WO2020141811A1 - Slide scanning system for microscope for bone marrow reading - Google Patents

Slide scanning system for microscope for bone marrow reading Download PDF

Info

Publication number
WO2020141811A1
WO2020141811A1 PCT/KR2019/018591 KR2019018591W WO2020141811A1 WO 2020141811 A1 WO2020141811 A1 WO 2020141811A1 KR 2019018591 W KR2019018591 W KR 2019018591W WO 2020141811 A1 WO2020141811 A1 WO 2020141811A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
region
interest
unit
bone marrow
Prior art date
Application number
PCT/KR2019/018591
Other languages
French (fr)
Korean (ko)
Inventor
최종호
이영득
Original Assignee
(주)유아이엠디
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by (주)유아이엠디 filed Critical (주)유아이엠디
Publication of WO2020141811A1 publication Critical patent/WO2020141811A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B21/00—Microscopes
    • G02B21/36—Microscopes arranged for photographic purposes or projection purposes or digital imaging or video purposes including associated control and data processing arrangements
    • G02B21/361—Optical details, e.g. image relay to the camera or image sensor
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B21/00—Microscopes
    • G02B21/34—Microscope slides, e.g. mounting specimens on microscope slides
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/60—Editing figures and text; Combining figures or text
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/18—Image warping, e.g. rearranging pixels individually

Definitions

  • the present invention relates to a scanning system for a microscope slide, and more particularly, to a microscope scanning scanning system for image processing and analysis of bone marrow cells.
  • the bone marrow is a soft tissue located in the inner space of the bone and is a hematopoietic organ that produces blood cells such as red blood cells, white blood cells, and platelets.
  • the bone marrow test is a function of bone marrow by observing bone marrow cells through a microscope and performing differential calculations after aspirating and biopsiing the bone marrow from large bones and staining them if necessary in the case of problems with blood cell production in the human body. This is a test to check for abnormal lesions.
  • This bone marrow examination is a task that requires a long time and fatigue of the doctor because it is necessary to observe various parts of the slide with high magnification as well as mobilizing the experience and knowledge of the working doctor in observing the slide.
  • the present invention has been derived based on the above needs, and the first object of the present invention is to reduce the amount of work of a doctor through rapid and efficient image processing and automatic image analysis of a scanned image on a smear slide of a bone marrow sample. It is to provide a slide scanning system for a microscope for bone marrow reading.
  • the second object of the present invention is a microscope slide for bone marrow reading that can reduce image processing time by dividing and binarizing stitching for a whole image and a region of interest image in stitching a divided image of a scanned slide into a whole image. It is to provide a scanning system.
  • the third object of the present invention is to accurately detect a region of interest among all images stitching a divided image of a scanned slide and extract features corresponding to the region of interest image, key point matching and accurate 3D information. It is to provide a slide scanning system for a microscope for bone marrow reading capable of outputting a natural region of interest image.
  • the fourth object of the present invention is to provide a slide scanning system for a microscope for bone marrow reading capable of automatically outputting a natural region of interest image by automatically editing a blank of a non-featured region with respect to the region of interest image.
  • a slide scanning system for a microscope for bone marrow reading scanning a smear slide prepared based on a bone marrow sample into a plurality of divided regions by motion control, but corresponding to a plurality of divided regions
  • a scanning unit acquiring a plurality of divided images
  • a lens distortion correction unit that corrects a lens distortion effect on the obtained multiple divided images
  • An encoder stitching unit that tiles the corrected plurality of divided images to stitch the entire image of the slide;
  • an image stitching unit stitching a plurality of divided images corresponding to a region of interest image.
  • the microscope-sized slide scanning system for bone marrow reading may further include a region-of-interest detection unit that detects a region-of-interest image containing at least one of particles and megakaryotic cells in the entire image stitched with the encoder.
  • the region of interest detector may be configured to detect an image of the region of interest based on the size of particles and megakaryocytes.
  • the image stitching unit a feature extraction unit for extracting a feature (Feature) from a plurality of divided images corresponding to the region of interest image, a key point matching unit for matching the adjacent adjacent images based on the position point of the feature, the obtained position point It may include an image warping unit for acquiring 3D information based on the cumulative value information of a feature and warping (dividing geometrically transformed) adjacent images with each other based on the obtained 3D information.
  • the image stitching unit edits the region of interest image by removing all of the segmented images of the non-featured region having no features among the warped divided images and one or more of the divided images of the non-featured region's width and height in the long axis direction. It may be to further include an image editing unit.
  • the microscope slide scanning system for bone marrow reading may further include a region of interest image output unit for outputting a warped region of interest image or an edited region of interest image.
  • the region of interest is automatically detected from the entire image stitched to the divided image of the scanned slide, and features of extracted images corresponding to the region of interest image are extracted, and accurate and natural region of interest images are output based on key point matching and 3D information. can do.
  • a natural region of interest image can be output by automatically editing the blank of the non-featured region with respect to the region of interest image.
  • FIG. 1 is a block diagram showing an embodiment configuration according to the slide scanning system for a microscope for the bone marrow reading of the present inventors
  • FIG. 2 is a view showing an example of an image in which a divided image obtained through slide scanning is assembled by an embodiment according to a slide scanning system for a microscope for bone marrow reading according to the present inventors;
  • LSC lens shading correction
  • LSC Lens Shading Correction
  • FIG. 4 is a view showing images comparing particles (left) and megakaryocytes (center) and normal cells (right), which are targets for detecting a region of interest image;
  • FIG. 5 is a view showing a plurality of regions of interest automatically detected by the region-of-interest detection unit according to an embodiment of the present invention according to a microscope slide scanning system for bone marrow reading;
  • Figure 6 is a configuration diagram showing the configuration of the image stitching unit in one embodiment configuration according to the present invention the microscope slide scanning system for bone marrow reading,
  • FIG. 7 is a view showing an example of 9 divided images constituting a region of interest image
  • FIG. 8 is a view showing a state in which a feature extraction unit extracts a feature (red) from an image of a region of interest during configuration of an embodiment according to a slide scanning system for a microscope for bone marrow reading according to the present inventors;
  • FIG. 9 is a diagram showing an image in which a non-featured region (a region having no features or a small number of features) in a region of interest image is detected,
  • FIG. 10 is a view showing an image of a region of interest (left) and a region of interest (image on the right) that is edited and output in the long axis direction of a blank, in which a non-feature region is processed as a blank (black portion).
  • One embodiment of a slide scanning system for a microscope for bone marrow reading prepares a bone marrow specimen prepared for bone marrow reading as a smear slide to acquire the image and performs processing and analysis of the slide image to read the obtained bone marrow image do. According to this embodiment, it is possible to increase the operation speed of the bone marrow reading, obtain and output a natural slide image in the process of assembling the divided image, and automatically search and output the Region of Interest of the working physician.
  • 1 is a configuration diagram showing a configuration of an embodiment according to a slide scanning system for a microscope for bone marrow reading according to the present inventors.
  • the scanning unit 10 the lens distortion correction unit 20, the divided image storage unit 30, the encoder stitching unit 40, the entire image output unit 50, interest, as shown in FIG. It comprises a region selection unit 60, an image stitching unit 70, a region of interest image output unit 80.
  • the scanning unit 10 scans the microscopic smear slide prepared based on the bone marrow sample into a plurality of divided areas by motion control, but acquires a plurality of divided images corresponding to the plurality of divided areas.
  • the scanning unit 10 includes motions such as imaging means including a low-magnification lens and an image sensor, supporting means for fixing the slide, lighting means for applying light, and imaging means for completely imaging a divided area of the slide.
  • Position control means for controlling the may be provided.
  • a plurality of sub-images obtained through scanning of the scanning unit 10 are overlapped with each other, and images of the slide area are completely captured to obtain a plurality of divided images.
  • the lens distortion correction unit 20 serves to correct the lens distortion effect on the obtained multiple divided images.
  • the image sensor constituting the scanning unit 10 and the lens disposed at the front end thereof have lens shading in which the peripheral portion is darker than the center of the image in the light receiving step. Therefore, stitching a plurality of divided images leads to unnatural results (the image on the left in FIG. 3), so lens shading correction (LSC) based on Mean-Image generation of the RGB image is performed.
  • LSC lens shading correction
  • the result of performing the LSC can be obtained as a natural image as shown in the right image of FIG. 3.
  • the divided image with lens shading correction is stored in the divided image storage unit 30.
  • the encoder stitching unit 40 serves to encoder stitch the entire image of the slide by tiling a plurality of corrected divided images.
  • stitching the entire image to obtain the entire image of the slide it may take an excessive amount of time because stitching hundreds of divided images.
  • Each of the plurality of low-magnification divided images has a large data size for each image, and the image processing process of comparing the divided images is inevitably more complicated. Therefore, the encoder stitching unit 40 performs stitching based on the positions of a plurality of divided images, but re-sizing the image of the divided images to perform tiled encoder stitching. Through this, the speed of acquiring the entire image of the slide can be improved.
  • the entire image stitched through the encoder stitching unit 40 can be visually output through the entire image output unit 50 and can be used as reading data of a working doctor.
  • the region of interest selection unit 60 serves to select a region of interest image corresponding to a region of interest from among the entire images stitched by the encoder. This may be a passive method based on an input signal from a working physician, or may be an automated method selected based on a detection signal to a region-of-interest detection unit (not shown) to be described later.
  • the image stitching unit 70 serves to stitch a plurality of divided images corresponding to the region of interest image.
  • the Region of Interest is the focus observation area of the working physician who must focus and observe for bone marrow reading.
  • the region of interest which is such an important observation region, is mainly a region in which particles or megakaryocytes are located, and the size of a megakaryocyte is about 5 times that of a cell, and the particle is about 100 times that of a normal cell. It has the above size. (See Fig. 4)
  • the particles are generic to those containing fat, bones, and the like.
  • this embodiment detects an image of the region of interest containing at least one of particles and megakaryotic cells based on the size of the encoder stitched whole image.
  • a region of interest detection unit (not shown) may be further included. As shown in the example shown in FIG. 5, a region of interest detector may output a total of 6 regions of interest images based on the size of the region of interest among all images of the slide.
  • the image stitching unit 70 a feature extraction unit 710 for extracting a feature (Feature) from a plurality of divided images corresponding to the region of interest image, based on the position point of the feature key points to match the adjacent adjacent images Matching unit 720, the image obtained by acquiring 3D information based on the accumulated value information of the feature at the acquired position point and warping each other adjacent images based on the obtained 3D information (Warping; geometric transformation)
  • a warping portion 730 may be included.
  • the feature extraction unit 710 extracts features based on color, texture, frequency series, size, and the like, of 9 divided images corresponding to the region of interest image as illustrated in FIG. 7.
  • the region of interest image illustrated in FIG. 8 is illustrated by processing a red region at a location point corresponding to a feature.
  • the key point matching unit 720 may naturally stitch mutually adjacent images by matching divided images adjacent to each other based on the location point of each feature.
  • the key point matching unit 720 utilizes the K-Nearest Neighbor (KNN) method in the Fast Livrary for Approximate Nearest Neighbors (FLANN), but various matching methods may be used.
  • KNN K-Nearest Neighbor
  • FLANN Approximate Nearest Neighbors
  • the image warping unit 730 warps the segmented image by adding depth information to the 2D information to prevent an error of simply stitching to a flat image since the bone marrow image is a slide image but has 3D information. .
  • the image editing unit 740 may include a divided image of a non-featured region having no features among warped (geometrically transformed) divided images, and one or more divided images adjacent in a long axis direction among the width and height of the divided image of the non-featured region. Removes and serves to edit the image of the region of interest.
  • the non-featured region may be output as a blank black portion without the configuration of the image editing unit 740, resulting in unnaturalness of the region of interest image. Therefore, the image editing unit 740 excludes the non-feature region from the region of interest image to form a rectangular image. Since such editing occurs when the cells are located at a portion other than the center, the editing process may be omitted when the cells are not biased.
  • the microscope slide scanning system for bone marrow reading may further include a region of interest image output unit 80 to output a warped region image or an edited region of interest image.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Optics & Photonics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

