KR102053890B1 - Intestinal obstruction detection Method Using a X-Ray monitoring - Google Patents

Abstract

The present invention relates to an intestinal obstruction detection method based on an X-ray image using PCM-based quantization. The intestinal obstruction detection method based on an X-ray image using PCM-based quantization comprises: a region-of-interest extraction step of extracting an end region of a lung and an upper end region of pelvis and setting a region between the end region of the lung and the upper end region of the pelvis as a region of interest; a Hough transform-based large intestine obstruction feature extraction step of applying Hough transform; and a PCM quantization-based large intestine and small intestine obstruction region extraction step of clustering pixel values into pixels having similar brightness by applying a PCM algorithm in an original X-ray image and extracting a small intestine obstruction candidate region by searching for a cluster group including morphological features of an air liquid layer extracted from the clustered clusters. Therefore, intestinal obstruction can be detected.

Description

Intestinal obstruction detection method using a X-ray monitoring using PCM-based quantization

The present invention relates to an X-ray image-based ileus detection method using PCM-based quantization. In particular, a region of interest extraction step of extracting the distal region of the lung and the upper region of the pelvis and setting the region there between as a region of interest, and Hough transform based colon occlusion feature extraction step applying transform, and morphological analysis of air liquid layer extracted from clustered clusters by clustering by applying PCM algorithm to pixels with similar contrast values in original X-ray image X-ray image-based intestinal obstruction detection using PCM-based quantization, which consists of PCM quantization-based colon and small intestine intestinal obstruction area extraction step to search for cluster groups that contain features to detect small intestine obstruction candidate areas. It is about a method.

Intestinal obstruction is a disease in which the contents of the intestine are completely blocked because part of the intestine is completely blocked.

Diagnosis of bowel obstruction includes X-ray imaging, abdominal auscultation, and CT imaging.

While abdominal auscultation is inexpensive, while accurate diagnosis is difficult, accurate diagnosis of bowel obstruction is possible with X-ray and CT scans.

CT imaging is a complex and expensive procedure for imaging, but X-ray imaging is simple and inexpensive.

When X-ray images of patients with intestinal obstruction are observed, an air-fluid level is observed in which the lower part of the intestine contents is horizontally straight.

The air liquid layer is clearly observed in the large intestine occlusion area, but the air liquid layer is indistinguishable in the small intestine obstruction area.

Therefore, when detecting the intestinal obstruction candidate region using the morphological features of the air liquid layer in the X-ray image of the intestinal obstruction patient, Clearly appearing colonic obstruction areas can be detected, but small intestinal obstructive areas are not detected.

Therefore, in this study, we have developed a method for detecting colorectal occlusion area and detecting a small intestine occlusion area by applying PCM algorithm and 8-way contour tracking technique.

references

[1] K. Yoh, et al., “Gallstone ileus: review of 112 patients in the Japanese literature,” The American Journal of Surgery, Vol. 140, No. 3 pp. 437-440. 1980.

[2] Nicolaou S, Kai B, Ho S, Su J, Ahamed K. Imaging of acute small-bowel obstruction. American Journal of Roentgenology. 2005 Oct; 185 (4): 1036-44.

[3] Reddy SR, Cappell MS. A systematic review of the clinical presentation, diagnosis, and treatment of small bowel obstruction. Current gastroenterology reports. 2017 Jun 1; 19 (6): 28.

[4] Lappas JC, Reyes BL, Maglinte DD. Abdominal radiography findings in small-bowel obstruction: relevance to triage for additional diagnostic imaging. American Journal of Roentgenology. 2001 Jan; 176 (1): 167-74.

[5] Maglinte DD, Heitkamp DE, Howard TJ, Kelvin FM, Lappas JC. Current concepts in imaging of small bowel obstruction. Radiologic Clinics. 2003 Mar 1; 41 (2): 263-83.

[6] Paulson EK, Thompson WM. Review of small-bowel obstruction: the diagnosis and when to worry. Radiology. 2015 Apr 23; 275 (2): 332-42.

[7] Jeffrey RB. Small bowel obstruction. Diagnostic imaging: abdomen, 2nd ed. Salt Lake City, Utah: Amirsys. 2010: 44-7.

