IR (Infrared Ray, Infrared) pixel. That is, a single RGB-IR sensor is constituted by a combination of R, G, B, and IR pixel arrays.

The RGB-IR sensor structure consists of a typical RGB sensor composed of one R, B pixel and two G pixels (this is widely known as the 'Bayer sensor structure'), by changing the one G pixel to an IR pixel. will do

Referring to the drawings, as shown by the dotted arrow, the camera 150 moves in an oval shape along the elliptical track 140 so that the camera 150 can take an oval-shaped image centered on the subject's chest on the bed 110 .

That is, since the camera 150 rotates along the elliptical camera track 140 in the direction of the arrow and operates to measure the chest of the subject on the bed, it can be seen that the focus is focused to accurately measure the chest.

The measuring device of the present invention uses an elliptical track to photograph the chest at the same distance in order to focus in the existing 3D scan method for photographing 360 degrees, and has a feature that can reduce the recording time because it does not photograph 360 degrees.

Shooting is measured at 30 frames per second.

The diagnosis unit 160 receives the image captured by the camera 150 and displays it on a display (not shown), and at the same time diagnoses the chest cavity from the captured image data by using the chest diagnosis algorithm, and displays the result on the display. to display

The diagnosis algorithm mounted in the diagnosis unit 160 will be described in detail below.

Reference numeral 130 denotes an upper frame for supporting the track 140 , reference numeral 120 denotes a support frame, and reference numeral 170 denotes a lower frame.

Referring to the basic flowchart for explaining the method for diagnosing concave chest of the present invention in FIG. 2 , first, measuring the chest through the above-described device for diagnosing concave chest (S100) and data measured in step S100 for diagnosing concave chest A step of acquiring raw data (200) and a step of pre-processing the raw data obtained in step S200 (300), and a data analysis step (400) of diagnosing concave chest by analyzing the pre-processed data are made.

Step S100 is taken by focusing on the chest using the oval track 140 so that the chest of the subject of the bed 110 can be photographed at the same distance using the concave chest diagnosis device.

Referring to the reference drawing for explaining the operation of the measuring device of FIG. 3 , the subject is placed on the bed and an image is taken based on the focus of the chest along the camera movement path (red arrow) on the track 140 .

A chest file is generated by acquiring raw data in step S200 from the image data measured in step S100.

Hereinafter, a process of generating a chest file will be described with reference to the drawings.

FIG. 4 is a view for explaining a process of creating a chest file by editing raw data, and FIG. 5 is a diagram of a state in which a part other than the chest is removed from a photographed image.

Referring to the drawing, an edited chest (PE.stl) file is created by removing parts unnecessary for analysis except for the chest among the captured data, such as arms and upper, from the measured raw data of the chest.

The stl file is composed of three vertices that make up a triangle and a corresponding vector.

The chest (PE.stl) file generated in step S100 performs a preprocessing process (S300) for analysis in the transverse plane.

Hereinafter, a pre-processing process will be described with reference to the drawings.

FIG. 6 is a view for explaining a process of rotating the chest file for each axis, and FIG. 7 is a reference view of a state in which the chest file is rotated for each axis.

Referring to FIG. 6 , the pre-processing process (S200) rotates the PE.stl file so that the cross-section of the chest and the ground are perpendicular for analysis in the transvers plane, and the area of the orthographic projection on a specific plane is the maximum. Find the angle that will form , and rotate it by that angle.

Step S310 is a step of performing rotation and orthographic projection based on the X-axis. After rotating at a rotation angle θx having the widest orthographic projection, the process proceeds to the step S320 of performing rotation and orthographic projection based on the Y-axis. In step S320, the area of the orthographic projection rotates at the widest rotation angle θy (S320), and in the step S330 of performing rotation and orthographic projection based on the Z axis (S330), the area of the orthographic projection rotates at the widest rotation angle θz.

Referring to the reference drawing of the state in which the chest file of FIG. 7 is rotated for each axis, the rotation result based on the X-axis is rotated by 171 degrees, the Y-axis is rotated 360 degrees, and the Z-axis is rotated 19 degrees, so that it is orthogonal in a specific plane. It can be seen that the area of is rotated by the angle that will achieve the maximum.

In steps S310 to S330, each axis is rotated at a rotation angle (θx, θy, θz ) having the widest orthographic area, and then corrected to data based on the floor (bed) (S340).

Referring to the reference drawing for explaining the process of processing the chest file based on the floor surface during the pre-processing process of FIG. 8 and the reference drawing for the process of obtaining the chest offset value of FIG. 9, the bed part to be used as the floor reference to analyze the data data (bed.stl) of

After pre-processing the reference to calculate the height of the chest based on the floor surface, edit the bed.stl by the rotation angle and movement distance that edited the chest file in steps S310 to S330, and the height based on the surface to calculate

The plane data (Dx, Dy, Dz) is acquired by shifting each point to the center based on each axis (S343), and θx, θy, θz and the obtained Dx, Dy, Dz are rotated based on the stored bed.stl data. and shift so that the y-axis becomes the height of the chest (S342).