An effect from one embodiment of the present invention is that the amount of work required of a doctor can be reduced through the rapid and efficient image processing and automatic image analysis of a scanned image of a smear slide of a bone marrow sample. To this end, in particular, one embodiment of the present invention relates to a slide scanning system for a microscope for bone marrow reading, the system comprising: a scanning unit for scanning a smear slide, prepared on the basis of a bone marrow sample, into a plurality of divided regions divided by means of motion control, and obtaining a plurality of divided images corresponding to the plurality of divided regions; a lens distortion correction unit for correcting a lens distortion effect on the plurality of obtained divided images; an encoder stitching unit for tiling the plurality of corrected divided images and encoder-stitching an overall image of the slide; a region of interest selection unit for selecting a region of interest image corresponding to a region of interest in the overall encoder-stitched image; and an image stitching unit for stitching the plurality of divided images corresponding to the region of interest image.

Description

골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템Microscope slide scanning system for bone marrow reading
본 발명은 현미경용 슬라이드의 스캐닝 시스템에 관한 것으로서, 더욱 상세하게는 골수세포의 이미지 처리 및 분석에 관한 현미경용 슬라이드 스캐닝 시스템에 관한 것이다.The present invention relates to a scanning system for a microscope slide, and more particularly, to a microscope scanning scanning system for image processing and analysis of bone marrow cells.
골수는 뼈의 안쪽 공간에 위치한 부드러운 조직으로 적혈구, 백혈구, 혈소판과 같은 혈액세포를 생성하는 조혈기관이다. 또한 골수 검사는 인체에서 혈구 생성에 문제가 생긴 경우 큰 뼈에서 골수를 흡인하고 생검하여 슬라이드로 제작하고 필요한 경우에는 염색한 후, 현미경을 통해 골수 세포를 관찰하고 감별 계산을 수행함으로써 골수의 기능 및 비정상적인 병변을 확인하는 검사이다.The bone marrow is a soft tissue located in the inner space of the bone and is a hematopoietic organ that produces blood cells such as red blood cells, white blood cells, and platelets. In addition, the bone marrow test is a function of bone marrow by observing bone marrow cells through a microscope and performing differential calculations after aspirating and biopsiing the bone marrow from large bones and staining them if necessary in the case of problems with blood cell production in the human body. This is a test to check for abnormal lesions.
이러한 골수 검사는 슬라이드의 관찰에 작업 의사의 경험과 지식이 동원될 뿐만아니라 슬라이드의 여러 부위를 고배율로 관찰해야 하므로 장시간이 소요되고 의사의 피로가 수반되는 작업이다.This bone marrow examination is a task that requires a long time and fatigue of the doctor because it is necessary to observe various parts of the slide with high magnification as well as mobilizing the experience and knowledge of the working doctor in observing the slide.
더군다나 스캐닝을 통해 획득한 다수의 분할된 슬라이드 이미지의 전체 이미지화 작업의 소요 시간이 길어지거나 슬라이드 이미지의 관심영역(ROI; Region of Interest)을 일일이 수동적으로 서치하는 것은 작업량과 소요시간을 가중시키는 문제점이 있었다.Moreover, it takes a long time to complete the entire imaging operation of a plurality of divided slide images acquired through scanning, or manually searching the region of interest (ROI) of the slide image is a problem that increases the amount of work and time required. there was.
따라서 골수 판독을 위한 슬라이드의 스캐닝 시스템에서 작업 의사의 작업 전에 슬라이드 이미지에 대한 최적의 이미지 처리 및 자동 분석에 관한 기술의 필요성이 대두된다.Therefore, in the scanning system of the slide for bone marrow reading, a need arises for a technique for optimal image processing and automatic analysis of the slide image before the work of the operator.
본 발명은 상기와 같은 필요성에 기하여 도출된 것으로서, 본 발명의 제1 목적은 골수 샘플의 도말 슬라이드에 대한 스캐닝된 이미지의 신속하고 효율적인 이미지 처리와 이미지 자동 분석을 통해 의사의 작업량을 감소시킬 수 있는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템을 제공하는 데 있다.The present invention has been derived based on the above needs, and the first object of the present invention is to reduce the amount of work of a doctor through rapid and efficient image processing and automatic image analysis of a scanned image on a smear slide of a bone marrow sample. It is to provide a slide scanning system for a microscope for bone marrow reading.
본 발명의 제2 목적은, 스캐닝된 슬라이드의 분할 이미지를 전체 이미지로 스티칭함에 있어서 전체 이미지와 관심 영역 이미지에 대한 스티칭을 이분화 및 이진화함으로써 이미지 처리 시간을 줄일 수 있는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템을 제공하는 데 있다.The second object of the present invention is a microscope slide for bone marrow reading that can reduce image processing time by dividing and binarizing stitching for a whole image and a region of interest image in stitching a divided image of a scanned slide into a whole image. It is to provide a scanning system.
본 발명의 제3 목적은, 스캐닝된 슬라이드의 분할 이미지를 스티칭한 전체 이미지 중 관심 영역을 자동으로 검출하고 관심 영역 이미지에 대응하는 분할 이미지들의 특징 추출, 키포인트 매칭 및 3 차원 정보에 기반하여 정확하고 자연스러운 관심 영역 이미지를 출력할 수 있는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템을 제공하는 데 있다.The third object of the present invention is to accurately detect a region of interest among all images stitching a divided image of a scanned slide and extract features corresponding to the region of interest image, key point matching and accurate 3D information. It is to provide a slide scanning system for a microscope for bone marrow reading capable of outputting a natural region of interest image.
본 발명의 제4 목적은, 관심 영역 이미지에 대하여 비특징 영역의 공백을 자동 편집하여 자연스러운 관심 영역 이미지를 출력할 수 있는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템을 제공하는 데 있다.The fourth object of the present invention is to provide a slide scanning system for a microscope for bone marrow reading capable of automatically outputting a natural region of interest image by automatically editing a blank of a non-featured region with respect to the region of interest image.