[8] Y-Z. Hsiao, S.-C. Pei, “Edge detection, color quantization, segmentation, texture removal, and noise reduction of color image using quaternion iterative filtering,” J. of Electronic Imaging, Vol. 23, No. 4, pp.043001, July, 2014.

[9] J. Malik, R. Dahiya, D. Girdhar, G. Sainarayanan, “Finger knuckle print authentication using Canny edge detection method,” International Journal of Signal and Imaging Systems Engineering, Vol. 9, Issue 6, 2016.

[10] R. Maini and H. Aggarwal, “Study and comparison of various image edge detection techniques,” International Journal of Image Processing (IJIP), vol. 3, no. 1, pp. 1-12, 2009.

[11] AM Wagh, SR Todmal, “Eyelids, Eyelashes Detection Algorithm and Hough Transform Method for Noise Removal in Iris Recognition,” International Journal of Computer Applications, Vol. 112, No. 3, pp.0975-8887, February, 2015 .

[12] Gideon Kanji Damaryam, “A Method to Determine End-Points of Straight Lines Detected Using the Hough Transform,” International Journal of Engineering Research and Applications, Vol. 6, Issue 1, pp.67-75, January 2016.

The present invention is to solve the above problems, the region of interest extraction step of extracting the distal region of the lung and the upper region of the pelvis and set the region between them as the region of interest, Huff transform-based colon to apply the Hough transform Occlusion feature extraction step, clustering by applying PCM algorithm to pixels with similar contrast values in original X-ray image, and searching cluster group including morphological features of air liquid layer extracted from clustered clusters This study aims to detect bowel obstruction by configuring PCM quantization-based colon and small bowel obstruction area extraction step to extract candidate obstruction candidate areas.

As a means for achieving the above object,

In the X-ray image-based intestinal obstruction detection method using PCM-based quantization, performed by a computer device, (a) the terminal area and the pelvis of the lung to set the intestinal area where the intestinal obstruction area exists as the region of interest A region of interest extraction step of extracting an upper region of the region and setting a region between the region of interest; (b) a Hough transform-based colon occlusion feature extraction step using Hough transform to take advantage of the morphological feature that the bottom of the air liquid layer appears horizontally in the X-ray image of the intestinal obstruction patient; (c) The small intestine occlusion area is characterized by similar contrast to the large intestine occlusion area, but the pixel values in the original X-ray image are clustered by applying the PCM algorithm. In addition, this study consists of PCM quantization-based colon and small intestine occlusion area extraction, which searches for a small group of small intestine occlusion candidates by searching for cluster groups that contain the morphological characteristics of the air-liquid layer extracted from clustered clusters.

The extracting of the ROI may include: (a1) applying average binarization to take advantage of the fact that the lung is located at the top of the image and has a relatively low contrast in the X-ray image; (a2) analyzing a horizontal histogram of black pixels in the upper portion of the image in the image to which the average binarization is applied; (a3) searching for the first bone on the histogram based on the maximum value in the histogram and setting it as the distal end of the lung in order to take advantage of the fact that the lower part of the lung becomes bright again; (a4) extracting the waist region to take advantage of the fact that the pelvis is located at the lower part of the waist and has a bright contrast in the X-ray image; (a5) extracting the waist and extracting the contour of the body by separating the body from the background and setting the closest region to the waist region based on the center of the image from the extracted contour; (a6) applying an Otus's binarization technique using the characteristic of the histogram having a bimodal shape to find an optimal threshold value in a grayscale X-ray image to separate a background and a body region; (a7) searching pixels from the outside of the image to the center in the image to which the Otus's binarization technique is applied and setting the first pixel to be outlined by the body; (a8) setting the waist where the nearest pixel is searched based on the center of the X-axis of the image from the contour image of the body; (a9) extracting the contour of the image by applying Canny Edge Detection to the X-ray original image to extract the contour of the pelvic region existing at the bottom of the waist; (a10) It is characterized in that it comprises the step of setting the brightest part of the original image from the contours located at the bottom with respect to the waist region from the contour extracted image to the pelvis region.

