KR20160133188A

KR20160133188A - Medical image reading apparatus and method for operating medical image reading apparatus

Info

Publication number: KR20160133188A
Application number: KR1020150065902A
Authority: KR
Inventors: 최병관; 이도훈; 오세옥; 김철민
Original assignee: 부산대학교 산학협력단
Priority date: 2015-05-12
Filing date: 2015-05-12
Publication date: 2016-11-22
Also published as: KR101810289B1

Abstract

Disclosed are a medical image reading apparatus and a medical image reading apparatus operating method. According to an embodiment of the present invention, the medical image reading apparatus includes: a determination unit which analyzes an inputted medical image and determines whether or not the image includes; an identification unit which identifies a location of a lesion when the lesion is included in the medical image; and a control unit which outputs code data generated by coding the location as a result of the input.

Description

Technical Field [0001] The present invention relates to a medical image reading apparatus and a medical image reading apparatus,

The present invention relates to a lesion localization technique using a variable triangular model, and more particularly, to a medical image reading apparatus and a medical image reading apparatus for automatically identifying a location of a lesion from a medical image (e.g., CT image) And a device operation method.

Currently, in-hospital medical information systems have large-scale storage of readout data (for example, brain imaging readouts) for medical images of patients, but usually readout data is written in the form of narrative data that is not coded.

In order to actually utilize such read data, a reader (e.g., a doctor in charge) must perform a task of finding code data corresponding to the read data over a complicated step, so that a lot of time, manpower, and cost are required, There is a problem that it is merely stored, but the utilization of the read data is remarkably deteriorated.

In addition, conventionally, a function of displaying and indexing a medical image of a patient selected from a patient list, a function of determining the length of a lesion in a medical image, and a function of outputting a report by formatting the read data coded by a reader Etc. have been developed and commercialized.

However, since there is no commercially available program that can automatically detect an anatomical location from a medical image and code an anatomical location, time and cost are incurred when a reader directly encodes a medical image, and in particular, In the case of medical images, the lesion is usually twisted, so it may be difficult to accurately locate the lesion.

Accordingly, there is a desperate need for an apparatus and a technique capable of automatically detecting a lesion from a medical image of a patient organs and coded the anatomical position of the lesion.

The embodiment of the present invention identifies the position of the lesion from the medical image and automatically outputs the coded data generated through coding when the medical image of the patient's organs is input so that the reader does not perform a complicated coding operation , And to easily obtain the code data related to the anatomical position of the lesion.

In addition, the embodiment of the present invention aims to allow a healthcare person to easily grasp the existence of a lesion and its anatomical position within a patient's organ by using only the code data.

It is also an object of the present invention to more accurately identify and code an anatomical location of a lesion based on a variable triangular model even for a medical image that includes a lesion in comparison with a normal state, do.

In addition, embodiments of the present invention can also be applied to a medical imaging system, such as the size, volume, shape, brightness, number and progress of a lesion identified through symmetrical comparison or numerical comparison of a plurality of triangles overlaid on the medical image based on a variable triangular model And to automatically code the medical information of the patient.

In addition, the embodiment of the present invention generates code data in a form that is relatively compact as compared with the conventional read-form data format and can be interpreted by an information terminal such as a computer, The purpose of this study is to improve the utilization of the results of analysis of medical images of patients in PHR (Personal Health Record System), clinical decision system and Big Data System.

A medical image reading apparatus according to an embodiment of the present invention includes a determination unit for analyzing an input medical image to determine whether a lesion is included, and a determination unit for determining whether a lesion is included in the medical image, And a control unit for outputting code data generated by encoding the position as a result of the input.

According to another aspect of the present invention, there is provided a method of operating a medical image reading apparatus, the method comprising: analyzing an input medical image to determine whether a lesion is included; And outputting the code data generated by coding the position as a result of the input.

According to an embodiment of the present invention, when a medical image of a patient's organs is input, the position of the lesion is identified from the medical image and the coded data generated through coding is automatically output so that the reader can perform a complicated coding operation It is possible to easily obtain the code data on the anatomical position of the lesion without performing it.

Also, according to the embodiment of the present invention, it is possible to allow the healthcare person to easily grasp the existence of the lesion and its anatomical position within the patient's organs by using only the code data.