To do this, first set the offset value using bed.stl to set the offset value of the floor, rotate it using the obtained rotation angles of θx, θy, θz, and then calculate the distance from the zero point on each axis to extract the offset value to extract the bed offset formed along the x and y axes.

That is, the chest offset (edited.stl) formed in the x and y axes is obtained by applying the offset value (bed offset) obtained through the bed.stl file to PE.stl.

By remeshing the chest offset obtained in step S342, distortion deformation is reduced to increase the reliability of the edited file (S350).

Referring to the drawing for explaining the process of reducing distortion through the remeshing of FIG. 10 and the reference drawing for explaining the remeshing process of FIG. 11 , first, the data edited in step S342 (edited.stl) is stored (S351) , remeshing is performed (S352) to obtain and store remeshed.stl data (S353).

11 is a photograph of changes in the cross section of the analysis point, and when the file before remesh at the top is compared with the file after remesh at the bottom, it can be seen that distortion deformation is reduced through remeshing.

When the remeshed.stl data is obtained through step S353, the step of diagnosing the presence or absence of concave chest is performed by applying it to the analysis algorithm (S400).

Referring to the flowchart for explaining the detailed data analysis process of FIG. 12 and the flowchart for explaining the process of forming point coordinates for observing the dent depth of FIG. 13, step S410 is a voxelize step (S410) and a cross-sectional point of the lowest point An extraction step (S420) and a step (S430) of obtaining tomography data for the cross section are performed.

The voxelize step (S410) is a step in which the edited stl file (S411) consisting of three vertices and the corresponding vector is voxelized (S412) to obtain a point in 3D (S412) to call the point coordinates. The voxelized file is a face colored plot. The depth of the depression can be observed.

Referring to the reference diagram of the point coordinate formation process of FIG. 14 , the upper part is a figure obtained by Voxelizing the remeshed.stl data, and the lower part is a figure displayed as a face colored plot, so that the depth can be known by color.

The data Voxelized in step S412 proceeds to the step of extracting the cross-sectional point of the lowest point (S420).

Referring to the flowchart for explaining the process of dividing the horizontal and vertical quarters for division of the central region of the sternum of FIG. 15, the step of extracting the cross-sectional point of the lowest point in step S420 is to first divide the analysis plane into 1/4, and (S421), extracting the lowest point of the divided surface (S422), and performing the step of extracting both points of the cross section of the lowest point (S423).

In step S421, when the lowest point is obtained from the edited PE.stl file, the part corresponding to the rib is extracted as the lowest point, and thus 1/4 division is performed horizontally and vertically to divide the central part corresponding to the sternum.

In the step (S423) of extracting both cross-sections of the lowest point, each cross-section of the patient is analyzed using the Haller index, and the cross-section of the chest is extracted with the largest ratio of the horizontal length of the sternum to the height of the sternum.

When analyzing the cross section corresponding to the smallest height of the sternum, the horizontal length of the chest and the height of the sternum are the largest, so the lowest point of the dividing plane is extracted and analyzed.

Referring to the flowchart for explaining the process of expressing the extracted point data in a graph of FIG. 16 , the step of obtaining tomography data for the cross section (S430) is a step of analyzing the cross section using the extracted point data (S431) ) and extracting the index (S432).

The step of analyzing in the cross section (S431) is a reference drawing illustrating the interpolated data from the extracted points of FIG. 17 as a graph. It is a drawing marked with

The interpolated data is graphed with 5 point data extracted as a cross section from the remeshed.stl data.

Also, the interpolation error is extracted using the error rate between the lowest point of the interpolated graph and the lowest point of the existing graph.

Referring to the reference diagram for explaining the error of the lowest point of FIG. 18, it can be seen that the error rate between the lowest point of the interpolated graph and the lowest point of the existing graph can be expressed as the error rate of the height and the error rate of the width.

In the drawing, it can be seen that the error rate of the height is 1.77348*10 ^-6 % and the error rate of the width is 8.3726*10 ^{-5 %.}

In the step of extracting the index (S432), referring to the reference drawings of the analysis process using the index in FIGS. 19 to 21, the depth point (Deepest Point (De) " Dmaxp-Dminp” is calculated and displayed, the External Hallers Index (EHI) is “Dw/Sminp” and the External Correction Index (ECI) is “(Dh-De)/Dh*100”.

20 is a diagram of obtaining a tomography index for a cross section as a reference diagram of an analysis process using an index. Referring to the drawings, the Deepest Point (De) is 17.29711, the External Hallers Index (EHI) is 2.93934, and the External Correction Index is (ECI) is marked 69.81063dmfh.

In addition, referring to the reference drawing of the analysis process using the index of FIG. 21 , the depth is indicated by color as a face colored plot.

In the present invention, if the External Hallers Index (EHI), expressed as the ratio of the chest width to the height, is 3.2 or more, it is considered a severe patient requiring surgery, and less than 3.2 is the normal group. Or, it is judged as a mild patient who does not require surgery and the depth of the depression is not high.

