KR101695079B1 - Apparatus and Method for Measuring Cobb angle - Google Patents

Abstract

 The present invention relates to an apparatus and method for measuring a cob angle, which can automatically measure a cobb angle for diagnosis of scoliosis using an x-ray imaging apparatus and an auxiliary device for x-ray imaging with a grid. The cob angle measuring apparatus includes a setting unit for setting an X-ray image of the patient as an image necessary for diagnosis of scoliosis, a vertebral outline extracting unit for extracting a vertebral outline in the set image to detect a straight line of the vertebra plate, And a cob angle measuring unit measuring an angle between all the straight lines of the spinal column and measuring an angle having the largest angle among the straight lines. According to the present invention, it is possible to automatically measure the cobb angle and to diagnose scoliosis more accurately.

Description

Technical Field [0001] The present invention relates to an apparatus for measuring a cob angle,

The present invention relates to a claw-angle measuring apparatus, and more particularly, to a claw-angle measuring apparatus for automatically measuring a claw angle for measuring a degree of curvature of a scoliosis using a patient's X- And a method thereof.

Scoliosis, one of the vertebrae diseases, is a disease in which the vertebra curves to the left and right to change its shape. Depending on the degree of curvature, it causes cosmetic problems and pain. In case of severe disease, surgical treatment is also required.

A method of measuring the cobb angle is widely used as a method of measuring the curvature of scoliosis. These cobble angles measure the degree of curvature of the scoliosis, and the measured cobb angle provides an important factor for the physician to determine what treatment is needed for the patient. For example, if the angle of curvature is more than 10 °, diagnosis is made as scoliosis. If it is more than 40 °, it is diagnosed that correction surgery is necessary. If the deformation is severe, the surrounding organs may be displaced or pressed to cause dysfunction and shorten the life span.

1 is a view for explaining a method of measuring a cob angle in general.

In this case, the upper spinal plate and the lower spinal plate, at which the tilting starts to be started with reference to the concave portion of the curvature, are respectively determined, and a line parallel to each upper end of the upper spinal plate and the lower spinal plate is drawn, The intersection angles of the lines are measured at the cob angle.

However, in order to measure the cob angle by the above-described method, a separate device is required.

In addition, there is a need to designate the upper spinal plate and the lower spinal plate for measuring the claw angle, and a line parallel to the upper spinal plate and the lower spinal plate. In this case, since the physician designates the spinal line directly in the X-ray image, there is a problem in that an error of the cob angle may occur. This leads to another problem of poor accuracy in the diagnosis of scoliosis.

Korean Patent Publication No. 10-2013-0102784 (Feb. 23, 2013. 23. Scoliosis Diagnosis System and Scoliosis Diagnosis Method) Korean Patent Laid-Open No. 10-2010-0093463 (Aug. 25, 2010)

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a Cobb angle measuring apparatus and method for improving the accuracy of diagnosis of scoliosis by minimizing the error of the Cobb angle.

It is another object of the present invention to provide a cob angle measuring apparatus and a method for measuring a cob angle by using only a spinal radiography image file.

According to an aspect of the present invention, there is provided a method for diagnosing spine scoliosis, comprising: setting a X-ray image of a patient as an image necessary for diagnosis of scoliosis; A spine contour extracting unit for extracting a contour of the spine in the set image so as to detect a straight line of the spine plate; And a cob angle measuring unit measuring an angle between all the straight lines of the spinal column and measuring a cob angle with an angle of the largest angle among the straight lines.

Wherein the setting unit comprises: a conversion unit for converting a size and a file format of the X-ray image; A removal unit for removing noise of the converted image; And an extracting unit for extracting only information of a predetermined number of pixels from the center of the noise-removed image to the left and right.

The pixel information extracted by the extraction unit may be set to 60 pixels on the left or right, or may be changed according to the specification of the setting unit or the X-ray photographing apparatus.

The spine contour extracting unit may include a canny edge extracting algorithm for detecting a contour line of the vertebrae in an image set by the setting unit; A contour detection algorithm for detecting a spine contour image by connecting the detected contour lines; And a hough transform algorithm for detecting a straight line of the vertebrae.