In addition, the Hough transform-based colon occlusion feature extraction step, converting a straight line passing through the (x, y) coordinate point of each edge extracted in order to detect a straight line into a curve on the (θ, r) coordinate plane using Equation 1 It is characteristic to let.

(Formula 1)

Θ is an angle formed by a straight line passing through the (x, y) coordinate point of each edge and a vertical line, and r is a distance from the origin to the straight line. x is the x intercept and y is the y intercept.

In addition, in the (r, θ) coordinate plane, a straight line of various angles passing through a point of the extracted edges is represented by a single Sin curve, and when the number n of Sin curves passing through a point is higher than a threshold, the point is expressed as an equation. It is characterized by converting to (x, y) coordinate plane using 2.

(Formula 2)

In addition, the PCM quantization-based colon and small intestine occlusion region extraction, the PCM algorithm uses a typicality (not membership) to express the degree to which each point belongs to the cluster, the PCM objective function is the same as Equation 3 to be.

(Formula 3)

값이 0이 되는 경우에 목적 함수가 최소화되는 자명해(trivial solution)을 제거하기 위해 첨가된 항이고, 수식(3)에서 c는 클러스터의 개수를 의미하며, m은 전형도를 나타내는 상수로 1.5로 설정하고,

는 k 번째 데이터 포인트와 i번째 클러스터 중심 사이의 거리를 의미한다.The second term in formula (3) is all

The term is added to remove the trivial solution in which the objective function is minimized when the value is 0. In formula (3), c is the number of clusters, and m is a constant representing the typicality. Set it up,

Denotes the distance between the k th data point and the i th cluster center.

는 k번째 데이터 포인트가 I번째 클러스터에 소속되는 전형도(Typicality)를 나타내며 식(4)와 같이 정의되는 것이 특징이다.In addition,

Denotes a typicality in which the k th data point belongs to the I th cluster and is defined as in Equation (4).

(Formula 4)

는 각 클러스터에 속하는 값들의 무게 중심을 나타내는 값으로 수식 (5)와 같이 계산하는 것이 특징이다.In addition,

Is a value representing a center of gravity of values belonging to each cluster, and is calculated as shown in Equation (5).

(Formula 5)

In addition, the PCM clustering technique can reduce the sensitivity by removing the Sum-to-One condition, and Equation (3) can be expressed as a sum of c values independent of each other as shown in Equation (6).

(Formula 6)

As described above, the present invention provides a region of interest extraction step of extracting the distal region of the lung and the upper region of the pelvis and setting the region therebetween as a region of interest; In the original X-ray image, the pixel values are clustered by using the PCM algorithm, and the cluster group including the morphological features of the air liquid layer extracted from the clustered clusters is searched for the small intestine occlusion candidate region. PCM quantization-based colon and small intestine occlusion area extraction step to extract the effect of detecting bowel obstruction.

1 is an X-ray image of the intestinal, ileus, pulmonary, and pelvic regions of the human body.
2 is a diagram for extracting a region of interest from an image to which binarization is applied;
2A is an X-ray circle image.
2B is an average binarization of the X-ray diagram.
2C is a horizontal histogram plot of black pixels in an image with mean binarization.
3 is a view for extracting the contour of the waist and pelvic region,
3A is an X-ray circle image diagram.
3B is a diagram of extracting contours;
Figure 3c is a view of the pelvic region extracted.
4 is a relation between r and θ on a coordinate plane.
5 is an image diagram converted to a curve on the r and θ coordinate planes.
6 is a diagram of detecting a straight line and an air liquid layer by applying a Hough transform.
6A is an image diagram of detecting a straight line by applying a Hough transform.
Figure 6b is a result of detecting the morphological characteristics of the air liquid layer using the detected straight line.
FIG. 7 is a diagram illustrating a result of extracting a long occlusion feature extracted by applying a Hough transform.
7A is a result image of applying a Hough transform.
7B is an intestinal occlusion feature extraction image by applying a Hough transform.
FIG. 8 is an image diagram of detecting an ileus region after searching for an object belonging to an extracted cluster and removing a background and noise.
8A is an X-ray original image.
8B is an image of the ileus region extraction result image;
9 is an intestinal obstruction extraction result screen.

