KR20210050830A - Apparatus and method for improving area segmentation performance in medical image data - Google Patents

Abstract

Disclosed are a device and method for improving region segmentation performance of medical image data. The method for improving region segmentation performance of medical image data according to an embodiment of the present application comprises: a step of performing region segmentation on a region of interest for each of the plurality of input 2D medical image data; a step of determining a candidate region of a boundary based on a change in pixel brightness at each point of the boundary for the region of interest obtained as a result of the region segmentation; a step of constructing a 3D model of the region of interest from the 2D medical image data based on the result of the region segmentation; a step of generating a plurality of cross-sections in a predetermined angular direction to include at least a portion of the candidate region from the 3D model; a step of determining, as a selected cross-section, a cross-section having improved segmentation performance for the region of interest based on the change in pixel brightness associated with the candidate region in each of the cross-sections; and a step of correcting the 3D model based on the selected cross-section. Accordingly, it is possible to improve the accuracy of boundary surface extraction during 3D modeling of the region of interest.

Description

Apparatus and method for improving area segmentation performance of medical image data {APPARATUS AND METHOD FOR IMPROVING AREA SEGMENTATION PERFORMANCE IN MEDICAL IMAGE DATA}

The present application relates to an apparatus and method for improving region division performance of medical image data.

Medical image data such as CT (Computed Tomography) is used for observing the patient's condition and lesion detection and changes, and contains a lot of information. By utilizing such CT, an internal cross section of the patient's body can be obtained, which is stored in the form of a 2D image and is used as a crucial tool for diagnosis.

The medical staff analyzes medical image data such as CT to grasp the condition of the subject such as the patient, and recognizes and judges the problem situation through in-depth analysis of each shooting slide. Various methods can be applied to facilitate such an analysis. For example, in the case of extracting the boundary of an organ or lesion using a segmentation technique, it is possible to easily observe items to be considered when diagnosing a subject, and Because it is easy to explain this, the technical field related to domain division is gradually developing.

In particular, CT, which is a tomography image of a subject such as a patient, is made in the form of Dicom standard, which is an international standard and has the advantage that it is easy to change to a still image format that can be checked on a computer.

In addition, in recent years, according to the development of anticancer treatment, the need for 3D stereoscopic information is gradually increasing beyond the existing 2D image data in order to increase the precision when establishing an operation plan or when evaluating anticancer drug reactivity. In particular, if a region segmentation method is applied using a plurality of 2D medical image data, 3D modeling of human organs, tissues, etc. can be performed. Since the resolution of the image data is lowered in the process of reconstructing the cross-sectional image of the interval, the 3D modeling result shows an unclear boundary surface in a specific area (for example, the edge of the liver, etc.) according to the characteristics of human organs and tissues. There is a limit that there are many.

The technology behind the present application is disclosed in Korean Patent Publication No. 10-1004342.

The present application is intended to solve the problems of the prior art described above, and uses a cross-sectional reconstruction through analysis in consideration of the anatomical characteristics of the region of interest such as human organs, organs, tissues, etc. from 2D medical image data such as CT. An object of the present invention is to provide an apparatus and method for improving region division performance of medical image data that can improve the accuracy of boundary surface extraction during 3D modeling.

The present application is to solve the problems of the prior art described above, and when three-dimensional modeling of a region of interest such as human organs, organs, tissues, etc., the boundary surface of the region where the boundary surface is difficult to extract due to the large curvature or sharp end surface in the region of interest It aims to extract accurately.

However, the technical problem to be achieved by the embodiments of the present application is not limited to the technical problems as described above, and other technical problems may exist.

As a technical means for achieving the above-described technical problem, the method for improving region division performance of medical image data according to an embodiment of the present application includes performing region division for an ROI for each of a plurality of input 2D medical image data. Determining a candidate region of the boundary based on a change in pixel brightness at each point of the boundary of the region of interest obtained as a result of the region division, from the plurality of 2D medical image data based on the region division result Constructing a 3D model for a region of interest, generating a plurality of cross-sections in a predetermined angular direction to include at least a portion of the candidate region from the 3D model, and associated with the candidate region in each of the plurality of cross-sections It may include determining a cross section with improved segmentation performance for the region of interest based on a change in pixel brightness as a selected cross section, and correcting the 3D model based on the selected cross section.

In the determining of the candidate region, a point of each point of the boundary that does not satisfy a condition for changing the brightness of a pixel in a vertical direction with respect to the ROI may be determined as the candidate region.

In addition, the pixel brightness change condition includes at least a part of a region in which the differential value of the pixel brightness change amount exceeds a preset upper limit value in a region close to a corresponding point by a predetermined level or more, and a preset lower limit value outside the region close to the predetermined level or more. If the following is maintained, it may be satisfied.

In addition, the step of constructing the 3D model may include deriving point cloud data for a region of interest in a 3D space based on the plurality of 2D medical image data.

In addition, the step of generating the plurality of cross-sections in the predetermined angular direction may generate a plurality of cross-sections in all angular directions for the candidate region.

In addition, the step of generating the plurality of cross sections in the predetermined angular direction may include a predetermined probability that the segmentation performance for the region of interest is improved based on the sampled 3D data for the candidate region is determined to be greater than or equal to a preset threshold probability. It is possible to generate the plurality of cross-sections by selecting the angular direction of.

In addition, the region of interest may include a predetermined intraperitoneal organ.