The contour detection algorithm uses the width and shape information of the connection area connecting the outline lines, and recognizes the vertebral area normally when the detected connection areas are in the stepping leg shape at regular intervals.

According to another aspect of the present invention, there is provided a method for setting an X-ray image of a patient, Extracting an outline of the vertebrae from the set image; Detecting a connection area connecting the detected outline line; Recognizing a vertebral region using the connection region; Detecting straight lines of the vertebrae included in the recognized vertebral region; Measuring the angle of the straight lines; And determining a largest angle among the measured angles as a cob angle.

The contour line extraction uses a canny edge extraction algorithm, the connection region detection uses a contour detection algorithm, and the straight line detection uses a hough transform algorithm.

The step of recognizing the vertebral region may include: a first calculation step of obtaining the width of each of the connection regions; A second calculating step of obtaining a width average of the connecting areas; A first determination step of determining the shape of the connection areas; And a second determination step of determining the vertebral region to be normal when the shape of the vertebral body is a spinal shape and a staggered leg shape in which the connection regions maintain a constant gap.

The connection area having the width average greater than or less than twice the standard deviation is excluded from the recognition step of the vertebral region.

If the difference in the interval between the connection regions exceeds a half of the smallest spinal region in the second determination step, the spinal region is not recognized.

The claw angle measuring apparatus and method according to the present invention has the following effects.

In the present invention, an x-ray image taken using an assistant device for x-ray imaging with a grid is set as an image for spine extraction, and a cobble angle is automatically measured from a set image using a series of algorithms.

Therefore, there is no need for a separate device for the conventional cob angle measurement. In addition, it is possible to reduce the error of the cob angle compared with the direct specification of the straight line of the spine for the cob angle measurement, Effect can be expected.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a view for explaining a method of measuring a cob angle in general;
2 is a diagram illustrating a claw-angle measuring apparatus according to a preferred embodiment of the present invention.
FIG. 3A is a perspective view of an x-ray imaging assistant used in the present invention,
Figure 3B illustrates one embodiment of a grid,
Fig. 4 is a schematic view of the claw-
FIG. 5 is a flowchart illustrating a method of measuring a claw angle using the claw angle measuring apparatus of FIG.
Fig. 6 is a diagram showing an example of image information according to a cob angle measurement method

The present invention relates to a technique for automatically diagnosing scoliosis by accurately measuring a cobble angle using a spinal x-ray image of a patient photographed using a grid-assisted x-ray imaging apparatus as a basic technical point do.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of a claw-angle measuring apparatus and method according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 2 is a view illustrating a claw-angle measuring apparatus according to a preferred embodiment of the present invention.

Referring to FIG. 2, the claw-angle measuring apparatus of the present invention includes an X-ray imaging apparatus 10 and an auxiliary apparatus for X-ray imaging (hereinafter, referred to as auxiliary apparatus) 20 having a grid. The assist device 20 may be used to reduce or eliminate scattered radiation in the radiation transmitted by the x-ray source. Examples of the auxiliary device 20 are shown in FIGS. 3A and 3B. FIG. 3A is a perspective view of an x-ray imaging assistant used in the present invention, and FIG. 3B is a view illustrating an embodiment of a grid.

A cob angle measuring device 100 for measuring a cob angle from a spinal X-ray image of a patient photographed by the X-ray imaging device 10 and the assistant device 20 is constructed. Here, the cob angle measuring device (00) measures the cob angle according to a series of programs. That is, it is implemented not in general hardware but in software.

The cob angle measuring apparatus 100 is shown in Fig. FIG. 4 is a schematic diagram illustrating the claw-angle measuring apparatus of FIG.