Hereinafter, with reference to the accompanying drawings and description will be described in detail the operating principle of the preferred embodiment of the present invention. However, the drawings and the following description shown below are for the preferred method among various methods for effectively explaining the features of the present invention, the present invention is not limited only to the drawings and description below.

In addition, in the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. Terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to intentions or customs of users or operators. Therefore, the definition should be made based on the contents throughout the present invention.

In addition, preferred embodiments of the present invention to be carried out below are already provided in each system functional configuration to efficiently describe the technical components constituting the present invention, or system functions commonly provided in the technical field to which the present invention belongs. The configuration will be omitted, and described mainly on the functional configuration to be additionally provided for the present invention.

If those skilled in the art to which the present invention pertains, it will be able to easily understand the function of the components already used in the prior art among the omitted functional configuration not shown below, and also the configuration omitted as described above The relationship between the elements and the components added for the present invention will also be clearly understood.

In addition, the following examples will be used to appropriately modify the terms so that those skilled in the art to clearly understand the technical features of the present invention to effectively understand, but the present invention It is by no means limited.

As a result, the technical spirit of the present invention is determined by the claims, and the following embodiment is a means for efficiently explaining the technical spirit of the present invention to those skilled in the art to which the present invention pertains. It's just

In addition, it is to be understood that all detailed descriptions, including the principles, aspects, and embodiments of the present invention, as well as listing specific embodiments, are intended to include structural and functional equivalents of these matters. In addition, these equivalents should be understood to include not only equivalents now known, but also equivalents to be developed in the future, that is, all devices invented to perform the same function regardless of structure.

Thus, for example, it should be understood that the block diagrams herein represent a conceptual view of example circuitry embodying the principles of the invention. Similarly, all flowcharts, state transitions, pseudocodes, and the like are understood to represent various processes performed by a computer or processor, whether or not the computer or processor is substantially illustrated on a computer readable medium and whether the computer or processor is clearly shown. Should be.

The functionality of the various elements shown in the figures, including functional blocks represented by a processor or similar concept, can be provided by the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functionality may be provided by a single dedicated processor, by a single shared processor or by a plurality of individual processors, some of which may be shared.

In addition, the explicit use of terms presented in terms of processor, control, or similar concept should not be interpreted exclusively as a citation to hardware capable of executing software, and without limitation, ROM for storing digital signal processor (DSP) hardware, software. (ROM), RAM, and non-volatile memory are to be understood to implicitly include. Other hardware for the governor may also be included.

In the claims of this specification, components expressed as means for performing the functions described in the detailed description include all types of software including, for example, a combination of circuit elements or firmware / microcode, etc. that perform the functions. It is intended to include all methods of performing a function which are combined with appropriate circuitry for executing the software to perform the function. The invention, as defined by these claims, is equivalent to what is understood from this specification, as any means capable of providing such functionality, as the functionality provided by the various enumerated means are combined, and in any manner required by the claims. It should be understood that.

The above objects, features, and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, whereby those skilled in the art may easily implement the technical idea of the present invention. There will be. In addition, in describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted.

Before describing the various embodiments of the present invention in detail, it will be appreciated that the application is not limited to the details of construction and arrangement of components described in the following detailed description or illustrated in the drawings. The invention can be implemented and carried out in other embodiments and can be carried out in various ways. In addition, device or element orientation (e.g., "front", "back", "up", "down", "top", "bottom" The expressions and predicates used herein with respect to terms such as "," "left", "right", "lateral", etc. are used merely to simplify the description of the present invention, and related apparatus. Or it will be appreciated that the element does not simply indicate or mean that it should have a particular direction.

1 is an X-ray image of the intestinal, ileus, pulmonary, and pelvic regions of the human body.

2 is a diagram for extracting a region of interest from an image to which binarization is applied;

2A is an X-ray circle image.

2B is an average binarization of the X-ray diagram.

2C is a horizontal histogram plot of black pixels in an image with mean binarization.

3 is a view for extracting the contour of the waist and pelvic region,

3A is an X-ray circle image diagram.