In addition, in the step of generating the plurality of cross-sections in the predetermined angular direction, the plurality of cross-sections may be generated by selecting a predetermined angular direction in consideration of an anatomical characteristic of an organ corresponding to the region of interest.

In addition, the step of determining the candidate region may be performed in consideration of anatomical characteristics of an organ corresponding to the region of interest.

In addition, in the determining of the candidate region, the candidate region may be determined at a point in which a curvature exceeds a predetermined threshold curvature among boundaries of the ROI.

In addition, the 2D medical image data may include CT image data.

Meanwhile, the apparatus for improving region segmentation performance of medical image data according to an exemplary embodiment of the present application includes a region segmentation unit that performs region segmentation for an ROI for each of a plurality of input 2D medical image data, and is obtained as a result of segment segmentation. A candidate region determining unit for determining a candidate region among the boundaries based on a change in pixel brightness at each point of the boundary with respect to the region of interest, and 3D for the region of interest from the plurality of 2D medical image data based on the region division result A three-dimensional conversion unit for constructing a model and a plurality of cross-sections in a predetermined angular direction to include at least a portion of the candidate area from the 3D model, and change in pixel brightness associated with the candidate area in each of the plurality of cross-sections A cross-section selection unit configured to determine a cross-section with improved division performance for the region of interest based on the selected cross-section.

In addition, the 3D conversion unit may correct the 3D model based on the selected cross section.

In addition, the cross-section selector may generate a plurality of cross-sections in all angular directions with respect to the candidate area, or a probability that the segmentation performance for the ROI is improved based on the sampled 3D data for the candidate area is preset. The plurality of cross-sections may be generated by selecting a predetermined angular direction determined to be greater than or equal to a critical probability, or the plurality of cross-sections may be generated by selecting a predetermined angular direction in consideration of an anatomical characteristic of the region of interest.

The above-described problem solving means are merely exemplary and should not be construed as limiting the present application. In addition to the above-described exemplary embodiments, additional embodiments may exist in the drawings and detailed description of the invention.

According to the above-described problem solving means of the present application, the 3D modeling of the region of interest by utilizing the cross-sectional reconstruction through analysis considering the anatomical characteristics of the region of interest such as human organs, organs, tissues, etc. from 2D medical image data such as CT An apparatus and method for improving region segmentation performance of medical image data capable of improving the accuracy of boundary surface extraction may be provided.

According to the above-described problem solving means of the present application, when three-dimensional modeling of a region of interest such as human organs, organs, tissues, etc., it is possible to more accurately extract the boundary surface of the region where the boundary surface is difficult to extract because the region of interest has a large or sharp end surface have.

According to the above-described problem solving means of the present application, as the boundary of the 3D model for the region of interest such as human organs, organs, tissues, etc. becomes clear, the accuracy of the volume calculation associated with the region of interest may be improved.

However, the effect obtainable in the present application is not limited to the above-described effects, and other effects may exist.

1 is a schematic configuration diagram of a medical image data analysis system including an apparatus for improving region division performance of medical image data according to an exemplary embodiment of the present disclosure.
2 is a diagram for explaining a change in brightness of a pixel in a vertical direction at each point of a boundary of an ROI.
3 is a graph schematically illustrating a change in pixel brightness at a point satisfying a pixel brightness change condition according to an exemplary embodiment of the present disclosure.
4 is a diagram for explaining angular directions for generating a plurality of cross sections in order to generate or correct a 3D model with improved region division performance.
5 to 7 are views exemplarily illustrating a plurality of cross sections generated by the apparatus for improving area division performance of medical image data according to an exemplary embodiment of the present disclosure based on various angular directions.
FIG. 8 is a diagram for explaining determining as a selected cross section a cross section having improved segmentation performance for an ROI for a determined candidate region.
9 is a schematic configuration diagram of an apparatus for improving region division performance of medical image data according to an exemplary embodiment of the present disclosure.
10 is a flowchart illustrating a method of improving segmentation performance of medical image data according to an exemplary embodiment of the present disclosure.

Hereinafter, embodiments of the present application will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art may easily implement the present application. However, the present application may be implemented in various different forms and is not limited to the embodiments described herein. In addition, in the drawings, parts irrelevant to the description are omitted in order to clearly describe the present application, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the present specification, when a part is said to be "connected" with another part, it is not only the case that it is "directly connected", but also "electrically connected" or "indirectly connected" with another element interposed therebetween. "Including the case.

Throughout the present specification, when a member is positioned "on", "upper", "upper", "under", "lower", and "lower" of another member, this means that a member is located on another member. This includes not only the case where they are in contact but also the case where another member exists between the two members.

In the entire specification of the present application, when a certain part "includes" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary.

The present application relates to an apparatus and method for improving region division performance of medical image data. In particular, the present application is an apparatus and method for improving the performance of segmentation for a specific area with respect to medical image data. An apparatus and method for improving the accuracy of boundary surface extraction by generating a cross section based on ).

For reference, in the description of the embodiments of the present application, terms related to a direction or a position (left direction, right direction, vertical direction, etc.) are described based on the arrangement state of each component shown in the drawings. For example, when viewed in FIG. 1, the left direction may be a 9 o'clock direction, the right direction may be a 3 o'clock direction, an upper direction may be a 12 o'clock direction, and a lower direction may be a 6 o'clock direction.