상기와 같은 본 발명의 목적은, 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 있어서, 골수 샘플에 기초하여 준비된 도말 슬라이드를 모션 제어에 의해 다수의 분할된 영역으로 스캐닝하되 다수의 분할된 영역에 대응하는 다수의 분할 이미지를 획득하는 스캐닝부; 획득된 다수의 분할 이미지에 대하여 렌즈 왜곡 효과를 보정하는 렌즈 왜곡 보정부; 보정된 다수의 분할 이미지를 타일링(Tiling)하여 슬라이드의 전체 이미지를 인코더 스티칭하는 인코더 스티칭부; 인코더 스티칭된 전체 이미지 중 관심 영역(Region of Interest)에 대응하는 관심 영역 이미지를 선택하는 관심 영역 선택부; 및 관심 영역 이미지에 대응하는 다수의 분할 이미지를 스티칭하는 이미지 스티칭부를 포함하는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템을 제공함으로써 달성될 수 있다.The object of the present invention as described above, in a slide scanning system for a microscope for bone marrow reading, scanning a smear slide prepared based on a bone marrow sample into a plurality of divided regions by motion control, but corresponding to a plurality of divided regions A scanning unit acquiring a plurality of divided images; A lens distortion correction unit that corrects a lens distortion effect on the obtained multiple divided images; An encoder stitching unit that tiles the corrected plurality of divided images to stitch the entire image of the slide; A region of interest selection unit for selecting a region of interest image corresponding to a region of interest from among the encoder stitched whole images; And an image stitching unit stitching a plurality of divided images corresponding to a region of interest image.
골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템은, 인코더 스티칭된 전체 이미지에서 파티클 및 거핵세포 중 적어도 1 이상이 포함된 관심 영역 이미지를 검출하는 관심 영역 검출부를 더 포함할 수 있다.The microscope-sized slide scanning system for bone marrow reading may further include a region-of-interest detection unit that detects a region-of-interest image containing at least one of particles and megakaryotic cells in the entire image stitched with the encoder.
여기서 관심 영역 검출부는 파티클 및 거핵세포를 사이즈에 기반하여 관심 영역 이미지를 검출하는 것일 수 있다.Here, the region of interest detector may be configured to detect an image of the region of interest based on the size of particles and megakaryocytes.
또한 이미지 스티칭부는, 관심 영역 이미지에 대응하는 다수의 분할 이미지로부터 특징(Feature)을 추출하는 특징 추출부, 특징 중 위치 포인트에 기반하여 상호 인접하는 분할 이미지들을 매칭시키는 키포인트 매칭부, 획득된 위치 포인트에서 특징의 누적값 정보에 기반하여 3차원 정보를 획득하고 획득된 3차원 정보에 기반하여 상호 인접하는 분할 이미지들을 서로 와핑(Warping; 기하학적 변환)하는 이미지 와핑부를 포함할 수 있다.In addition, the image stitching unit, a feature extraction unit for extracting a feature (Feature) from a plurality of divided images corresponding to the region of interest image, a key point matching unit for matching the adjacent adjacent images based on the position point of the feature, the obtained position point It may include an image warping unit for acquiring 3D information based on the cumulative value information of a feature and warping (dividing geometrically transformed) adjacent images with each other based on the obtained 3D information.
그리고 이미지 스티칭부는, 와핑된 분할 이미지들 중 특징이 없는 비특징 영역의 분할 이미지와, 비특징 영역의 분할 이미지의 너비와 높이 중 장축 방향으로 인접한 1 이상의 분할 이미지를 모두 제거하여 관심 영역 이미지를 편집하는 이미지 편집부를 더 포함하는 것일 수 있다.In addition, the image stitching unit edits the region of interest image by removing all of the segmented images of the non-featured region having no features among the warped divided images and one or more of the divided images of the non-featured region's width and height in the long axis direction. It may be to further include an image editing unit.
한편 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템은, 와핑된 관심 영역 이미지 또는 편집된 관심 영역 이미지를 출력하는 관심 영역 이미지 출력부를 더 포함하는 것일 수 있다.Meanwhile, the microscope slide scanning system for bone marrow reading may further include a region of interest image output unit for outputting a warped region of interest image or an edited region of interest image.
상기와 같은 본 발명의 일 실시예에 의하면, 골수 샘플의 도말 슬라이드에 대한 스캐닝된 이미지의 신속하고 효율적인 이미지 처리와 이미지 자동 분석을 통해 의사의 작업량을 감소시킬 수 있는 효과가 있다.According to an embodiment of the present invention as described above, it is possible to reduce the amount of work of a doctor through rapid and efficient image processing and automatic image analysis of a scanned image of a smear slide of a bone marrow sample.
또한 스캐닝된 슬라이드의 분할 이미지를 전체 이미지로 스티칭함에 있어서 전체 이미지와 관심 영역 이미지에 대한 스티칭을 이분화 및 이진화함으로써 이미지 처리 시간을 줄일 수 있는 효과가 있다.In addition, in stitching the divided image of the scanned slide into the entire image, it is possible to reduce image processing time by dividing and binarizing stitching for the entire image and the region of interest image.
그리고 스캐닝된 슬라이드의 분할 이미지를 스티칭한 전체 이미지 중 관심 영역을 자동으로 검출하고 관심 영역 이미지에 대응하는 분할 이미지들의 특징 추출, 키포인트 매칭 및 3 차원 정보에 기반하여 정확하고 자연스로운 관심 영역 이미지를 출력할 수 있다.Then, the region of interest is automatically detected from the entire image stitched to the divided image of the scanned slide, and features of extracted images corresponding to the region of interest image are extracted, and accurate and natural region of interest images are output based on key point matching and 3D information. can do.
더욱이 관심 영역 이미지에 대하여 비특징 영역의 공백을 자동 편집하여 자연스러운 관심 영역 이미지를 출력할 수 있는 효과가 있다.Furthermore, there is an effect that a natural region of interest image can be output by automatically editing the blank of the non-featured region with respect to the region of interest image.
도 1은 본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 따른 일 실시예 구성을 나타낸 구성도,1 is a block diagram showing an embodiment configuration according to the slide scanning system for a microscope for the bone marrow reading of the present inventors,
도 2는 본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 따른 일 실시예에 의해 슬라이드 스캐닝을 통해 획득한 분할 이미지가 조립된 이미지의 일 예를 나타낸 도면,2 is a view showing an example of an image in which a divided image obtained through slide scanning is assembled by an embodiment according to a slide scanning system for a microscope for bone marrow reading according to the present inventors;