3B is a diagram of extracting contours;

Figure 3c is a view of the pelvic region extracted.

4 is a relation between r and θ on a coordinate plane.

5 is an image diagram converted to a curve on the r and θ coordinate planes.

6 is a diagram of detecting a straight line and an air liquid layer by applying a Hough transform.

6A is an image diagram of detecting a straight line by applying a Hough transform.

Figure 6b is a result of detecting the morphological characteristics of the air liquid layer using the detected straight line.

FIG. 7 is a diagram illustrating a result of extracting a long occlusion feature extracted by applying Hough transform

7A is a result image of applying a Hough transform.

7B is an intestinal occlusion feature extraction image by applying a Hough transform.

FIG. 8 is an image diagram of detecting an ileus region after searching for an object belonging to an extracted cluster and removing a background and noise.

8A is an X-ray original image.

8B is an image of the ileus region extraction result image;

9 is an intestinal obstruction extraction result screen.

First, in the X-ray image of the body, the lung region is located at the top of the image and has a low contrast. The upper region is located at the bottom of the left lung and is characterized by a flat morphology due to the air layer inside the stomach. The pelvic region is located in the lower region of the body and is characterized by a relatively light intensity. The intestinal region is between the lung and pelvic regions. X-ray image including the intestinal obstruction region is shown in FIG.

Hereinafter, the intestinal obstruction detection method will be described.

1. Region Of Interest Extraction

In order to set the intestinal region where the intestinal obstruction region exists as the region of interest, the distal region of the lung and the upper region of the pelvis are extracted and the region between them is set as the region of interest.

In the X-ray image, average binarization is applied as shown in FIG. 2B to take advantage of the fact that the lung is located at the top of the image and has a relatively low contrast.

In the image to which the average binarization is applied, the horizontal histogram of the black pixels is analyzed in the upper portion of the image as shown in FIG.

In order to take advantage of the fact that the lower part of the lung becomes bright again, the first bone in the histogram is searched based on the maximum value in the histogram and set as the end of the lung.

In FIG. 2C, the red line represents the maximum value on the histogram and the blue line is the area searched for the distal end of the lung.

The pelvis is located at the lower part of the waist and is characterized by bright contrast on X-ray images. In order to use this, a waist region is extracted.

To extract the waist, the body is separated from the background and the body outline is extracted, and the closest point is set as the waist region based on the center of the image.

Otus ’s binarization is applied to separate the background from the body. Otus ’s binarization technique uses the histogram's bimodal shape to find the optimal threshold in grayscale X-ray images.

In the image applied with Otus ’s binarization technique, the pixel that is met for the first time is set as the outline of the body by searching for pixels from the outside of the image to the center. In the image from which the contour of the body is extracted, the waist where the nearest pixel is searched based on the center of the X-axis of the image is set. 3B is an extracted result image of the waist region, and the red line means the waist region. In order to extract the contour of the pelvic area at the lower back, Canny Edge Detection is applied to the X-ray original image to extract the contour of the image. The brightest part of the original image is set as the pelvis region among the contours positioned at the bottom of the waist region from the contour extracted image. 3C is an image of a pelvic region extracted from an image to which Canny Edge Detection is applied, and a blue line is a pelvic region.

2. Huff Transform-based Colon Occlusion Feature Extraction

Hough Transform is a technique for detecting features of an image and is frequently applied in image processing. The Hough transform is characterized by the fact that it is possible to detect figures that can be represented by mathematical expressions. In this study, Hough transform is applied to take advantage of the morphological characteristics of the horizontal flat bottom of the air liquid layer in X-ray images of patients with bowel obstruction. In order to detect straight lines in X-ray image, Hough transform method is applied after edge extraction using Canny Edge Detector. In order to detect the straight line, the straight line passing through the (x, y) coordinate points of the extracted edges is converted into a curve on the (θ, r) coordinate plane using Equation (1).

(Equation 1)

θ is an angle formed by a straight line passing through the (x, y) coordinate point of each edge and a vertical line, and r is a distance from the origin to a straight line. x is the x intercept and y is the y intercept. 4 shows r and θ on the (x, y) coordinate plane.