However, such a direction setting may vary depending on a direction in which medical image data input to the present apparatus is captured, a direction in which the medical image data is viewed, and the like. For example, if necessary, medical image data may be acquired so that the upward direction of FIG. 1 faces a horizontal direction (left and right direction), and as another example, the medical image data may be rotated so that the upward direction of FIG. 1 faces an oblique oblique direction. I can.

1 is a schematic configuration diagram of a medical image data analysis system including an apparatus for improving region division performance of medical image data according to an exemplary embodiment of the present disclosure.

Referring to FIG. 1, the medical image data analysis system 10 of the present application may include an apparatus 100 for improving region division performance of medical image data according to an exemplary embodiment of the present application. In addition, referring to FIG. 1, the apparatus 100 for improving region division performance of medical image data according to an exemplary embodiment of the present disclosure receives, as an input, a plurality of 2D medical image data 1 captured as a 2D image. Thus, it may be implemented to provide a 3D model 2 derived from a plurality of 2D medical image data. According to an exemplary embodiment of the present disclosure, the 2D medical image data 1 may include CT image data.

In addition, (a) of FIG. 1 is an embodiment of a 3D model (2) provided by the apparatus 100 for improving region division performance of medical image data according to an embodiment of the present application, and (b) of FIG. In the implemented 3D model (2), the boundary surface of the area where it is difficult to clearly distinguish the boundary surface due to reasons such as having a sharp or sharp edge or having difficulty in distinguishing it from other organs adjacent to other organs has been corrected according to the present application. This is an enlarged view of the dimensional mesh generation result.

In addition, referring to FIG. 1, the apparatus 100 for improving region division performance of medical image data according to an embodiment of the present application sets the liver, which is an organ in the abdominal cavity, as a region of interest, and after segmenting the region for the liver region, 3D modeling It is possible to create a volume (3D object) according to. At this time, depending on the anatomical shape and location characteristics of the liver, there are parts that can be divided well and parts that cannot be divided, and it is difficult to obtain a 3D model that satisfies the division of areas in all areas by setting only one condition. It may be more pronounced in narrow or dense areas (eg, areas at the edge of the liver).

In addition, according to an embodiment of the present application, the apparatus 100 for improving region segmentation performance of medical image data (hereinafter referred to as “region segmentation performance enhancement apparatus 100”) according to an exemplary embodiment of the present disclosure captures a medical image. A plurality of 2D medical image data 1 may be received from an apparatus (not shown). In addition, the 3D model 2 generated by the region division performance improving apparatus 100 may be transmitted to and stored in a user terminal (not shown), or may be implemented to be displayed by a display module of the user terminal (not shown). In addition, according to an exemplary embodiment of the present disclosure, the apparatus 100 for improving region division performance may be provided as a device included under a medical image capturing apparatus (not shown). As another example, the region division performance improving apparatus 100 may be provided as a device included under a user terminal (not shown).

According to an embodiment of the present application, the medical imaging apparatus (not shown) may include a medical computed tomography (CT) scanner, an X-ray imaging device, a magnetic resonance imaging (MRI) device, etc. It is not limited only. That is, the medical imaging apparatus of the present application and medical image data to be analyzed are not limited to a specific type, and include a wide range of conventional medical field imaging devices and imaging devices to be developed in the future for imaging the inside of the body of a subject such as a patient. It is desirable to be understood as a concept.

For example, a user terminal (not shown) includes a smartphone, a smart pad, a tablet PC, etc. and a personal communication system (PCS), a global system for mobile communication (GSM), a personal digital cellular (PDC), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) It can include all kinds of wired/wireless communication devices such as terminals.

According to an exemplary embodiment of the present disclosure, the above-described medical imaging apparatus, region division performance improvement apparatus 100, and a user terminal may communicate with each other through a network (not shown).

Here, the network (not shown) is exemplified by way of example 3GPP (3rd Generation Partnership Project) network, LTE (Long Term Evolution) network, 5G network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), Bluetooth network, Wi-fi network, satellite broadcasting network, analog broadcasting network , Digital Multimedia Broadcasting (DMB) networks, and the like are included, but are not limited thereto.

The apparatus 100 for improving region segmentation performance may perform region segmentation on an ROI for each of the plurality of input 2D medical image data 1. Here, performing region division may be understood as extracting a boundary (contour) for the region of interest.

According to an embodiment of the present application, the apparatus 100 for improving region segmentation performance uses a semantic segmentation technique, or corresponds to the intensity or intensity distribution of each region of the 2D medical image data 1, and the region of interest. In consideration of morphological information such as organs, organs, etc., the primary region can be divided for the region of interest.

According to an embodiment of the present disclosure, the region of interest may include a predetermined intraperitoneal organ. For example, the region of interest may include the liver, kidneys, spleen, lungs, stomach, and the like. However, the region of interest in the present application is not limited to a specific organ, and may widely include various elements present in the subject's body that can be identified from medical image data such as blood vessels, tissues, tumor lesions, bones, and pelvis. In particular, the present application not only extracts the external boundary surface of organs (eg, heart, kidney, spleen, etc.) having a smooth curve (curved surface), but also externally of organs (eg, liver, etc.) having a large or sharp end surface. The interface also has an effect that can be easily extracted.

The apparatus 100 for improving region division performance may determine a candidate region among the boundaries of the region of interest based on a change in pixel brightness at each point of the boundary of the region of interest obtained as a result of region division. Here, the candidate region may be determined as a region having an unclear or ambiguous boundary among boundaries of the ROI obtained as a result of the preceding region division. In other words, when the 3D model 2 is constructed from a plurality of 2D medical image data 1, the candidate region may mean a region in which the boundary surface of the ROI is likely to appear inaccurately. That is, the candidate region may mean a region in which the boundary surface is to be newly analyzed.