According to an embodiment of the present invention, an anatomical location of a lesion can be more accurately identified and coded on the basis of a variable triangular model even for a medical image including a lesion and a long-term twisted compared to a normal state.

Also, according to an embodiment of the present invention, the size, volume, shape, brightness, number, and size of a lesion identified through a symmetrical comparison or numerical comparison of a plurality of triangles overlaid on the medical image based on a variable triangular model The medical information such as the progress degree can be automatically coded.

Further, according to the embodiment of the present invention, by generating code data in a form that is relatively compact compared to the conventional read-only data type and can be interpreted by an information terminal such as a computer, It is possible to more easily transmit the result of analyzing the medical image of the patient to the medical device, the personal health record system (PHR), the clinical decision system, and the big data system, thereby increasing the utilization.

1 is a block diagram illustrating an internal configuration of a medical image reading apparatus according to an embodiment of the present invention.
2 is a diagram showing an example of creating a virtual figure (triangle) in the input medical image based on the variable triangular model.
FIG. 3 is a diagram illustrating an example of determining the presence or absence of a lesion through a triangular symmetric comparison based on a variable triangular model. FIG.
FIG. 4 is a diagram illustrating an example of determining the presence or absence of a lesion by comparing the distribution of pixels having a predetermined brightness.
FIG. 5 is a diagram showing an example of identifying a position of a lesion by confirming a separation distance from a feature point forming a triangle.
6 is a diagram showing an example of identifying a tissue in which a lesion has occurred in an organ.
7 is a diagram showing an example of code data generated by coding the position of a lesion.
Fig. 8 is a diagram showing an example of coding medical information about the degree of progress of a lesion identified according to a change in the size of a triangle.
9 is a flowchart illustrating a procedure of a method of operating a medical image reading apparatus according to an embodiment of the present invention.

Hereinafter, an apparatus and method for updating an application program according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to or limited by the embodiments. Like reference symbols in the drawings denote like elements.

1 is a block diagram illustrating an internal configuration of a medical image reading apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the medical image reading apparatus 100 may include a determination unit 110, an identification unit 120, and a control unit 130.

The determination unit 110 analyzes the input medical image to determine whether a lesion is included.

The determination unit 110 may be configured to input medical images of a patient's organs (e.g., brain, vertebrae, etc.) from a medical device such as a CT or MRI or a medical image from a storage device (e.g., USB memory, MicroSD card, If it is read and input, the presence or absence of a lesion can be determined from the input medical image.

Here, a lesion can refer to a human tissue in which an abnormality occurs due to a change in pathological action. Generally, if a lesion occurs in a patient's organs, the lesion may cause the organs to twist to the left or right, and the degree of twisting may increase depending on the progress of the lesion.

Accordingly, the determination unit 110 can symmetrically or numerically compare features of left and right triangles based on the variable triangulation model to determine the degree of deformation of the organ due to the lesion in the medical image to determine whether the medical image includes a lesion .

Specifically, the determination unit 110 assigns a singular point to the medical image, creates a virtual figure that connects the singular point with a specified number, and performs symmetric comparison or numerical comparison between the figures to determine whether or not the lesion is present It can be judged.

Here, a singular point (landmark) can be recognized even when a patient's organs (for example, brain) are twisted due to a lesion, or can be assigned to a main structure in a long term in a position where the presence or absence of a lesion can be confirmed. That is, the determination unit 110 determines any one of a center point of the organs associated with the medical image, an outlier point on the outline constituting the organ, and a transition point where the shade is switched over to a predetermined value, Can be assigned.

For example, referring to FIG. 2 (a), the determination unit 110 determines the center point 4 of the brain, the outline points 1, 3, 5, 8 on the brain contour , 9,10), and transition points (0, 2, 6, 7) that are shifted to a value greater than or equal to a predetermined value in the medical image can be assigned as a singular point.

2 (b), the determination unit 110 overlays sixteen left and right symmetrical triangles connecting the respective singular points (0 to 10) on the medical image, The shape of the organ (brain) can be patterned. At this time, the determination unit 110 can give symbols (a1 to a8, b1 to b8) to each triangle, and in particular, in order to facilitate identification of triangles formed in symmetrical positions, Quot; a1 " and " b1 ", for example.