By using such an analysis method, the present invention optimizes the existing analysis algorithm and analyzes the analysis time that takes about 13 minutes to within 1 minute, and enables analysis at the cross-section of the maximum depression point, as well as using the existing Haller Index. This made it possible to measure.

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

100: concave chest diagnosis device 110: bed
120: support frame 130: upper frame
140: Track 150: Depth Camera
160: diagnostic unit 170: lower frame

Claims (14)

(a) measuring the chest using a concave chest diagnosis device;
(b) acquiring raw data for diagnosing concave chest from the data measured in step (a);
(c) pre-processing the raw data obtained in step (b); and
(d) analyzing the pre-processed data to diagnose concave chest;
Concave chest diagnosis algorithm comprising a. The method according to claim 1,
The concave chest diagnosis device in step (a) is
Depth camera;
the bed on which the subject is located;
A track for moving the Depth camera in an elliptical shape so that the camera can take a cross section of the chest of the subject at the upper part spaced apart from the bed by a certain distance; And
a diagnosis unit that acquires image data captured by the depth camera and diagnoses serratus anterior chest using a convex chest diagnosis algorithm;
Concave chest diagnosis algorithm comprising a. The method according to claim 1,
Step (b) is
The measured raw data for the chest is a concave chest diagnosis algorithm that includes the step of creating a chest (PE.stl) file in which the raw data is edited by removing the parts unnecessary for analysis except for the chest among the captured data such as arms and upper chest. 4. The method according to claim 3,
Step (c) is
(c-1) Rotate the PE.stl file so that the cross-section of the chest and the ground are perpendicular for analysis in the transvers plane, find the angle at which the orthographic area in a specific plane is the maximum, and find the corresponding angle rotating with; and
(c-2) obtaining a chest offset value based on the bottom surface of the chest file rotated at the maximum rotation angle in step (c-1);
Concave chest diagnosis algorithm comprising a. 5. The method according to claim 4,
The step (c-1) is
It includes the steps of performing rotation and orthographic projection based on the X axis, performing rotation and orthographic projection based on the Y axis, and performing rotation and orthographic projection based on the Z axis, wherein the widest rotation angle (θx, Concave chest diagnosis algorithm, characterized in that rotation by θy, θz). 5. The method according to claim 4,
The step (c-2) is
Acquiring the bed data (bed.stl) of the bed area to be used as the floor standard;
Set the offset value using bed.stl, rotate it using the obtained rotation angle of θx, θy, θz, calculate the distance from the zero point on each axis, extract the offset value, and form it as the x, y axis Steps to extract the bed offset and
Applying the offset value (bed offset) obtained through the bed.stl file to PE.stl to obtain a chest offset (edited.stl) formed in the x and y axes;
Concave chest diagnosis algorithm comprising the step of reducing distortion deformation through remeshing the chest offset (edited.stl) obtained in the above step. 7. The method of claim 6,
Step (d) is
A concave chest diagnosis algorithm for acquiring the remeshed.stl data and applying it to an analysis algorithm to analyze and determine whether there is a concave chest. 8. The method of claim 7,
Step (d) is
If the ratio of chest width to height is 3.2 or more, it is judged that there is a concave chest and treated as a severe patient requiring surgery. Concave chest diagnosis algorithm that judges by
Depth camera;
the bed on which the subject is located;
A track for moving the Depth camera in an elliptical shape so that the camera can take a cross section of the chest of the subject at the upper part spaced apart from the bed by a certain distance; And
a diagnosis unit that acquires image data captured by the depth camera and diagnoses serratus anterior chest using a convex chest diagnosis algorithm;
A device for diagnosing concave chest comprising a. 10. The method of claim 9,
the track is
A concave chest diagnosis device comprising left and right elliptical tracks so that the camera can take an oval-shaped image of the subject's chest on the bed. 10. The method of claim 9,
The diagnostic unit
An apparatus for diagnosing concave chest by acquiring raw data for diagnosing concave chest from chest measurement data taken using the camera, pre-processing the obtained raw data, and then analyzing the pre-processed data to diagnose the presence or absence of concave chest. 12. The method of claim 11,
The diagnostic unit
The measured raw data for the chest is a concave chest diagnosis device that includes the step of creating a chest (PE.stl) file in which raw data is edited by removing parts unnecessary for analysis except for the chest among the captured data such as arms and upper chest. 13. The method of claim 12,
The diagnostic unit
Rotate the PE.stl file so that the cross section of the chest and the ground are perpendicular for analysis in the transvers plane, find the angle at which the orthogonal area is maximized on a specific plane, and rotate it to the corresponding angle, Acquire chest offset value based on the chest file rotated at the maximum rotation angle based on the bottom surface, and reduce distortion deformation through remeshing the acquired chest offset (edited.stl) to acquire remeshed.stl data, and apply it to the analysis algorithm A device for diagnosing concave chest by analyzing and judging the presence or absence of concave chest. 14. The method of claim 13,
The diagnostic unit
If the ratio of chest width to height is 3.2 or more, it is judged that there is a concave chest and treated as a severe patient requiring surgery. A device for diagnosing concave chest.

Images