Hereinafter, with reference to FIGS. 2 and 3, a process of determining a candidate area by the apparatus 100 for improving divisional performance will be described in detail.

2 is a diagram for explaining a change in brightness of a pixel in a vertical direction at each point of a boundary of an ROI. Referring to FIG. 2, (a) of FIG. 2 shows a graph of a change in pixel brightness at a point where a change in pixel brightness in a vertical direction is relatively clear, and FIG. 2 (b) is a point where a change in pixel brightness in a vertical direction is somewhat unclear. This is a graph of the pixel brightness change at.

Referring to FIG. 2, the apparatus 100 for improving region division performance is based on a change in pixel brightness at each point of a boundary of an ROI (referring to FIG. 2, a boundary line indicated in red) obtained as a result of region division. Can determine the candidate area. In more detail, the apparatus 100 for improving region division performance may determine, as a candidate region, a point of each point of the boundary that does not satisfy a condition for changing the brightness of a pixel in a vertical direction with respect to the region of interest.

Here, in the pixel brightness change condition according to the exemplary embodiment of the present disclosure, there is at least a part of a region in which the differential value of the pixel brightness change amount exceeds a preset upper limit value in a region close to a corresponding point by a predetermined level or more, and is close to the corresponding point by a predetermined level or more. Outside of the region, it may be satisfied if it is maintained below a preset lower limit. As described above, the present application introduces the threshold values (upper limit and lower limit values) among them as a criterion for determining the candidate region, so that the candidate region requiring resetting of the boundary can be more closely detected.

3 is a graph schematically illustrating a change in pixel brightness at a point satisfying a pixel brightness change condition according to an exemplary embodiment of the present disclosure.

Referring to FIG. 3, FIG. 3(a) is a graph showing the amount of change in pixel brightness based on a predetermined point selected among the boundaries, and FIG. 3(b) is a derivative (for example) of the amount of change in pixel brightness in (a). For example, this is a graph showing a differential value of a pixel brightness change amount obtained by first-order differentiation based on a selected predetermined point. The predetermined points selected in FIGS. 3A and 3B are indicated by dotted lines based on the horizontal axis.

Referring to (b) of FIG. 3, in the case of a point that satisfies the pixel brightness change condition of the present application shown in FIG. 3, an area exceeding a preset upper threshold exists in an area close to the point by a predetermined level or more. It can be seen that all of the conditions and conditions for maintaining a preset lower threshold or less are satisfied outside of the condition and a region close to the corresponding point by a predetermined level or more. If the pixel brightness change condition is satisfied as described above, it can be determined that the region segmentation has been performed relatively accurately so that the boundary of the region of interest is clearly distinguished at the corresponding point. In this case, the corresponding point that satisfies the pixel brightness change condition is a candidate It may not be determined (not selected) as a region.

Conversely, as in the case of (b) of FIG. 2, when a change in pixel brightness does not appear clearly based on a corresponding point (in other words, when the pixel brightness change condition of the present application is not satisfied), the corresponding point is a candidate region. It is determined (selected), and thereafter, a process for clarifying the boundary for the corresponding candidate region proceeds.

According to an embodiment of the present application, the value of the range (or length, width, etc.) of the proximity region at a corresponding point, which is a criterion for whether a predetermined upper limit value, a preset lower limit value, and an upper limit value is met, and whether the lower limit value is met, for determining a candidate region is of interest. It may be determined based on at least one of a region type, a resolution of the medical image data, an average pixel value of the 2D medical image data 1, a median value of the pixel value, and a range of a pixel value of the 2D medical image data 1. In addition, as another example, the value of the range (or length, width, etc.) of the proximity area at the corresponding point, which is the criterion for whether the predetermined upper limit value, the preset lower limit value, and the upper limit value is met, and whether the lower limit value is met, for determining the candidate area, is determined by user input. It can be implemented to be adjusted. In this case, the user is required by variously adjusting at least one of the range of the proximity region at the corresponding point, which is a criterion for satisfying the predetermined upper limit value, the preset lower limit value, and whether the lower limit value is met or not, through user input. It is possible to determine at what level the candidate region is to be determined according to, and furthermore, by changing the specifications including the accuracy and resolution of the 3D model 2 to be obtained through the region division performance improvement device 100, It can be made to meet the requirements of the 3D model (2).

In addition, according to an exemplary embodiment of the present disclosure, the apparatus 100 for improving region division performance may be designed to determine a candidate region at a point in which a curvature exceeds a preset critical curvature among boundaries of a region of interest. Here, among the boundaries of the region of interest, a point whose curvature is less than a preset critical curvature can be understood as a region in which boundary detection is relatively well performed because the shape of the region of interest adjacent to the point is relatively smooth. The point of the boundary for which the curvature exceeds the preset critical curvature may be understood as a region having a high probability of incorrectly performing boundary detection, such as the shape of the region of interest adjacent to the corresponding point having a sharp shape or a sharp edge. As described above, the apparatus 100 for improving the area segmentation performance can infer the difficulty of detecting the boundary at the corresponding point based on the curvature information (for example, the radius of curvature, etc.) of each point of the first detected boundary. It may be implemented to preferentially select a candidate region that is likely to require re-detection of the boundary surface based on the detection difficulty.