The determination unit 110 symmetrically compares at least one of the size, the width, the shape, and the angle formed by the two sides of the triangles formed at the left and right symmetrical positions, and when the left and right symmetry is not established, (Abnormal). For example, when the right triangle of the medical image 310 shown in the left side of FIG. 3 is symmetrical to the right triangle, the determination unit 110 determines that the medical image 310 is normal without the lesion, If the right triangle is skewed as in the medical image 320 shown in the right side of FIG. 3, the medical image 320 can be judged to be abnormal including a lesion.

In addition, the determination unit 110 symmetrically compares the distributions of pixels having brightness of, for example, 0 to 255 at both triangles (e.g., a1 and b1) created at symmetrical positions in the medical image, , It can be determined that the lesion is included in the medical image.

The determination unit 110 determines whether the number of pixels having the predetermined brightness (e.g., 60 to 70) within the triangles (e.g., a1 and b1) exceeds a certain range (e.g., 10) , It can be determined that the lesion is included in the medical image. The determination unit 110 compares the brightness values within the triangles (e.g., a1 and b1) by pixels and determines that the lesion is included in the medical image if the number of pixels having different brightnesses exceeds a predetermined range can do.

4, the determination unit 110 compares the brightness within the triangles 420 and 430 included in the medical image 410 by pixels and calculates the brightness of the triangle 420 A peak is generated at the brightness '75' in the histogram (ii) regarding the brightness in the triangle 430, while the peak in the histogram (i) It can be judged that the lesion is included.

In addition, the determination unit 110 may designate the range of the peak value of the brightness in the triangle as, for example, '95 to 105 ', and when it is out of the range of the designated maximum value, it is determined that the lesion is included in the medical image .

In addition, the determination unit 110 may compare the feature of the triangle created in the medical image with the medical image of another patient in a normal state to determine whether the medical image includes a lesion.

When the lesion is included in the medical image, the identification unit 120 identifies the position of the lesion.

For example, when the lesion is included in the medical image, the identification unit 120 can identify the position of the lesion by checking whether the number of pixels having the predetermined brightness in each figure exceeds a certain range.

For example, referring to FIG. 4, the determination unit 110 compares the brightness within the triangles 420 and 430 included in the medical image 410 by pixels to determine whether the number of pixels having different brightness is within a predetermined range (For example, 10), it can be determined that the lesion is included in the medical image 410, and the discrimination unit 120 judges that the lesion is included in the triangle 430 having the number of pixels having brightness 60 to 70 exceeding 10 It is possible to identify that the lesion is included. That is, the identification unit 120 can identify the position of the triangle including the lesion, that is, the lesion, by comparing the number of bright pixels in the triangles using the histogram to see if the difference is within a predetermined range.

In another example, when the lesion is included in the medical image, the identification unit 120 can identify the position of the lesion by checking the separation distance from the minutiae forming the respective figures.

5, when the triangle formed on the medical image is determined to be abnormal including the lesion 510, the discrimination unit 120 detects the lesion 510 from the vertices a, b and c forming the triangle By identifying the distance or distance ratio, it is possible to more precisely identify the location of the lesion 510 within the triangle.

In this manner, the identification unit 120 can easily identify the position of the lesion in the organ (for example, the brain) by using the minutiae and sides of the figure (triangle).

In addition, the identification unit 120 may identify at least one medical information, such as size, volume, shape, brightness, number, progress, and contrast enhancement, for the lesion based on the variable triangular model. For example, the identification unit 120 may check the length of a boundary (perimeter) of a value having a brightness of a predetermined value or more to determine whether the shape is close to a circle, Of the lesion.

The control unit 130 outputs the code data generated by coding the position as a result of the input.

For example, the control unit 130 may output the code data according to the position of the identified lesion. Here, the code data may be designated by an International Classification of Diseases (ICD).

For example, referring to FIG. 7 (i), the controller 130 determines whether the lesion is located at a distance from the central point of the brain, that is, in the left frontal cortex, at the boundary line between the triangle a8 and the triangle a1 , It is possible to output the code data '26972005' corresponding to the left frontal cortex.

7 (b), the control unit 130 determines that the lesion is located in the region where the brightness appears in the triangle 'a3' having the brightness of 0 to 255, , It is possible to output the code data '55228006' corresponding to the side fissure.