도 3은 본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 따른 일 실시예 구성 중 렌즈 쉐이딩 보정(LSC; Lens Shading Correction)을 적용하지 않은 경우(왼편)와, 렌즈 쉐이딩 보정(LSC; Lens Shading Correction)을 적용한 경우 각각의 스티칭된 전체 이미지를 나타낸 도면,3 is a case in which the lens shading correction (LSC; Lens Shading Correction) is not applied (left) and the lens shading correction (LSC) during the embodiment configuration according to the slide scanning system for a microscope for bone marrow reading by the present inventors ) Is applied to each stitched full image,
도 4는 관심 영역 이미지 검출의 대상이 되는 파티클(왼편) 및 거핵세포(중앙)와 일반 세포(오른편)를 비교한 이미지들을 나타낸 도면,4 is a view showing images comparing particles (left) and megakaryocytes (center) and normal cells (right), which are targets for detecting a region of interest image;
도 5는 본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 따른 일 실시예 구성 중 관심 영역 검출부가 자동으로 검출한 다수의 관심 영역 이미지를 나타낸 도면,5 is a view showing a plurality of regions of interest automatically detected by the region-of-interest detection unit according to an embodiment of the present invention according to a microscope slide scanning system for bone marrow reading;
도 6은 본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 따른 일 실시예 구성 중 이미지 스티칭부의 구성을 나타낸 구성도,Figure 6 is a configuration diagram showing the configuration of the image stitching unit in one embodiment configuration according to the present invention the microscope slide scanning system for bone marrow reading,
도 7은 관심 영역 이미지를 구성하는 9 장의 분할 이미지의 예시를 나타낸 도면,7 is a view showing an example of 9 divided images constituting a region of interest image;
도 8은 본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 따른 일 실시예 구성 중 특징 추출부가 관심 영역 이미지로부터 특징(빨강색)을 추출한 상태를 나타낸 도면,8 is a view showing a state in which a feature extraction unit extracts a feature (red) from an image of a region of interest during configuration of an embodiment according to a slide scanning system for a microscope for bone marrow reading according to the present inventors;
도 9는 관심 영역 이미지에서 비특징 영역(세포가 존재하지 않아 특징이 없거나 특징 수가 적은 영역)을 검출한 이미지를 나타낸 도면,9 is a diagram showing an image in which a non-featured region (a region having no features or a small number of features) in a region of interest image is detected,
도 10은 비특징 영역이 공백(검은색 부분)으로 처리되어 출력되는 관심영역 이미지(왼편)와 공백의 장축 방향으로 편집되어 출력된 관심 영역 이미지(오른편)를 나타낸 도면이다.FIG. 10 is a view showing an image of a region of interest (left) and a region of interest (image on the right) that is edited and output in the long axis direction of a blank, in which a non-feature region is processed as a blank (black portion).
이하 첨부 도면들 및 첨부 도면들에 기재된 내용들을 참조하여 본 발명의 실시예를 상세하게 설명하지만, 본 발명이 실시예에 의해 제한되거나 한정되는 것은 아니다.Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and the contents described in the accompanying drawings, but the present invention is not limited or limited by the embodiments.
아래 설명하는 실시예들에는 다양한 변경이 가해질 수 있다. 아래 설명하는 실시예들은 실시 형태에 대해 한정하려는 것이 아니며, 이들에 대한 모든 변경, 균등물 내지 대체물을 포함하는 것으로 이해되어야 한다.Various modifications may be made to the embodiments described below. The examples described below are not intended to be limiting with respect to the embodiments, and should be understood to include all modifications, equivalents, or substitutes thereof.
한편, 본 발명을 설명함에 있어서, 관련된 공지 기능 또는 구성에 대한 구체적인 설명이 본 발명의 요지를 불필요하게 흐릴 수 있다고 판단되는 경우에는, 그 상세한 설명을 생략할 것이다. 그리고, 본 명세서에서 사용되는 용어(terminology)들은 본 발명의 실시예를 적절히 표현하기 위해 사용된 용어들로서, 이는 사용자, 운용자의 의도 또는 본 발명이 속하는 분야의 관례 등에 따라 달라질 수 있다. 따라서, 본 용어들에 대한 정의는 본 명세서 전반에 걸친 내용을 토대로 내려져야 할 것이다.On the other hand, in the description of the present invention, when it is determined that a detailed description of related known functions or configurations may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted. In addition, terms used in the present specification (terminology) are terms used to properly represent an embodiment of the present invention, which may vary depending on a user, an operator's intention, or a custom in a field to which the present invention pertains. Therefore, definitions of these terms should be made based on the contents throughout the present specification.
골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템Microscope slide scanning system for bone marrow reading
본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템의 일 실시예는 골수 판독을 위해 준비된 골수 표본을 도말 슬라이드로 제작하여 그 이미지를 획득하고 획득된 골수 이미지를 판독하기 위해 슬라이드 이미지의 처리 및 분석을 수행한다. 본 실시예에 의하면 골수 판독의 작업 속도를 증가시키고 분할 이미지의 조립 과정에서 자연스러운 슬라이드 이미지를 획득하여 출력할 수 있으며 자동으로 작업 의사의 관심 영역(Region of Interest)을 서치 및 출력할 수 있다.One embodiment of a slide scanning system for a microscope for bone marrow reading according to the present inventors prepares a bone marrow specimen prepared for bone marrow reading as a smear slide to acquire the image and performs processing and analysis of the slide image to read the obtained bone marrow image do. According to this embodiment, it is possible to increase the operation speed of the bone marrow reading, obtain and output a natural slide image in the process of assembling the divided image, and automatically search and output the Region of Interest of the working physician.
도 1은 본 발명인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 따른 일 실시예 구성을 나타낸 구성도이다. 본 실시예는 도 1에 도시된 바와 같이, 스캐닝부(10), 렌즈 왜곡 보정부(20), 분할 이미지 저장부(30), 인코더 스티칭부(40), 전체 이미지 출력부(50), 관심 영역 선택부(60), 이미지 스티칭부(70), 관심영역 이미지 출력부(80)를 포함하여 구성된다.1 is a configuration diagram showing a configuration of an embodiment according to a slide scanning system for a microscope for bone marrow reading according to the present inventors. 1, the scanning unit 10, the lens distortion correction unit 20, the divided image storage unit 30, the encoder stitching unit 40, the entire image output unit 50, interest, as shown in FIG. It comprises a region selection unit 60, an image stitching unit 70, a region of interest image output unit 80.