In addition, according to the exemplary embodiment of the present disclosure, the apparatus 100 for improving region division performance may determine a candidate region in consideration of anatomical characteristics of an organ corresponding to the region of interest. In other words, the apparatus 100 for improving region division performance does not perform a candidate region search for all parts of the region of interest in consideration of the anatomical characteristics of the organ corresponding to the region of interest, but a region with distinct characteristics of the organ (e.g. For example, it can be operated to preferentially determine a candidate region, particularly in a sharp portion, an angled portion, etc.).

Specifically, for example, when the region of interest is'liver', generally, when viewed as a whole, the left region of the liver is gentle, and the right region may include relatively sharp and sharp edges. Accordingly, the difficulty of detecting the boundary in the right area may be higher than in the left area. In consideration of the anatomical characteristics of the region of interest, the apparatus 100 for improving region division performance may operate to preferentially determine a candidate region in the right region than in the left region of the liver. For example, the apparatus 100 for improving area division performance includes a region with a high probability of having a relatively gentle shape based on the general shape of the relevant organ and sharp or sharp edges based on information on which organ the region of interest corresponds to. A preset upper limit value, a preset lower limit value, and whether the upper limit value and the lower limit value are met for determining a candidate region in a region having a high probability of having a may be set differently.

In addition, the apparatus 100 for improving region division performance may construct a 3D model 2 ′ for an ROI from a plurality of 2D medical image data 1 based on a result of region division. Here, the 3D model 2 ′ of the constructed ROI may include an ambiguous or inaccurate boundary surface. In other words, the 3D model 2 ′ generated in this step may refer to a 3D model before correction (border reset) for the previously determined candidate region is performed.

The 3D object of the ROI is converted into a 2D image by CT scan and stored, but if the loss of 2D image information is not significant, it is possible to process it into a 3D object in a 3D space again. At this time, the lost information can be generated by interpolation.

In this regard, according to an embodiment of the present application, the apparatus 100 for improving region division performance is based on a plurality of 2D medical image data 1 in order to construct a 3D model 2'. It may be to derive data for point cloud. In other words, the area segmentation performance improvement apparatus 100 may utilize a method of synthesizing 2D images photographed from various directions and converting them into a point cloud (point cloud) in a 3D space as a technique for building a 3D model. It is not limited to this. In addition, the region division performance improving apparatus 100 may generate a new section 1'using an arbitrary axis or an arbitrary angular direction based on the 3D model 2 or reconstructed 3D data that is generated thereafter. .

The region segmentation performance enhancement apparatus 100 determines all pixel values for the x-axis, y-axis, and z-axis of each portion of an ROI in a 3D space based on a plurality of 2D medical image data 1 at predetermined intervals. Can be assigned (can be filled). The 3D model 2'(in other words, the 3D model before correction) built by the region segmentation performance improvement device 100 is used to obtain a cross section in different angular directions to improve segmentation performance for the candidate region. I can.

The region division performance improvement apparatus 100 may generate a plurality of cross sections 1 ′ in a predetermined angular direction to include at least a part of the candidate region from the constructed 3D model 2 ′.

Hereinafter, a process of generating a plurality of cross sections 1 ′ in a predetermined angular direction by the region division performance improving apparatus 100 will be described in detail with reference to FIGS. 4 to 7.

4 is a diagram for explaining angular directions for generating a plurality of cross sections in order to generate or correct a 3D model with improved region division performance.

Referring to FIG. 4, cross-sections 1 ′ for all arbitrary angular directions θ may be generated based on a predetermined point. In addition, referring to FIG. 4, the angular direction θ may be measured in different ways based on the direction in which the subject's body is viewed ( θ ₁ Or θ ₂ ) .

Referring to Figure 4, Figure 4 position (coordinates) of the reference position for measuring the angular orientation (θ) for convenience relative to the center point of the subject body, but the angular orientation (θ) calculation of the description is determined candidate regions Can be determined as

According to an embodiment of the present application, when a predetermined angular direction θ is determined, an axis for generating a cross section 1'including at least a portion of the candidate region is determined accordingly, and the region division performance is improved. The device 100 may be one that generates a cross section 1 ′ based on a determined arbitrary axis.

5 to 7 are views exemplarily illustrating a plurality of cross sections generated by the apparatus for improving area division performance of medical image data according to an exemplary embodiment of the present disclosure based on various angular directions.

Referring to FIGS. 5 to 7, the apparatus 100 for improving the area division performance includes a plurality of cross-sections 1 ′ in all angular directions with respect to the determined candidate area (e.g., a 360-degree cross-section based on the candidate area). (1')) can be created.

In addition, according to an embodiment of the present application, the apparatus 100 for improving the segmentation performance of the region determines that the probability that the segmentation performance for the region of interest is improved is greater than or equal to a preset threshold probability based on the sampled 3D data for the candidate region. A plurality of cross-sections 1 ′ may be generated by selecting a predetermined angular direction. In the case of the above-described method of generating the cross section 1'for all angular directions, since the cross section 1'can be generated in consideration of data for all angular directions, it has the advantage of obtaining an accurate result. This is because the calculation time (for example, the time required to generate the cross section 1'in all angular directions) can be increased by considering all angular directions. In other words, in order to shorten the operation time, the region division performance apparatus 100 does not consider all angular directions, but uses the sampled 3D data to determine an approximate amount of change (for example, the amount of change in pixel brightness), and Among them, the most probable angular direction (in other words, the most probable axial change) can be selected to generate the cross section 1'.