In another example, the control unit 130 may generate code data by combining the feature points of a figure (triangle) determined to include lesions and the ratio of the distance from each feature point to the lesion. For example, referring to FIG. 5, the controller 130 calculates a ratio of a minutiae a, b, c of a triangle including the lesion 510 to a lesion 510 to a lesion a: b: c = 2: 1: 2.5 'can be combined to generate the code data' a2 b1 c2.5 '.

As another example, the controller 130 may output code data designated to a part of the overlapping segment when overlapping segments forming a figure are overlapped on a region where the lesion is located. For example, as shown in FIG. 6, when the lesion 610 is relatively large and the lesion 610 extends over a plurality of triangles including the triangle a3, the controller 130 determines the position of the lesion 610, It is possible to output the code data specified in the line segment 602 in which the disease name is registered.

In this way, the controller 130 can easily code the position of the lesion 610 even when the size of the lesion 610 included in the medical image is large.

According to the embodiment, when the disease category is identified by the identification unit 120 from the position of the lesion, the control unit 130 codes the disease category and outputs the code in the code data.

Specifically, the identification unit 120 may identify one or more pathologies that are related to the location of the lesion. That is, the identifying unit 120 can identify at least one disease name assigned to the triangle judged to contain the lesion. The control unit 130 can output the code data according to each disease name.

For example, referring to (ii) of FIG. 7, when the area having brightness of 30 to 50 disappears in the triangle 'a3' and instead, a structure such as a bright band appears, Can be confirmed. The control unit 130 may output the code data '21454007' designated for the lesion 'side fistula bleeding'.

Also, according to the embodiment, the identification unit 120 identifies at least one medical information of the size, volume, shape, brightness, number, and degree of progress of the lesion based on the variable triangular model, and the control unit 130 The identified medical information may be coded and included in the code data and output.

For example, in the medical image judged that the lesion is included, the identification unit 120 identifies the progression of the lesion as 'initial' if the size change of the triangle is below a predetermined level, Can be coded as medical information that diagnoses the degree of " early ".

For example, referring to FIG. 8, the identification unit 120 determines whether the triangle 802 within the medical image 820, which is determined to include a lesion, is a triangle 801 in the medical image 810 when the patient is normal , The control unit 130 identifies the progression degree of the disease category 'side tear bleeding' associated with the lesion or the lesion as 'initial', and when the progression degree 'initial' is smaller than 's1' Quot; 21454007 " designated to the ' side < / RTI > hemorrhage ', and outputs it as '21454007 s1'.

In the medical image judged to include the lesion, the identification unit 120 re-identifies the degree of progress of the lesion as 'severe' when the triangle is confirmed to be destroyed, and the control unit 130 re- My progress can be updated to 'severe' and coded.

For example, referring to FIG. 8, when the triangle 803 is almost extinguished in the medical image 830 in which the lesion is judged to be included in the lesion, the discrimination unit 120 judges whether the lesion or the disease category ' The controller 130 re-identifies the progression of the hemorrhage of the side of the patient to be severe and the control unit 130 codes the progression severity as s4 to generate the code data 21454007 designated for the side- And update it as '21454007 s4'.

As described above, according to the embodiment of the present invention, when a medical image of a patient's organs is input, the position of the lesion is identified from the medical image and the coded data generated through coding is automatically output, It is possible to easily obtain the code data related to the anatomical position of the lesion without performing the coding operation.

2 is a diagram showing an example of creating a virtual figure (triangle) in the input medical image based on the variable triangular model.

Referring to FIG. 2, the medical image reading apparatus judges whether a lesion is included in a medical image of a patient organs and, if so, identifies an anatomical location of the lesion within the organ, Based analysis.

2 (a), the medical image reading apparatus includes a central point 4 of the brain, an outer point 1, 3, 5, 8, 9, 10 on the brain contour, , And transition points (0, 2, 6, 7) that are shifted to a value equal to or greater than a predetermined value in the medical image can be assigned as a singular point.

Referring to FIG. 2B, the medical image reading apparatus creates 16 symmetrical triangles symmetrically connecting the respective singularities (0 through 10) over the medical image, and forms a brain shape Can be patterned.