스캐닝부(10)는 골수 샘플에 기초하여 준비된 현미경용 도말 슬라이드를 모션 제어에 의해 다수의 분할된 영역으로 스캐닝하되 다수의 분할된 영역에 대응하는 다수의 분할 이미지를 획득하는 역할을 한다. 이러한 스캐닝부(10)는 저배율 렌즈와 이미지 센서를 포함하는 촬상 수단과, 슬라이드를 고정하는 지지 수단과, 조명을 가하는 조명 수단, 그리고 슬라이드의 분할된 영역을 빠짐없이 촬상하기 위해 촬상 수단 등의 모션을 제어하는 위치 제어 수단 등이 구비될 수 있다. The scanning unit 10 scans the microscopic smear slide prepared based on the bone marrow sample into a plurality of divided areas by motion control, but acquires a plurality of divided images corresponding to the plurality of divided areas. The scanning unit 10 includes motions such as imaging means including a low-magnification lens and an image sensor, supporting means for fixing the slide, lighting means for applying light, and imaging means for completely imaging a divided area of the slide. Position control means for controlling the may be provided.
스캐닝부(10)의 스캐닝을 통해 획득되는 다수의 부할 이미지는, 도 2에 도시된 바와 같이, 분할된 이미지가 서로 오버랩되며 슬라이드 영역을 빠짐없이 촬상하여 다수의 분할 이미지를 획득한다.As shown in FIG. 2, a plurality of sub-images obtained through scanning of the scanning unit 10 are overlapped with each other, and images of the slide area are completely captured to obtain a plurality of divided images.
렌즈 왜곡 보정부(20)는 획득된 다수의 분할 이미지에 대하여 렌즈 왜곡 효과를 보정하는 역할을 한다. 스캐닝부(10)를 구성하는 이미지 센서와 그 앞단에 배치되는 렌즈는 수광 단계에서 이미지의 가운데에 비하여 주변부가 어두워지는 렌즈 쉐이딩(Lens shading)이 있게 된다. 따라서 다수의 분할된 이미지를 스티칭(stitching)하면 부자연스런 결과(도 3의 왼편 이미지)로 이어지므로 RGB 이미지의 Mean-Image generation에 기반한 렌즈 쉐이딩 보정(LSC; Lens Shading Correction)이 수행된다. LSC가 수행된 결과는 도 3의 오른편 이미지와 같이 자연스러운 이미지로 획득될 수 있다. 렌즈 쉐이딩 보정이 이루어진 분할 이미지는 분할 이미지 저장부(30)에 저장된다.The lens distortion correction unit 20 serves to correct the lens distortion effect on the obtained multiple divided images. The image sensor constituting the scanning unit 10 and the lens disposed at the front end thereof have lens shading in which the peripheral portion is darker than the center of the image in the light receiving step. Therefore, stitching a plurality of divided images leads to unnatural results (the image on the left in FIG. 3), so lens shading correction (LSC) based on Mean-Image generation of the RGB image is performed. The result of performing the LSC can be obtained as a natural image as shown in the right image of FIG. 3. The divided image with lens shading correction is stored in the divided image storage unit 30.
인코더 스티칭부(40)는 보정된 다수의 분할 이미지를 타일링하여 슬라이드의 전체 이미지를 인코더 스티칭하는 역할을 한다. 슬라이드의 전체 이미지를 획득하기 위해 전체 이미지를 스티칭하는 경우 수백 개의 분할 이미지를 스티칭하므로 과다한 시간이 소요될 수 있다. 다수의 저배율 분할 이미지 각각은 이미지 별 데이터 크기가 크고 분할 이미지 상호 간을 비교하는 이미지 처리 과정은 더욱 복잡할 수 밖에 없다. 따라서 인코더 스티칭부(40)는 다수의 분할 이미지의 위치에 기반하여 스티칭하되 분할 이미지의 영상 크기를 재조정함으로써 타일링(tiling) 방식의 인코더 스티칭을 수행한다. 이를 통해 슬라이드의 전체 이미지를 획득하는 속도를 향상 시킬 수 있다. 인코더 스티칭부(40)를 통해 스티칭된 전체 이미지는 전체 이미지 출력부(50)를 통해 시각적으로 출력될 수 있으며 작업 의사의 판독 자료로 활용될 수 있다.The encoder stitching unit 40 serves to encoder stitch the entire image of the slide by tiling a plurality of corrected divided images. When stitching the entire image to obtain the entire image of the slide, it may take an excessive amount of time because stitching hundreds of divided images. Each of the plurality of low-magnification divided images has a large data size for each image, and the image processing process of comparing the divided images is inevitably more complicated. Therefore, the encoder stitching unit 40 performs stitching based on the positions of a plurality of divided images, but re-sizing the image of the divided images to perform tiled encoder stitching. Through this, the speed of acquiring the entire image of the slide can be improved. The entire image stitched through the encoder stitching unit 40 can be visually output through the entire image output unit 50 and can be used as reading data of a working doctor.
관심 영역 선택부(60)는 인코더 스티칭된 전체 이미지 중 관심 영역(Region of Interest)에 대응하는 관심 영역 이미지를 선택하는 역할을 한다. 이는 작업 의사의 입력 신호에 의한 수동적 방식일 수도 있으며 이후 설명할 관심 영역 검출부(도시되지 않음)에 검출 신호에 기반하여 선택되는 자동화 방식일 수 있다.The region of interest selection unit 60 serves to select a region of interest image corresponding to a region of interest from among the entire images stitched by the encoder. This may be a passive method based on an input signal from a working physician, or may be an automated method selected based on a detection signal to a region-of-interest detection unit (not shown) to be described later.
이미지 스티칭부(70)는 관심 영역 이미지에 대응하는 다수의 분할 이미지를 스티칭하는 역할을 한다. 관심 영역(Region of Interest)는 골수 판독을 위해 중점적으로 관찰하고 판독해야 하는 작업 의사의 중점 관찰 영역이다. 이러한 중점 관찰 영역인 관심 영역은 주로 파티클(particle) 또는 거핵세포(Megakaryocyte)가 위치하는 영역으로서 일반 세포(Cell)에 비하여 거핵세포는 약 5 배의 크기를 갖고 파티클은 일반 세포에 비하여 약 100 배 이상의 크기를 갖는다.(도 4 참조) 여기서 파티클은 지방(Fat), 골격(Bones) 등이 포함된 것을 총칭한다.The image stitching unit 70 serves to stitch a plurality of divided images corresponding to the region of interest image. The Region of Interest is the focus observation area of the working physician who must focus and observe for bone marrow reading. The region of interest, which is such an important observation region, is mainly a region in which particles or megakaryocytes are located, and the size of a megakaryocyte is about 5 times that of a cell, and the particle is about 100 times that of a normal cell. It has the above size. (See Fig. 4) Here, the particles are generic to those containing fat, bones, and the like.
관심 영역은 하나의 슬라이드에서 여러 개가 발견될 수 있으며 이를 자동으로 검출하기 위해 본 실시예는, 인코더 스티칭된 전체 이미지에서 파티클 및 거핵세포 중 적어도 1 이상이 포함된 관심 영역 이미지를 사이즈에 기반하여 검출하는 관심 영역 검출부(도시되지 않음)를 더 포함할 수 있다. 도 5에 도시된 일 예와 같이, 관심 영역 검출부는 슬라이드의 전체 이미지 중에서 관심 영역의 사이즈에 기반하여 총 6 개의 관심 영역 이미지가 출력될 수 있다.Multiple regions of interest can be found on one slide, and in order to automatically detect them, this embodiment detects an image of the region of interest containing at least one of particles and megakaryotic cells based on the size of the encoder stitched whole image. A region of interest detection unit (not shown) may be further included. As shown in the example shown in FIG. 5, a region of interest detector may output a total of 6 regions of interest images based on the size of the region of interest among all images of the slide.