For example, the region division performance improving apparatus 100 extracts only line data in a predetermined direction for the candidate region as sampled 3D data, and then uses a change in brightness of the sampled pixel value in the corresponding line data. It can be implemented to select the angular direction. Here, the selection of a predetermined direction for line data extraction may be performed only for a partial direction sampled by selecting a value of θ having a preset interval.

Referring to FIGS. 6 and 7, the apparatus 100 for improving area division performance selects angular directions of 90 degrees, 45 degrees, 13 degrees, and 0 degrees instead of all angular directions using sampled 3D data. Create a section (1') (Fig. 6) or select angular directions of 22.5 degrees, 45 degrees, 67.5 degrees, 90 degrees, 112.5 degrees, 135 degrees, and 157.5 degrees instead of all angular directions to create a section (1') (Fig. 7) You can.

In addition, according to an exemplary embodiment of the present application, the apparatus 100 for improving region division performance selects a predetermined angular direction in consideration of the anatomical characteristics of an organ corresponding to the region of interest, thereby forming a plurality of cross sections 1 ′. Can be generated. For example, when the region of interest is'liver', generally, when viewed as a whole, the left region of the liver is gentle, and the right region may include relatively sharp and sharp edges. Accordingly, the difficulty of detecting the boundary in the right area may be higher than in the left area. In consideration of the anatomical characteristics of the region of interest, the apparatus 100 for improving region division performance is based on the position information of the determined candidate region in the region of interest, and the like, in consideration of different inclination characteristics between different angular directions, etc. In the case of generating the cross section 1'by selecting, there is an effect of reducing the operation time as in the case of using the above-described sampled 3D data.

According to an embodiment of the present application, a method of generating a cross section for all the angular directions described above for one candidate region, a method of generating a cross section by selecting a predetermined angular direction based on sampled 3D data, and a region of interest The method of selecting a predetermined angular direction in consideration of the anatomical characteristics of the organ corresponding to is the type of the region of interest, the resolution of the medical image data, the average pixel value of the secured 2D medical image data (1), and the median value of the pixel value. And a pixel value range of the 2D medical image data 1. As another example, selecting any one of the above-described three section generation methods may be performed in a manner selected by the user based on a separate user input applied to the region division performance improving apparatus 100.

The apparatus 100 for improving region division performance may determine a section with improved division performance for the region of interest as the selected section based on a change in pixel brightness associated with the candidate region in each of the generated plurality of sections 1 ′. Here, the section with improved segmentation performance refers to a section (1) that includes data to more clearly distinguish the boundary of the candidate region compared to the 2D medical image data (1) that was used when the previously constructed 3D model (2) was created. If') exists among the generated cross-sections (1'), the boundary surface for the region of interest is clear if the corrected 3D model (2) is created by correcting the previously constructed 3D model (2') based on this. It can be understood as referring to a cross section that can be made.

The apparatus 100 for improving region division performance according to an exemplary embodiment of the present disclosure may determine a cross section 1 ′ in which a point satisfying the above-described pixel brightness change condition among a plurality of cross sections 1 ′ exists as the selected cross section. . In addition, according to an embodiment, when there are a plurality of cross-sections 1'in which points satisfying the above-described pixel brightness change condition exist, the apparatus 100 for improving region division performance is All of the plurality of cross-sections 1 ′ may be determined as the selected cross-section. According to an exemplary embodiment, detailed conditions (eg, an upper limit value, a lower limit value, etc.) of a pixel brightness change condition for determining a candidate region and a pixel brightness change condition for determining a selected cross section may be determined differently from each other.

As another example, the region division performance improving apparatus 100 includes a pixel brightness change value associated with a candidate region in each of a plurality of cross-sections 1'(pixel brightness change at a point corresponding to a candidate region in the cross-section 1'). Value) may be implemented so that a plurality of cross-sections 1'are arranged in a large order, and the upper-order cross-section as many as a preset number of cross-sections is determined as a selection cross-section.

As another example, the region division performance improving apparatus 100 includes a pixel brightness change value associated with a candidate region in each of a plurality of cross-sections 1'(pixel brightness at a point corresponding to a candidate region in the cross-section 1'). The change value) may be implemented to determine all the cross-sections 1 ′ having a predetermined threshold change value or more as the selected cross-section.

After the selected cross section is determined, the region division performance improving apparatus 100 may correct the 3D model 2 ′ constructed based on the selected cross section. Here, the meaning of being able to correct the constructed 3D model (2') means that the existing ambiguous boundary surface is changed to a clear boundary surface (contour line) based on the determined selected cross-section, so that it is closer to the actual region of interest (long term, etc.). It can be understood that it is possible to create a model (2).

If the boundary point for the region of interest in the selected section is located at a point different from the point at the boundary (in other words, the candidate region) determined based on the preceding region division result, the apparatus 100 for improving region division performance May reset the boundary point in the selected section to the boundary point on the 3D model 2. In other words, the region division performance improving apparatus 100 may search for a boundary point closer to an actual boundary of the ROI from the selected cross-section. Resetting the boundary point and setting the coordinates of the reset boundary point in the three-dimensional space may be performed by the following Equations 1-1 and 1-2.

[Equation 1-1]

Here, x , y , z are the coordinates of the existing candidate area (the coordinates of the current point), x' , y' , z'are the coordinates of the changed boundary point, and Rx(θ) , Ry(θ) and Rz(θ ) Can be calculated by the following equation 1-2.