In the medical image reading apparatus, symbols (a1 to a8, b1 to b8) can be assigned to each triangle. Particularly, in a triangle formed at a symmetrical position, symbols 'a1' and 'b1' .

FIG. 3 is a diagram illustrating an example of determining the presence or absence of a lesion through a triangular symmetric comparison based on a variable triangular model. FIG.

3, the medical image reading apparatus symmetrically compares at least one of the size, the width, the shape, and the angle formed by the two sides of the triangles formed at the positions symmetrical to each other in a bilaterally symmetrical manner, It can be judged that the lesion is included (abnormal).

That is, when the right triangle of the medical image 310 shown in the left side of FIG. 3 is symmetric with the right triangle as shown in the left side of FIG. 3, the medical image reading apparatus determines that the medical image 310 is normal without the lesion, The medical image 320 may be judged to be abnormal including a lesion when the right triangle is skewed as in the medical image 320 shown in FIG.

FIG. 4 is a diagram illustrating an example of determining the presence or absence of a lesion by comparing the distribution of pixels having a predetermined brightness.

FIG. 4 shows a CT image 410 of a brain of a patient having a stroke. Referring to FIG. 4, the medical image reading apparatus checks whether the number of pixels having brightness of 60 to 70 in triangle a1 420 and triangle b1 430 exceeds a certain range (for example, 10) , It can be determined that the lesion is included in the CT image 410.

The medical image reading apparatus compares the brightness in the triangles 420 and 430 included in the medical image 410 on a pixel by pixel basis and compares the brightness in the histogram (i) A peak is generated at the brightness '75' in the histogram (ii) regarding the brightness in the triangle 430, so that it can be determined that the lesion is included in the inside of the triangle 430.

In addition, if a triangle shape is adopted as a standard in the world and used to search for lesions, a standard normal histogram may be prescribed throughout the world.

For example, a medical image reading device should be "no more than 10 pixels with a brightness value of 60 to 70 in the left triangle bra-74 of the brain." Or "the peak value of the brightness value of the same triangle shall be 100 ± 5". And the medical image deviating from these rules can be judged to be abnormal (including lesions). These rules can be set differently depending on age or gender, such as, for example, "the younger the age, the higher the peak value".

Also, the medical image reading device can be used to check the lesion, such as the width and height of the triangle, and the angle produced by the two sides.

FIG. 4 is an example for determining which part of the brain (white matter or gray matter) of the lesion has developed. In the histogram shown in FIG. 4 (i) On the other hand, in the histogram shown in FIG. 4 (ii), the number of pixels having the same brightness value is reduced to 150 or less and the brightness value of 200 or more is greatly increased. Accordingly, the medical image reading apparatus can determine that a lesion has occurred in a portion of the triangle 430 corresponding to the white matter of the brain.

FIG. 5 is a diagram showing an example of identifying a position of a lesion by confirming a separation distance from a feature point forming a triangle.

Referring to FIG. 5, when the medical image is determined to be abnormal including the lesion 510 based on the variable triangular model, the medical image reading apparatus calculates the distance from the vertexes a, b, Or distance, the position of the lesion 510 can be more precisely identified within the triangle.

That is, the medical image reading apparatus can confirm the position of the area apart from each point by dividing the distances in the points a, b and c in the shape of a symmetrical triangle. The medical image reading device can use the points and sides to determine which position of the brain the lesion corresponds to.

6 is a diagram showing an example of identifying a tissue in which a lesion has occurred in an organ.

6, when the lesion 610 is spread over several triangles including the triangle a3 because the lesion 610 is relatively large in size, It is possible to output the code data specified in the line segment 602 in which the category (disease name) is registered.

Here, the concentric circle may indicate the distance from the center 601 of the brain to the lesion 610. 4, when the lesion 610 is relatively large and the lesion 610 extends over several triangles, the medical image reading apparatus determines the distance from the center 601 of the brain to the lesion 610, It is possible to identify whether the lesion 610 is at the center of the brain or at the edge of the brain.

In addition, the medical image reading apparatus can output the code data designated to a part of the overlapping line segment when the line segments forming the figure overlap on the region where the lesion is located. That is, when a line segment 602 of the triangle belongs to the lesion 610, the medical image reading apparatus can recognize which part of the brain is abnormal when the pixel value corresponding to the line segment 602 is abnormal , And the specific segment 602 can be coded. For example, the medical image reading apparatus can output the code data specified in the line segment 602 in which the pathology is registered, among the line segments where the position overlaps the lesion 610.