The cob angle measuring apparatus 100 is configured first with an information lead unit 110 for reading a spinal X-ray image of the patient according to entered patient information. And a setting unit 120 for setting the read original image. The setting unit 120 serves to perform a preliminary work on the spinal X-ray image to diagnose scoliosis. That is, the cob angle measuring apparatus 100 is intended to create an image condition for extracting a spinal image from an X-ray image. To this end, the setting unit 120 is configured with a conversion unit 122 for converting an X-ray image into a predetermined size and file format. That is, it is difficult for the cob angle measuring apparatus 100 to read the size original of the X-ray image. Reducing size can also reduce capacity. In addition, X-ray images are usually stored in DICOM (Medical Imaging and Communication in Medicine) format, but can be stored in other image file formats such as TIFF, JPG, and PNG as needed. In addition, the setting unit 120 includes a removal unit 124 for removing noise from a file whose size / file has been converted, and an extracting unit 126 for extracting only information corresponding to a predetermined number of left and right pixels from the center of the converted file. According to the setting unit 120, an optimal image file for scoliosis diagnosis is provided. The optimized file is referred to as a 'first image'.

The cob angle measuring apparatus 100 includes a vertebral outline extracting unit 130 for extracting a vertebral outline from the first image and a claw angle measuring unit 140 for measuring a claw angle. The vertebral outline extractor 130 includes a canny edge extraction algorithm 132, a contour detection algorithm 134 and a hough transform algorithm 136. The canny outline extraction algorithm 132 detects an outline line of the spine in the first image and the outline detection algorithm 134 connects the detected outline line to detect a spine outline image. And the Hough transform algorithm 136 detects the straight line of the spine. The image detected up to the straight line of the vertebra will be referred to as a 'second image'.

The cob angle measuring unit 140 measures the cob angle from the second image.

Next, the operation of the above-described claw-angle measuring apparatus will be described. This will be described with reference to FIG. 5, which is a flowchart for explaining the method of measuring the claw angle, and FIG. 6, which is an exemplary image generated by the claw angle measuring method.

In order to obtain a diagnosis of scoliosis, a patient must be photographed with his / her spine using an x-ray imaging device 10 and an assistant device 20 (s100). The X-ray image thus taken is shown in FIG. 6A. The X-ray image is stored as a DICOM file, which is a medical image file.

Patient information for receiving diagnosis of scoliosis is input to the cob angle measuring device 100 (s102).

Then, the information lead unit 110 reads the previously stored X-ray image of the corresponding patient (s104).

Then, the setting unit 120 sets the X-ray image to the first image, which is optimal image information for diagnosing scoliosis (S110). In the setting process, the converting unit 122 converts the size and the file format of the X-ray image into a file format optimized for diagnosis (s112), and the removal unit 124 removes the noise included in the X- (S114), and the extracting unit 126 extracts an image of a certain pixel (for example, 60 piexl) in the left-right direction from the center of the noise-removed image to extract an image of the vertebral region (s116). Here, the extraction unit 126 may extract an image of the spinal region to be extracted by specifying the region to be 60 pixels or more or less than 60 pixels without specifying the image to be 60 pixels, that is, the X-ray imaging apparatus or the setting unit 120 ), And the like. The extracted setting image, that is, the first image is shown in FIG. 6B.

As shown in FIG. 6B, when the first image is provided, the vertebral outline extracting unit 130 provides an image for measuring the claw angle. If the first image shown in FIG. 6B is provided, the canine contour extraction algorithm 132 starts by extracting the contour lines of the vertebrae (s118). Then, the contour detection algorithm 134 is used to detect connected regions (hereinafter referred to as " connection regions ") of the contour lines of the vertebrae (s119).

When the connection area is detected, the contour detection algorithm 134 recognizes the spinal region through the following process. First, the width of each of the detected connection areas is obtained (s122), and an average between the widths of the connection areas is obtained. At this time, the part having a width larger than or smaller than twice the standard deviation based on the obtained average is excluded. In this case, it is determined whether the shape of the connection regions detected in that state is a square shape having a general shape of a vertebra and both corners are flat (s124). In the embodiment, it is judged to be in the form of a hexagon. Then, it is determined whether the detected connection regions have a predetermined interval and are connected to each other in step S126. As a result of the determination, when it is determined that the stepped bridge has a predetermined interval, the spinal region is regarded as normally recognized (s128). However, if the connection region includes more than half of the smallest spinal region, the spinal region is considered to be unrecognized. In this case, the process of loosening and detecting the width and shape of the initially detected connection area is repeated.