이미지 스티칭부(70)는, 관심 영역 이미지에 대응하는 다수의 분할 이미지로부터 특징(Feature)을 추출하는 특징 추출부(710)와, 특징 중 위치 포인트에 기반하여 상호 인접하는 분할 이미지들을 매칭시키는 키포인트 매칭부(720)와, 획득된 위치 포인트에서 특징의 누적값 정보에 기반하여 3차원 정보를 획득하고 획득된 3차원 정보에 기반하여 상호 인접하는 분할 이미지들을 서로 와핑(Warping; 기하학적 변환)하는 이미지 와핑부(730)를 포함할 수 있다.The image stitching unit 70, a feature extraction unit 710 for extracting a feature (Feature) from a plurality of divided images corresponding to the region of interest image, based on the position point of the feature key points to match the adjacent adjacent images Matching unit 720, the image obtained by acquiring 3D information based on the accumulated value information of the feature at the acquired position point and warping each other adjacent images based on the obtained 3D information (Warping; geometric transformation) A warping portion 730 may be included.
특징 추출부(710)는 도 7에 도시된 바와 같은 관심 영역 이미지에 대응하는 9 장의 분할 이미지에 대하여 색깔, 질감, 주파수 계열, 사이즈 등에 기반하여 특징(feature)을 추출한다. 도 8에 도시된 관심 영역 이미지는 특징에 해당하는 위치 포인트에 붉은 색 영역을 처리하여 도시했다. 키포인트 매칭부(720)는 각 특징의 위치 포인트에 기반하여 상호 인접하는 분할 이미지들을 매칭시킴으로써 상호 인접하는 이미지를 자연스럽게 스티칭할 수 있다. 구체적으로 키포인트 매칭부(720)는 FLANN(Fast Livrary for Approximate Nearest Neighbors) 내에서 KNN(K-Nearest Neighbor) 방식을 활용하였지만 이외에도 다양한 매칭 방식을 활용할 수 있을 것이다. 이미지 와핑부(730)는 골수 이미지가 슬라이드 이미지이긴 하지만 3차원 정보를 가지고 있으므로 단순히 평면 이미지로 스티칭하는 오류를 방지하기 위해 2차원 정보에 깊이 정보를 더하여 분할 이미지를 와핑(Warping; 기하학적 변환)한다.The feature extraction unit 710 extracts features based on color, texture, frequency series, size, and the like, of 9 divided images corresponding to the region of interest image as illustrated in FIG. 7. The region of interest image illustrated in FIG. 8 is illustrated by processing a red region at a location point corresponding to a feature. The key point matching unit 720 may naturally stitch mutually adjacent images by matching divided images adjacent to each other based on the location point of each feature. Specifically, the key point matching unit 720 utilizes the K-Nearest Neighbor (KNN) method in the Fast Livrary for Approximate Nearest Neighbors (FLANN), but various matching methods may be used. The image warping unit 730 warps the segmented image by adding depth information to the 2D information to prevent an error of simply stitching to a flat image since the bone marrow image is a slide image but has 3D information. .
이미지 편집부(740)는 와핑(Warping; 기하학적 변환)된 분할 이미지들 중 특징이 없는 비특징 영역의 분할 이미지와, 비특징 영역의 분할 이미지의 너비와 높이 중 장축 방향으로 인접한 1 이상의 분할 이미지를 모두 제거하여 관심 영역 이미지를 편집하는 역할을 한다.The image editing unit 740 may include a divided image of a non-featured region having no features among warped (geometrically transformed) divided images, and one or more divided images adjacent in a long axis direction among the width and height of the divided image of the non-featured region. Removes and serves to edit the image of the region of interest.
비특징 영역은 이미지 편집부(740)의 구성이 없으면 공백로 검은 색 부분으로 출력될 수 있으므로 관심 영역 이미지의 부자연스러움을 초래한다. 따라서 이미지 편집부(740)는 이러한 비특징 영역을 관심 영역 이미지에서 배제하여 사각형의 이미지를 형성한다. 이러한 편집은 세포들이 센터가 아닌 부분에 위치하는 경우 발생되므로 세포가 치우쳐 있지 않은 경우에는 편집 과정이 생략될 수 있다.The non-featured region may be output as a blank black portion without the configuration of the image editing unit 740, resulting in unnaturalness of the region of interest image. Therefore, the image editing unit 740 excludes the non-feature region from the region of interest image to form a rectangular image. Since such editing occurs when the cells are located at a portion other than the center, the editing process may be omitted when the cells are not biased.
한편 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템은, 관심 영역 이미지 출력부(80)가 더 구비되어 와핑(Warping; 기하학적 변환)된 관심 영역 이미지 또는 편집된 관심 영역 이미지를 출력할 수 있다.Meanwhile, the microscope slide scanning system for bone marrow reading may further include a region of interest image output unit 80 to output a warped region image or an edited region of interest image.
이상 첨부된 도면을 참조하여 본 발명의 실시 예를 설명하였지만, 상술한 본 발명의 기술적 구성은 본 발명이 속하는 기술 분야의 당 업자가 본 발명의 그 기술적 사상이나 필수적 특징을 변경하지 않고서 다른 구체적인 형태로 실시될 수 있다는 것을 이해할 수 있을 것이다. 그러므로 이상에서 기술한 실시 예들은 모든 면에서 예시적인 것이며 한정적인 것이 아닌 것으로서 이해되어야 한다. 아울러, 본 발명의 범위는 상기의 상세한 설명보다는 후술하는 특허청구범위에 의하여 나타내어진다. 또한, 특허청구범위의 의미 및 범위 그리고 그 등가 개념으로부터 도출되는 모든 변경 또는 변형된 형태가 본 발명의 범위에 포함되는 것으로 해석되어야 한다.The embodiments of the present invention have been described above with reference to the accompanying drawings, but the technical configuration of the present invention described above is another specific form without a person skilled in the art to which the present invention pertains does not change the technical spirit or essential features of the present invention. It will be understood that can be carried out. Therefore, the above-described embodiments are to be understood in all respects as illustrative and not restrictive. In addition, the scope of the invention is indicated by the claims below, rather than the detailed description above. In addition, all modifications or variations derived from the meaning and scope of the claims and equivalent concepts should be construed as being included in the scope of the present invention.
[부호의 설명][Description of codes]
10: 스캐닝부10: scanning unit
20: 렌즈 왜곡 보정부20: lens distortion correction unit
30: 분할 이미지 저장부30: split image storage
40: 인코더 스티칭부40: encoder stitching unit
50: 전체 이미지 출력부50: full image output
60: 관심 영역 선택부60: region of interest selection unit
70: 이미지 스티칭부70: image stitching unit
80: 관심 영역 이미지 출력부80: image output section of interest
710: 특징 추출부710: feature extraction unit
720: 키포인트 매칭부720: key point matching unit
730: 이미지 와핑부730: image warping part
740: 이미지 편집부740: image editing department