[Equation 1-2]

Here, θ may be a rotation angle corresponding to the selected cross section.

In addition, according to an exemplary embodiment of the present disclosure, when Equation 1-1 and Equation 1-2 generate a plurality of cross-sections 1 ′ in a predetermined angular direction, a point matching the determined candidate region is generated as a cross-section 1 ′. ) Can also be used in the case of converting to coordinates within.

In summary, the region division performance improvement apparatus 100 generates cross-sections 1'in various angular directions for a candidate area corresponding to an ambiguous boundary surface, and clearly identifies boundary information on the region of interest among the generated cross-sections 1'. By selecting a section (1') that can be provided according to a predetermined standard to create a 3D model (2), or by correcting a pre-built 3D model (2'), the boundary of the selected area is clear even in sharp or angular areas. It is possible to provide a calibrated 3D model (2) that is classified accordingly.

FIG. 8 is a diagram for explaining determining as a selected cross section a cross section having improved segmentation performance for an ROI for a determined candidate region.

Referring to FIG. 8, (a) of FIG. 8 shows a candidate region corresponding to an ambiguous boundary obtained as a result of segmentation, and (b) of FIG. 8 is generated in a predetermined angular direction to include at least a portion of the candidate region. An example of a plurality of cross-sections (1') is shown, and (c1) and (c2) of FIG. 3 are cross-sections (1') determined as a selected cross-section by improving the segmentation performance based on the change in pixel brightness associated with the candidate area. Is shown by way of example.

For example, (c1) may be a selected cross section in the angular direction in which the θ value is 20 degrees, and (c2) may be a selected cross section in the angular direction in which the θ value is 110 degrees.

9 is a schematic configuration diagram of an apparatus for improving region division performance of medical image data according to an exemplary embodiment of the present disclosure.

Referring to FIG. 9, the apparatus 100 for improving region division performance of medical image data according to an embodiment of the present disclosure includes a region divider 110, a candidate region determiner 120, a 3D transform unit 130, and It may include a cross-section selection unit 140.

The region dividing unit 110 may perform region division on an ROI for each of the plurality of input 2D medical image data 1.

The candidate region determiner 120 may determine a candidate region among the boundaries based on a change in pixel brightness at each point of the boundary of the region of interest obtained as a result of region division.

The 3D conversion unit 130 may construct a 3D model 2 ′ for an ROI from a plurality of 2D medical image data 1 based on a result of segmentation. In addition, the 3D conversion unit 130 may correct the 3D model based on the selected cross section determined by the cross section selection unit 140.

The cross-section selection unit 140 may generate a plurality of cross-sections 1 ′ in a predetermined angular direction to include at least a portion of the candidate region from the 3D model 2 ′. In addition, the cross-section selection unit 140 may determine a cross-section with improved segmentation performance for the region of interest based on a change in pixel brightness associated with the candidate region in each of the plurality of cross-sections 1 ′ as the selected cross-section.

Hereinafter, based on the details described above, the operation flow of the present application will be briefly described.

10 is a flowchart illustrating a method of improving segmentation performance of medical image data according to an exemplary embodiment of the present disclosure.

The method for improving region division performance of medical image data illustrated in FIG. 10 may be performed by the apparatus 100 for improving region division performance described above. Accordingly, even if omitted below, the description of the apparatus 100 for improving region division performance may be equally applied to a description of a method for improving region division performance of medical image data.

Referring to FIG. 10, in step S1010, the region dividing unit 110 may perform region division on an ROI for each of the inputted plurality of 2D medical image data 1.

Next, in operation S1020, the candidate region determiner 120 may determine a candidate region among the boundaries based on a change in pixel brightness at each point of the boundary with respect to the region of interest obtained as a result of region division.

Specifically, in step S1020, the candidate region determiner 120 may determine a point of each point of the boundary that does not satisfy a condition for changing the brightness of a pixel in the vertical direction with respect to the region of interest as the candidate region. In particular, the pixel brightness change condition is, according to an embodiment of the present application, at least a part of a region in which the differential value of the pixel brightness change amount exceeds a preset upper limit value in a region close to a corresponding point by a predetermined level or more, and a region close to a predetermined level It may be satisfied if it is kept below a preset lower limit value outside of

In addition, in step S1020, the candidate region determiner 120 may determine the candidate region in consideration of the anatomical characteristics of the organ corresponding to the region of interest.

In addition, according to an exemplary embodiment of the present disclosure, in step S1020, the candidate region determiner 120 may determine a candidate region at a point in which a curvature exceeds a preset critical curvature among the boundaries of the region of interest.

Next, in step S1030, the 3D transform unit 130 may construct a 3D model 2'for the ROI from the plurality of 2D medical image data 1 based on the result of segmentation.

Specifically, in step S1030, the 3D transform unit 130 may derive point cloud data for an ROI in a 3D space based on a plurality of 2D medical image data 1.

Next, in step S1040, the cross-section selection unit 140 may generate a plurality of cross-sections 1'in a predetermined angular direction to include at least a part of the candidate area from the 3D model 2'built in step S1030. have.

For example, in step S1040, the cross-section selection unit 140 may generate a plurality of cross-sections 1 ′ in all angular directions with respect to the candidate region.