An example of the image in which the vertebral region is normally recognized is shown in FIG. 6C. 6C shows an example in which two straight lines are detected by using the Hough transform algorithm 136. In this case, That is, when the Huff transformation algorithm 136 is practically performed, a straight line is detected for all spinal plates of the vertebrae (s130).

Thereafter, the cobb angle measuring unit 140 measures the angle between the straight lines of all the vertebra plates (s142). Then, the cob angle measuring unit 140 determines the angle of the two or more angles having the largest angle as the cob angle (s144). 6D shows an example in which the claw angle measuring unit 140 measures the final claw angle between the spinal plates.

On the other hand, the claw angle measuring unit 140 automatically stores the measured claw angle and the measurement date in the patient information.

As described above, the present embodiment automatically measures the claw angle from a X-ray image using a series of algorithms. The diagnosis of scoliosis is convenient, accuracy is improved, and patient's disease can be managed continuously.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It will be apparent that modifications, variations and equivalents of other embodiments are possible. Therefore, the true scope of the present invention should be determined by the technical idea of the appended claims.

10: X-ray imaging device 20: X-ray assist device
100: Cob angle measuring device 110: Information lead part
120: setting unit 122:
124: Removal section 126:
130: vertebral outline extracting unit 140:

Claims (10)

A setting unit for setting an X-ray image of a patient photographed using a grid-equipped X-ray imaging assistant apparatus as an image necessary for diagnosis of scoliosis;
A spine contour extracting unit for extracting a contour of the spine in the set image so as to detect a straight line of the spine plate; And
And a cob angle measuring unit for measuring an angle between all the straight lines of the spinal column and measuring an angle having the largest angle among the straight lines,
The vertebral outline extracting unit extracts,
A canny edge extraction algorithm for detecting an outline of the vertebrae in an image set by the setting unit;
A contour detection algorithm for detecting a spine contour image by connecting the detected contour lines; And
And a hough transform algorithm for detecting a straight line of the vertebra,
Wherein the contour detection algorithm uses the width and shape information of the connection area connecting the outline lines and recognizes the spine area when the detected connection areas are in the stepping leg shape at regular intervals. Measuring device. The method according to claim 1,
The setting unit may include:
A conversion unit for converting a size and a file format of the X-ray image;
A removal unit for removing noise of the converted image; And
And an extracting unit for extracting only information of a predetermined number of pixels from the center of the noise-removed image to the left and right. 3. The method of claim 2,
The pixel information extracted by the extracting unit may be set to 60 pixels (left and right)
And changes according to a specification of the setting unit or the X-ray photographing apparatus. delete delete A setting step of setting an X-ray image of the patient to which the cob angle measuring device is input according to the environment of the cob angle measuring device;
Extracting an outline of the vertebrae from the set image; Detecting a connection area connecting the detected outline line;
Recognizing a vertebral region using the connection region;
Detecting straight lines of the vertebrae included in the recognized vertebral region;
Measuring the angle of the straight lines; And
Determining a largest angle among the measured angles as a cob angle,
The step of recognizing the vertebral region comprises:
A first calculation step of obtaining the width of each of the connection areas;
A second calculating step of obtaining a width average of the connecting areas;
A first determination step of determining the shape of the connection areas; And
And a second judging step of judging a vertebral region to be normal when the shape of the vertebrae is a spinal shape and a staggered leg shape in which the connecting regions maintain a constant gap. The method according to claim 6,
The outer line extraction uses a canny edge extraction algorithm,
The connection region detection uses a contour detection algorithm,
Wherein the straight line detection uses a hough transform algorithm. delete The method according to claim 6,
Wherein the connecting region is excluded from the recognition step of the vertebral region when the width average is greater than or less than twice the standard deviation. The method according to claim 6,
Wherein if the difference between the connection regions is greater than half of the smallest spinal region in the second determination step, the spinal region is not recognized.

Images