Using equation (1), a straight line passing through each edge is represented by a curve on the (r, θ) coordinate plane. In this study, the result of limiting the range of θ to 87 to 93 in order to detect a horizontal straight line is shown in FIG. 5.

In the (r, θ) coordinate plane, a straight line of various angles passing through a point of the extracted edges is represented by a single Sin curve. When n Sin curves drawn on the (r, θ) coordinate plane pass through a point, the straight line that appears when the point is converted to the (x, y) coordinate plane passes through n edges. Therefore, when the number n of Sin curves passing through a point is higher than the threshold, the point is converted into the (x, y) coordinate plane using Equation (2). In Equation (2), H is the width of the image and W is the width of the image. FIG. 6 is an image of detecting a straight line using the Hough transform technique, and FIG. 7B is a result of detecting morphological characteristics of the air liquid layer using the detected straight line.

(Formula 2)

3. Extraction of Colon and Small Bowel Occlusion Areas Based on PCM Quantization

The small intestine occlusion area is not clear in morphological features but is characterized by similar intensity to the large intestine occlusion area. These characteristics are analyzed and clustered by applying the PCM algorithm to pixels with similar contrast values in the original X-ray image. The small intestine occlusion candidate region is extracted by searching the cluster group including the morphological characteristics of the air liquid layer extracted from the clustered clusters. The PCM algorithm is a clustering technique that uses typicality to make the number of clusters insensitive. The PCM Clustering technique is as follows. The PCM algorithm uses typicality rather than membership to represent the degree to which each point belongs to a cluster. The objective function of PCM is shown in Equation (3).

(Formula 3)

값이 0이 되는 경우에 목적 함수가 최소화되는 자명해(trivial solution)을 제거하기 위해 첨가된 항이다. 수식(3)에서 c는 클러스터의 개수를 의미한다. m은 전형도를 나타내는 상수로 1.5로 설정하며

는 k 번째 데이터 포인트와 i번째 클러스터 중심 사이의 거리를 의미한다.

는 k번째 데이터 포인트가 I번째 클러스터에 소속되는 전형도(Typicality)를 나타내며 수식(4)와 같이 정의된다.The second term in equation (3) is

The term is added to remove the trivial solution where the objective function is minimized when the value is zero. In Equation (3), c means the number of clusters. m is a constant representing the typicality, set to 1.5

Denotes the distance between the k th data point and the i th cluster center.

Denotes a typicality in which the k th data point belongs to the I th cluster and is defined as shown in Equation (4).

(Formula 4)

는 각 클러스터에 속하는 값들의 무게 중심을 나타내는 값으로 일반적으로 수식(5)와 같이 계산한다.Equation (4) only considers the distance between one data point and one cluster center.

Is a value representing the center of gravity of the values belonging to each cluster, and is generally calculated as shown in Equation (5).

(Formula 5)

The PCM clustering technique can reduce the sensitivity by eliminating the sum-to-one condition. Equation (3) can be expressed as a sum of c values independent of each other, as shown in Equation (6).

(Formula 6)

)이 이전에 학습된 값(

)보다 작으면 재학습을 시도한다.Typical value derived from Equation (4)

) Is the previously learned value (

If less than), try relearning.

The cluster group including the morphological features of the detected colorectal obstruction is searched by applying the Hough transform among clusters of clustered images using the PCM algorithm, and the region belonging to the searched cluster is extracted as the small intestine obstruction as shown in FIG. 7. .

The intestinal occlusion area is detected after searching for an object belonging to the extracted cluster and removing the background and noise. The detected result is shown in FIG. 8.

9 shows a software analysis result screen for automatically detecting the intestinal obstruction of the developed software.