As another example, in step S1040, the section selector 140 determines a predetermined angular direction that is determined to be greater than or equal to a preset threshold probability that the probability of improving the segmentation performance for the region of interest based on the sampled 3D data for the candidate region. By selecting, it is possible to create a plurality of cross-sections (1').

As another example, in step S1040, the cross-section selection unit 140 may generate a plurality of cross-sections 1'by selecting a predetermined angular direction in consideration of the anatomical characteristics of the organ corresponding to the region of interest.

Next, in step S1050, the section selector 150 may determine a section with improved segmentation performance for the region of interest as the selected section based on a change in pixel brightness associated with the candidate region in each of the plurality of sections 1'. have.

Next, in step S1060, the 3D conversion unit 130 may correct the 3D model based on the selected cross section.

In the above description, steps S1010 to S1060 may be further divided into additional steps or may be combined into fewer steps, according to an embodiment of the present disclosure. In addition, some steps may be omitted as necessary, or the order between steps may be changed.

The method for improving area division performance of medical image data according to an exemplary embodiment of the present disclosure 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, etc. alone or in combination. The program instructions recorded in the medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operation of the present invention, and vice versa.

In addition, the above-described method for improving area division performance of medical image data may be implemented in the form of a computer program or application executed by a computer stored in a recording medium.

The foregoing description of the present application is for illustrative purposes only, and those of ordinary skill in the art to which the present application pertains will be able to understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present application. Therefore, it should be understood that the embodiments described above are illustrative in all respects and are not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present application is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present application.

10: medical image data analysis system
100: Apparatus for improving area division performance of medical image data
110: area division
120: candidate area determination unit
130: 3D transform unit
140: section selector
1: 2D medical image data
1': cross section
2': 3D model before correction
2: 3D model after calibration

Claims (13)

In the method for improving area division performance of medical image data,
Performing region division on an ROI for each of the plurality of input 2D medical image data;
Determining a candidate region of the boundary based on a change in pixel brightness at each point of the boundary of the ROI obtained as a result of the division of the region;
Constructing a 3D model of an ROI from the plurality of 2D medical image data based on a result of the division of the region;
Generating a plurality of cross sections in a predetermined angular direction to include at least a portion of the candidate region from the 3D model;
Determining a cross section with improved segmentation performance for the region of interest as a selection cross section based on a change in pixel brightness associated with the candidate region in each of the plurality of cross sections; And
Correcting the 3D model based on the selected cross section,
Comprising, the method for improving the region division performance. The method of claim 1,
The step of determining the candidate region,
And determining a point of each point of the boundary that does not satisfy a condition for changing the brightness of a pixel in a vertical direction with respect to the region of interest as the candidate region. The method of claim 2,
The pixel brightness change condition is,
It is satisfied when there is at least a part of an area in which the differential value of the pixel brightness change amount exceeds a preset upper limit value in an area close to the point by a predetermined level or more, and maintains a preset lower limit value or less outside the area close to the predetermined level , How to improve segmentation performance. The method of claim 1,
Building the 3D model,
And deriving point cloud data for a region of interest in a 3D space based on the plurality of 2D medical image data. The method of claim 1,
The step of generating a plurality of cross-sections in the predetermined angular direction,
To generate a plurality of cross-sections in all angular directions for the candidate region, the method for improving region division performance. The method of claim 1,
The step of generating a plurality of cross-sections in the predetermined angular direction,
To generate the plurality of cross-sections by selecting a predetermined angular direction determined to be greater than or equal to a preset threshold probability that the probability of improving the segmentation performance for the region of interest based on the sampled 3D data for the candidate region How to improve partitioning performance. The method of claim 1,
The region of interest is to include a predetermined intraperitoneal organ (Organ), the method for improving the region division performance. The method of claim 7,
The step of generating a plurality of cross-sections in the predetermined angular direction,
To generate the plurality of cross-sections by selecting a predetermined angular direction in consideration of the anatomical characteristics of the organ corresponding to the region of interest. The method of claim 7,
The step of determining the candidate region,
A method for improving region division performance, characterized in that it is performed in consideration of the anatomical characteristics of an organ corresponding to the region of interest. The method of claim 9,
The step of determining the candidate region,
And determining the candidate region at a point in which a curvature exceeds a preset critical curvature among boundaries of the ROI. The method of claim 1,
The 2D medical image data includes CT image data. In the apparatus for improving area division performance of medical image data,
A region dividing unit for performing region division on an ROI for each of the plurality of input 2D medical image data;
A candidate region determining unit determining a candidate region among the boundaries based on a change in pixel brightness at each point of the boundary with respect to the region of interest obtained as a result of dividing the region;
A 3D conversion unit for constructing a 3D model for a region of interest from the plurality of 2D medical image data based on a result of the region division; And
A plurality of cross-sections in a predetermined angular direction are generated from the 3D model to include at least a portion of the candidate area, and segmentation performance for a region of interest based on a change in pixel brightness associated with the candidate area in each of the plurality of cross-sections A section selector that determines this enhanced section as the selected section,
Including,
The 3D conversion unit corrects the 3D model based on the selected cross section. The method of claim 12,
The cross-section selection unit,
Create a plurality of cross sections in all angular directions for the candidate region, or
Based on the sampled 3D data for the candidate region, the plurality of cross sections are generated by selecting a predetermined angular direction determined to be greater than or equal to a preset threshold probability that the probability of improving the segmentation performance for the region of interest, or
To generate the plurality of cross-sections by selecting a predetermined angular direction in consideration of the anatomical characteristics of the region of interest.

Images