Further, the medical image reading apparatus can identify the position of the lesion by simultaneously using the distance from the feature point and the position of the lesion in the triangle. When the lesion 610 is generated in the two triangles a3 and a1 as shown in Fig. 6, it can be seen that the lesion 610 is formed in the medical image reading apparatus far from the center 601 of the brain when only the triangle a3 is viewed . Therefore, the medical image reading apparatus can identify the position of the lesion in the medical image using the relationship between each triangle and the feature point.

7 is a diagram showing an example of code data generated by coding the position of a lesion.

Referring to (i) of FIG. 7, in the medical image reading apparatus, when the lesion is identified as being located at a position distant from the central point of the brain, that is, in the left frontal cortex, at the boundary line between the triangle a8 and the triangle a1 , And code data '26972005' corresponding to the left frontal cortex can be output.

7 (ii), the medical image reading apparatus discriminates that the lesion is located in the region where the brightness appears to be 30 to 50 in the triangle 'a3' having brightness of 0 to 255, that is, in the side fissure , It is possible to output the code data '55228006' corresponding to the side fissure. At this time, when the area having the brightness of 30 to 50 disappears in the triangle 'a3' and instead, a structure like a bright band appears, the medical image reading device confirms that the hemorrhage occurs in the side fissure, Quot; 21454007 ", which is designated for the " 1 "

Fig. 8 is a diagram showing an example of coding medical information about the degree of progress of a lesion identified according to a change in the size of a triangle.

8 shows the triangular distribution in the medical images 820 and 830 of the hydrocephalic patient and the triangular distribution in the medical image 810 of the patient in the normal case. Referring to FIG. 8, the medical image reading apparatus can determine the presence or absence of a lesion and progress of hydrocephalus using the size and shape of a triangle.

In the medical image 810, the triangle b1 801 and the triangle a1 symmetrical to the triangle b1 801 can be discriminated. On the other hand, in the medical image 820 of the initial hydrocephalus patient, the triangle b1 802 has a small size (width) And the triangle b1 (803) is annihilated in the medical image 830 of the patient with severe hydrocephalus.

A triangle a4 and a triangle b4 in the medical image 830 are located in a medical image 810 in a medical image 830 of a patient with severe hydrocephalus and the characteristic point 0 is located outside a line segment between the characteristic point 1 and the characteristic point 5, , And when the ratio of the widths of the triangles a1 and a4 in the medical image 830 is indexed, it can be used for the diagnosis of hydrocephalus and the tracking of the therapeutic effect (Fig. 2 (a), b)).

The medical image reading device can utilize the numerical objective index for the progress of the hydrocephalus according to the size change of the lesion such as increasing or decreasing hydrocephalus. In addition, the medical image reading apparatus may define the ratio of the value of each line segment constituting the triangle a1, or may use the angle of the angle between the minutiae points 0, 1, 5 as an index of hydrocephalus. For example, in the medical image reading apparatus, by using the fact that the angle connecting the characteristic points '0', '1', and '5' in the medical image 830 of the patient with severe hydrocephalus is close to 0, Can also diagnose hydrocephalus, identify progress, and track treatment effects.

Hereinafter, the operation flow of the medical image reading apparatus 100 according to the embodiments of the present invention will be described in detail with reference to FIG.

9 is a flowchart illustrating a procedure of a method of operating a medical image reading apparatus according to an embodiment of the present invention.

The method of operating the medical image reading apparatus according to the present embodiment can be performed by the medical image reading apparatus 100 described above.

Referring to FIG. 9, in step 910, the medical image reading apparatus 100 analyzes the input medical image to determine whether a lesion is included.

The medical image reading apparatus 100 can determine whether the lesion is included in the medical image by symmetric comparison or numerical comparison of the characteristics of left and right triangles based on the variable triangular model with respect to the degree of deformation of the organ due to the lesion in the medical image.

For example, referring to FIG. 2, the medical image reading apparatus 100 may include a central point 4 of a brain, an outer point of a brain contour line 1, 3, 5, 8, 9 , 10), and mutation points (0, 2, 6, 7) that are switched to a value greater than or equal to a predetermined value in the medical image are assigned to a singular point, and 16 symmetrical triangles Can be created by overlaying on the medical image, and the shape of the organ (brain) can be patterned by the combination of the specified number of triangles.