Claims (9)

In the X-ray image-based ileus detection method performed by a computer device, using PCM-based quantization,
(a) a region of interest extraction step of extracting a terminal region of the lung and an upper region of the pelvis and setting a region between the region of interest in order to set the intestinal region in which the intestinal obstruction region exists;
(b) a Hough transform-based colon occlusion feature extraction step using Hough transform to take advantage of the morphological feature that the bottom of the air liquid layer appears horizontally in the X-ray image of the intestinal obstruction patient;
(c) The small intestine occlusion area is characterized by similar contrast to the large intestine occlusion area, but the pixel values in the original X-ray image are clustered by applying the PCM algorithm. Using PCM-based quantization, it consists of PCM quantization-based colon and small intestine occlusion region extraction to search for cluster groups containing morphological features of air liquid layers extracted from clustered clusters. X-ray image based ileus detection method. The method of claim 1,
The ROI extraction step,
(a1) applying average binarization to take advantage of the fact that in the X-ray image the lung is located at the top of the image and has a relatively low contrast;
(a2) analyzing a horizontal histogram of black pixels in the upper portion of the image in the image to which the average binarization is applied;
(a3) searching for the first bone on the histogram based on the maximum value in the histogram and setting it as the distal end of the lung in order to take advantage of the fact that the lower part of the lung becomes bright again;
(a4) extracting the waist region to take advantage of the fact that the pelvis is located at the lower part of the waist and has a bright contrast in the X-ray image;
(a5) extracting the waist and extracting the contour of the body by separating the body from the background and setting the closest region to the waist region based on the center of the image from the extracted contour;
(a6) applying an Otus's binarization technique using the characteristics of the histogram having a bimodal shape to find an optimal threshold in grayscale X-ray images to separate the background and body regions;
(a7) searching pixels from the outside of the image to the center in the image to which the Otus's binarization technique is applied and setting the first pixel to be outlined by the body;
(a8) setting the waist where the nearest pixel is searched based on the center of the X-axis of the image from the contour image of the body;
(a9) extracting the contour of the image by applying Canny Edge Detection to the X-ray original image to extract the contour of the pelvic region existing at the bottom of the waist;
(a10) X-Ray using PCM-based quantization, comprising the step of setting the brightest part of the original image as the pelvis region among the contours positioned at the bottom of the waist region from the contour extracted image Image-based intestinal obstruction detection method. The method of claim 1,
The Hough transform-based colon occlusion feature extraction step,
X-Ray using PCM-based quantization, which converts a straight line passing through the (x, y) coordinate points of the extracted edges into a curve on the (θ, r) coordinate plane using Equation 1 to detect the straight line. Image-based intestinal obstruction detection method.
(Formula 1)

Θ is an angle formed by a straight line passing through the (x, y) coordinate point of each edge and a vertical line, and r is a distance from the origin to the straight line. x is the x intercept and y is the y intercept. The method of claim 3, wherein
In the (r, θ) coordinate plane, a straight line of various angles passing through a point of extracted edges is represented by a single Sin curve, and when the number n of Sin curves passing through a point is higher than a threshold, Equation 2 is expressed as X-ray image-based ileus detection method using PCM-based quantization, characterized in that the conversion to the (x, y) coordinate plane using.
(Formula 2)

The method of claim 1,
The PCM quantization-based colon and small intestine occlusion area extraction,
In the PCM algorithm, X-ray image-based ileus detection using PCM-based quantization is characterized by using typicality rather than membership to represent the degree to which each point belongs to a cluster. Way.
(Formula 3)

The second term in formula (3) is all

The term is added to remove the trivial solution in which the objective function is minimized when the value is 0. In equation (3), c is the number of clusters, and m is a constant representing the typicality. Set it up,

Denotes the distance between the k th data point and the i th cluster center. The method of claim 5,

X-Ray image-based ileus detection method using PCM-based quantization, characterized in that the k-th data point belongs to the I-th cluster (Typicality) and is defined as in Equation (4).
(Formula 4)

The method of claim 5,

X-ray image-based ileus detection method using a PCM-based quantization, characterized in that calculated as shown in equation (5) as a value representing the center of gravity of the values belonging to each cluster.
(Formula 5)

The method of claim 5,
The PCM clustering technique can reduce the sensitivity by removing the Sum-to-One condition, and Equation (3) can be expressed as a sum of c independent values, as shown in Equation (6). X-ray image-based bowel obstruction detection method.
(Formula 6)

A computer-readable recording medium having recorded thereon a program for performing the method according to any one of claims 1 to 8.

Images