Claims (6)

  1. 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템에 있어서,In the slide scanning system for a microscope for bone marrow reading,
    골수 샘플에 기초하여 준비된 도말 슬라이드를 모션 제어에 의해 다수의 분할된 영역으로 스캐닝하되 상기 다수의 분할된 영역에 대응하는 다수의 분할 이미지를 획득하는 스캐닝부;A scanning unit scanning a smear slide prepared based on a bone marrow sample into a plurality of divided regions by motion control, and obtaining a plurality of divided images corresponding to the divided regions;
    상기 획득된 다수의 분할 이미지에 대하여 렌즈 왜곡 효과를 보정하는 렌즈 왜곡 보정부;A lens distortion correcting unit which corrects a lens distortion effect on the obtained multiple divided images;
    상기 보정된 다수의 분할 이미지를 타일링(Tiling)하여 상기 슬라이드의 전체 이미지를 인코더 스티칭하는 인코더 스티칭부;An encoder stitching unit that tiles the corrected plurality of divided images to stitch the entire image of the slide;
    상기 인코더 스티칭된 전체 이미지 중 관심 영역(Region of Interest)에 대응하는 관심 영역 이미지를 선택하는 관심 영역 선택부; 및A region of interest selection unit for selecting a region of interest image corresponding to a region of interest from among all the stitched encoder images; And
    상기 관심 영역 이미지에 대응하는 다수의 분할 이미지를 스티칭하는 이미지 스티칭부를 포함하는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템.A microscope scanning scanning system for bone marrow reading comprising an image stitching unit stitching a plurality of divided images corresponding to the region of interest image.
  2. 제1 항에 있어서,According to claim 1,
    상기 인코더 스티칭된 전체 이미지에서 파티클 및 거핵세포 중 적어도 1 이상이 포함된 상기 관심 영역 이미지를 검출하는 관심 영역 검출부를 더 포함하는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템.Further comprising a region of interest detection unit for detecting the region of interest image containing at least one of the particles and megakaryocytes in the encoder stitched whole image Microscope slide scanning system for bone marrow reading.
  3. 제2 항에 있어서,According to claim 2,
    상기 관심 영역 검출부는 상기 파티클 및 상기 거핵세포를 사이즈에 기반하여 상기 관심 영역 이미지를 검출하는 것인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템.The region-of-interest detection unit detects the image of the region of interest based on the size of the particles and the megakaryocytes.
  4. 제2 항에 있어서,According to claim 2,
    상기 이미지 스티칭부는,The image stitching unit,
    상기 관심 영역 이미지에 대응하는 다수의 분할 이미지로부터 특징(Feature)을 추출하는 특징 추출부, 상기 특징 중 위치 포인트에 기반하여 상호 인접하는 분할 이미지들을 매칭시키는 키포인트 매칭부, 상기 획득된 위치 포인트에서 상기 특징의 누적값 정보에 기반하여 3차원 정보를 획득하고 상기 획득된 3차원 정보에 기반하여 상호 인접하는 분할 이미지들을 서로 와핑(Warping)하는 이미지 와핑부를 포함하는 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템.A feature extracting unit extracting a feature from a plurality of split images corresponding to the region of interest image, a key point matching unit matching the split images adjacent to each other based on a location point among the features, and the obtained location point A microscope scanning scanning system for bone marrow reading, comprising an image warping unit that acquires 3D information based on the accumulated value information of a feature and warps mutually adjacent divided images based on the acquired 3D information.
  5. 제4 항에 있어서,According to claim 4,
    상기 이미지 스티칭부는,The image stitching unit,
    상기 와핑된 분할 이미지들 중 상기 특징이 없는 비특징 영역의 분할 이미지와, 상기 비특징 영역의 분할 이미지의 너비와 높이 중 장축 방향으로 인접한 1 이상의 분할 이미지를 모두 제거하여 상기 관심 영역 이미지를 편집하는 이미지 편집부를 더 포함하는 것인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템.Editing the region of interest image by removing all of the segmented images of the non-featured region without the feature from the warped divided images and at least one of the divided images of the non-featured region of the width and height adjacent to each other in the long axis direction. Microscopic slide scanning system for bone marrow reading further comprising an image editing unit.
  6. 제4 항 또는 제5 항에 있어서,The method of claim 4 or 5,
    상기 와핑된 관심 영역 이미지 또는 상기 편집된 관심 영역 이미지를 출력하는 관심 영역 이미지 출력부를 더 포함하는 것인 골수 판독을 위한 현미경용 슬라이드 스캐닝 시스템.And a region of interest image output unit for outputting the warped region of interest image or the edited region of interest image.
PCT/KR2019/018591 2018-12-31 2019-12-27 Slide scanning system for microscope for bone marrow reading WO2020141811A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2018-0173623 2018-12-31
KR1020180173623A KR102105489B1 (en) 2018-12-31 2018-12-31 Microscopy based slide scanning system for bone marrow interpretation

Publications (1)

Publication Number Publication Date
WO2020141811A1 true WO2020141811A1 (en) 2020-07-09

Family

ID=70455969

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2019/018591 WO2020141811A1 (en) 2018-12-31 2019-12-27 Slide scanning system for microscope for bone marrow reading

Country Status (2)

Country Link
KR (1) KR102105489B1 (en)
WO (1) WO2020141811A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023096125A1 (en) * 2021-11-25 2023-06-01 연세대학교 산학협력단 Method for confirming change in bone formation, and program and device therefor

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102288332B1 (en) * 2019-10-16 2021-08-10 (주)유아이엠디 Bone marrow imaging method using bone marrow interpretation
KR102613961B1 (en) * 2020-12-10 2023-12-14 (주)유아이엠디 Determination method for cell zone of slide sample image smeared with bone-marrow and high magnification imaging method of the same cell zone
KR102425195B1 (en) * 2021-04-07 2022-07-27 (주)유아이엠디 Detection method of optimal inspection area for smear slide
KR20230009718A (en) * 2021-07-09 2023-01-17 (주)유아이엠디 Blood-cell detecion and classification method based on designation of inspection area

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6674884B2 (en) * 1996-08-23 2004-01-06 Bacus Laboratories, Inc. Apparatus for remote control of a microscope
KR20090006295A (en) * 2007-07-11 2009-01-15 하숙태 Apparatus, system and method for examination of cell slide and readable medium thereof
KR20140031742A (en) * 2012-09-05 2014-03-13 삼성전기주식회사 Apparatus and method for feature extraction of image, and image processing system using the same
KR20170013855A (en) * 2014-05-30 2017-02-07 더 보드 오브 트러스티스 오브 더 리랜드 스탠포드 쥬니어 유니버시티 Methods and devices for imaging large intact tissue samples
KR20180091024A (en) * 2015-12-02 2018-08-14 클리어라이트 다이어그노스틱스 엘엘씨 Methods for preparing and analyzing tumor tissue samples for detection and monitoring of cancer

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4988057B1 (en) * 2011-10-11 2012-08-01 アキュートロジック株式会社 Omnifocal image generation method, omnifocal image generation device, omnifocal image generation program, subject height information acquisition method, subject height information acquisition device, and subject height information acquisition program

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6674884B2 (en) * 1996-08-23 2004-01-06 Bacus Laboratories, Inc. Apparatus for remote control of a microscope
KR20090006295A (en) * 2007-07-11 2009-01-15 하숙태 Apparatus, system and method for examination of cell slide and readable medium thereof
KR20140031742A (en) * 2012-09-05 2014-03-13 삼성전기주식회사 Apparatus and method for feature extraction of image, and image processing system using the same
KR20170013855A (en) * 2014-05-30 2017-02-07 더 보드 오브 트러스티스 오브 더 리랜드 스탠포드 쥬니어 유니버시티 Methods and devices for imaging large intact tissue samples
KR20180091024A (en) * 2015-12-02 2018-08-14 클리어라이트 다이어그노스틱스 엘엘씨 Methods for preparing and analyzing tumor tissue samples for detection and monitoring of cancer

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023096125A1 (en) * 2021-11-25 2023-06-01 연세대학교 산학협력단 Method for confirming change in bone formation, and program and device therefor

Also Published As

Publication number Publication date
KR102105489B1 (en) 2020-04-28

Similar Documents

Publication Publication Date Title
WO2020141811A1 (en) Slide scanning system for microscope for bone marrow reading
CN111275016B (en) Slide scanning image acquisition and analysis method and device
KR102237696B1 (en) A artificial intelligence computer aided diagnosis system for automated analysis of cervical cancer cell image and its control method
US5796861A (en) Mosaic construction, processing, and review of very large electronic micrograph composites
US11226280B2 (en) Automated slide assessments and tracking in digital microscopy
EP1300713A3 (en) Method and apparatus for automated image analysis of biological specimens
CA2901164C (en) Referencing in multi-acquisition slide imaging
US20020090120A1 (en) System and method for finding regions of interest for microscopic digital montage imaging
WO2020141812A1 (en) Region of interest labeling device for bone marrow interpretation, and region of interest detection system comprising same
CA2689950A1 (en) Color space transformations for use in identifying objects of interest in biological specimens
BRPI0610115A2 (en) methods for image analysis based on chromogen separation
CN102822671B (en) Image processing apparatus, image reading apparatus, image processing method and image processing program
JP2006209698A (en) Target tracking device, microscope system and target tracking program
US20090103798A1 (en) Image generation
CN106973258A (en) Pathological section information quick obtaining device
JP3758894B2 (en) Mammogram image diagnosis support device
US7653260B2 (en) System and method of registering field of view
EP0595506A2 (en) Automated detection of cancerous or precancerous tissue by measuring malignancy associated changes
US20240152692A1 (en) Information processing device, information processing method, information processing system, and conversion model
CN115541578A (en) High-throughput super-resolution rapid scanning and analyzing system for cervical cell pathological section
WO2022050109A1 (en) Image processing device, image processing method, and image processing system
WO2017065591A1 (en) System and method for automatically calculating effective radiation exposure dose
US8649576B2 (en) Arborization reconstruction
JP3338625B2 (en) Object identification device
CN115917594A (en) Entire slide annotation transfer using geometric features

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19907182

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19907182

Country of ref document: EP

Kind code of ref document: A1