3, the medical image reading apparatus 100 may be configured such that when the right triangle is right-and-left symmetrical as in the medical image 310 shown in the left side of FIG. 3, If it is determined that the medical image 320 is normal and the triangles hatched in the medical image 320 shown in the right side of FIG. 3 are asymmetric, the medical image 320 may be determined to be abnormal including a lesion.

In step 920, the medical image reading apparatus 100 identifies the position of the lesion when the lesion is included in the medical image.

For example, when the lesion is included in the medical image, the medical image reading apparatus 100 can identify the position of the lesion by checking whether the number of pixels having the predetermined brightness in each figure exceeds a certain range.

For example, referring to FIG. 4, the medical image reading apparatus 100 compares the brightness in the triangles 420 and 430 included in the medical image 410 by pixels to calculate the number of pixels having different brightness If the lesion is included in the medical image 410, the lesion is included in the triangle 430 having the number of pixels having brightness 60 to 70 exceeding 10, Can be identified. That is, the medical image reading apparatus 100 can compare the number of bright pixels in the triangles using the histogram to determine whether the difference is detected within a predetermined range, thereby identifying the position of the lesion in the triangle containing the lesion have.

5, when the medical image reading apparatus 100 determines that the triangle formed on the medical image is abnormal including the lesion 510, the lesion 510 is detected from the vertices a, b, and c forming the triangle By identifying the distance or distance ratio, it is possible to more precisely identify the location of the lesion 510 within the triangle.

As described above, the medical image reading apparatus 100 can easily identify the position of the lesion in a long term (e.g., the brain) by using the minutiae and sides of the figure (triangle).

In step 930, the medical image reading apparatus 100 outputs the code data generated by coding the position as a result of the input.

For example, if the medical image reading apparatus 100 is identified as being located at a distance from the central point of the brain, that is, in the left frontal cortex, at the boundary between the triangle 'a8' and the triangle 'a1', the left frontal lobe Quot; 26972005 " corresponding to the cortex (see (i) of Fig. 7).

In addition, when the medical image reading apparatus 100 is identified as a zone in which the brightness is in a range of 30 to 50 in a triangle 'a3' having a brightness of 0 to 255, that is, a lesion is located in a side fissure, Quot; 55228006 " (refer to (ii) of Fig. 7).

At this time, when the area having the brightness of 30 to 50 disappears in the triangle 'a3' and instead, a structure such as a bright band appears, the medical image reading apparatus 100 displays the code data ' Quot; 21454007 ".

That is, the medical image reading apparatus 100 can identify the disease name (disease category) related to the lesion location and output the code data according to each disease name.

Also, the medical image reading apparatus 100 combines the minutiae points a, b, c of the triangle including the lesion and the ratio 'a: b: c: 2: 1: 2.5' , And the code data 'a2 b1 c2.5' (see FIG. 5).

Further, when the lesion is spread over a plurality of triangles, the medical image reading apparatus 100 can output the code data specified in the line segment (see 602 in FIG. 6) in which the pathology is registered, among the line segments where the lesion and the position overlap each other .

Further, according to the embodiment, the medical image reading apparatus 100 codes at least one medical information of the size, volume, shape, brightness, number, and degree of progress of the lesion identified based on the variable triangular model, And outputs it.

For example, referring to FIG. 8, the medical image reading apparatus 100 determines whether a triangle 802 within a medical image 820, which is determined to include a lesion, is a triangle 802 in a medical image 810, The degree of progression of the disease category 'side fissure bleeding' associated with the lesion is identified as 'initial' and is coded as 's1', and when it is designated for 'side fissure bleeding' Code data '21454007' and output it as '21454007 s1'.

In the medical image reading apparatus 100, when the triangle 803 is almost extinguished in the medical image 830 judged to include the lesion, the degree of progress of the disease category 'side fissure bleeding' , 'C4', and '21454007 s4' in the code data '21454007' designated for the 'hemorrhagic fissure'.

The method according to an embodiment of the present invention may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

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. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

100: Medical image reading device
110:
120: Identification unit
130:

Claims

A judgment unit for analyzing the input medical image to judge whether a lesion is included;
An identification unit for identifying a position of the lesion when the lesion is included in the medical image; And
And outputting code data generated by coding the position as a result of the input,
And a medical image reading device.

The method according to claim 1,
When the disease category is identified from the location by the identification unit,
Wherein,
The disease category is coded and included in the code data and output
Medical image reading device.

The method according to claim 1,
Wherein,
Assigning a singularity point in the medical image,
A virtual figure connecting the singular points is created by a specified number,
The presence or absence of the lesion is judged through symmetrical comparison or numerical comparison between the figures
Medical image reading device.

The method of claim 3,
Wherein,
Any one of a center point of the organ associated with the medical image, an outer point on the contour line constituting the organ, and a transition point where the shade is switched over to a predetermined value is assigned to the singular point
Medical image reading device.

The method of claim 3,
When a lesion is included in the medical image,
Wherein,
The distance between the feature points constituting each figure is checked and the position of the lesion is identified
Medical image reading device.

The method of claim 3,
When a lesion is included in the medical image,
Wherein,
It is determined whether or not the number of pixels having the predetermined brightness in each figure exceeds a certain range, and the position of the lesion is identified
Medical image reading device.

The method according to claim 1,
Wherein,
Identifying at least one medical information of size, volume, shape, brightness, number and progression for the lesion based on a variable triangular model,
Wherein,
The identified medical information is coded and included in the code data and output
Medical image reading device.

8. The method of claim 7,
In a medical image judged to contain a lesion, if the size change of the triangle is below a certain level,
The identification unit identifies the progression of the lesion as 'initial'
Wherein the control unit codes the medical information that diagnoses the progress of the " initial "
Medical image reading device.

9. The method of claim 8,
In the medical image judged to include the lesion, if the triangle is confirmed to be extinct,
The identification unit re-identifies the progression of the lesion as 'severe'
Wherein the control unit updates the degree of progress of the medical information to 'severe'
Medical image reading device.

Analyzing the input medical image to determine whether a lesion is included;
Identifying a lesion in the medical image if the lesion is included in the medical image; And
And outputting code data generated by coding the position as a result of the input
The method comprising the steps of:

11. The method of claim 10,
If a disease category is identified from the location,
Wherein the outputting step comprises:
Categorizing the disease category into the code data and outputting
The method comprising the steps of:

11. The method of claim 10,
Wherein the determining step comprises:
Assigning a singularity in the medical image;
Creating a virtual figure connecting the singular points by a specified number; And
Determining whether the lesion is present through symmetric comparison or numerical comparison between the figures;
The method comprising the steps of:

13. The method of claim 12,
Wherein the step of allocating the singular point comprises:
Assigning any one of a center point of the organs associated with the medical image, an outlier point on the outline of the organs, and a transition point where the shade is switched over to a predetermined value, to the outlier
The method comprising the steps of:

13. The method of claim 12,
When a lesion is included in the medical image,
Wherein the identifying comprises:
Identifying a position of the lesion by checking a separation distance from a feature point constituting each figure,
The method comprising the steps of:

13. The method of claim 12,
When a lesion is included in the medical image,
Wherein the identifying comprises:
Identifying the position of the lesion by checking whether the number of pixels having the predetermined brightness in each figure exceeds a certain range,
The method comprising the steps of:

11. The method of claim 10,
If at least one of the size, volume, shape, brightness, number, and degree of progress of the lesion is identified based on the variable triangular model,
Wherein the outputting step comprises:
Categorizing the identified medical care information into the code data and outputting
The method comprising the steps of:

17. The method of claim 16,
In a medical image judged to contain a lesion, if the size change of the triangle is below a certain level,
Wherein the identifying comprises:
Identifying the progression of the lesion as " early "
Lt; / RTI >
Wherein the outputting step comprises:
The step of coding the medical information diagnosing the progress of the 'initial'
The method comprising the steps of:

18. The method of claim 17,
In the medical image judged to include the lesion, if the triangle is confirmed to be extinct,
Wherein the identifying comprises:
Identifying the progression of the lesion as 'severe'
Lt; / RTI >
Wherein the outputting step comprises:
Updating the progress of the medical information to 'severe' and coding
The method comprising the steps of: