Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020141812 - REGION OF INTEREST LABELING DEVICE FOR BONE MARROW INTERPRETATION, AND REGION OF INTEREST DETECTION SYSTEM COMPRISING SAME

Publication Number WO/2020/141812
Publication Date 09.07.2020
International Application No. PCT/KR2019/018592
International Filing Date 27.12.2019
IPC
G06T 7/187 2017.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
187involving region growing; involving region merging; involving connected component labelling
G06T 7/136 2017.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
136involving thresholding
G06T 3/40 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
3Geometric image transformation in the plane of the image
40Scaling of a whole image or part thereof
A61B 5/00 2006.01
AHUMAN NECESSITIES
61MEDICAL OR VETERINARY SCIENCE; HYGIENE
BDIAGNOSIS; SURGERY; IDENTIFICATION
5Measuring for diagnostic purposes; Identification of persons
CPC
A61B 5/00
AHUMAN NECESSITIES
61MEDICAL OR VETERINARY SCIENCE; HYGIENE
BDIAGNOSIS; SURGERY; IDENTIFICATION
5Measuring for diagnostic purposes
A61B 5/0059
AHUMAN NECESSITIES
61MEDICAL OR VETERINARY SCIENCE; HYGIENE
BDIAGNOSIS; SURGERY; IDENTIFICATION
5Measuring for diagnostic purposes
0059using light, e.g. diagnosis by transillumination, diascopy, fluorescence
G06T 2207/30024
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
30Subject of image; Context of image processing
30004Biomedical image processing
30024Cell structures in vitro; Tissue sections in vitro
G06T 3/40
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
3Geometric image transformation in the plane of the image
40Scaling the whole image or part thereof
G06T 7/136
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
136involving thresholding
G06T 7/187
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
187involving region growing; involving region merging; involving connected component labelling
Applicants
  • (주)유아이엠디 UIMD INC. [KR]/[KR]
Inventors
  • 최종호 CHOI, Jong Ho
  • 이영득 LEE, Young Deuk
Agents
  • 신성기 SHIN, Sungki
Priority Data
10-2018-017362731.12.2018KR
Publication Language Korean (KO)
Filing Language Korean (KO)
Designated States
Title
(EN) REGION OF INTEREST LABELING DEVICE FOR BONE MARROW INTERPRETATION, AND REGION OF INTEREST DETECTION SYSTEM COMPRISING SAME
(FR) DISPOSITIF DE MARQUAGE DE RÉGION D'INTÉRÊT POUR ANALYSE DE MOELLE OSSEUSE ET SYSTÈME DE DÉTECTION DE RÉGION D'INTÉRÊT COMPRENANT CE DERNIER
(KO) 골수 판독을 위한 관심영역 레이블링 장치 및 이를 포함하는 관심영역 검출 시스템
Abstract
(EN)
According to an embodiment of the present invention, a region of interest may be automatically labelled and detected for bone marrow interpretation. To this end, an embodiment of the present invention may comprise a region of interest labeling device for bone marrow interpretation, the region of interest labeling device comprising: an image scaling unit which downscales a stitching image obtained by stitching multiple divided images obtained through motion control imaging of a bone marrow specimen slide; a color conversion unit which converts the RGB color type of the scaled stitching image into an image-processing color space; a segmentation unit which segments, by applying multi-thresholding including high thresholding and low thresholding, the stitching image that has been converted into the image-processing color space; a morphology conversion unit which enhances a cell nucleus image by performing a morphology opening operation on the stitching image if the stitching image is segmented by applying the high thresholding, and enhances a cell continuity image by performing a morphology dilation operation on the stitching image if the stitching image is segmented by applying the low thresholding; a labeling unit which detects, on the basis of the enhanced cell nucleus image, multiple reliable cells to label the detected reliable cells, and detects, on the basis of the enhanced cell continuity image, particles or abnormal cells to label the detected particles or abnormal cells; and an occlusion cell filtering unit which removes, from the multiple labeled reliable cells, one or more occlusion cells occluded by the labeled particles or the labeled abnormal cells.
(FR)
Selon un mode de réalisation de la présente invention, une région d'intérêt peut être automatiquement marquée et détectée pour une analyse de moelle osseuse. À cet effet, un mode de réalisation de la présente invention peut comprendre un dispositif de marquage de région d'intérêt pour l'analyse de moelle osseuse, le dispositif de marquage de région d'intérêt comprenant : une unité de mise à l'échelle d'image qui réduit une image de jointure obtenue par la jointure de multiples images divisées obtenues par le biais d'une imagerie de commande de mouvement d'une lame porte-échantillon de moelle osseuse ; une unité de conversion de couleur qui convertit le type de couleur RVB de l'image de jointure mise à l'échelle en un espace de couleur de traitement d'image ; une unité de segmentation qui segmente, par l'application d'un seuillage multiple comprenant un seuillage haut et un seuillage bas, l'image de jointure qui a été convertie en l'espace de couleur de traitement d'image ; une unité de conversion de morphologie qui améliore une image de noyau cellulaire par la réalisation d'une opération d'ouverture de morphologie sur l'image de jointure si l'image de jointure est segmentée par l'application du seuillage haut et améliore une image de continuité cellulaire par la réalisation d'une opération de dilatation de morphologie sur l'image de jointure si l'image de jointure est segmentée par l'application du seuillage bas ; une unité de marquage qui détecte, sur la base de l'image de noyau cellulaire améliorée, de multiples cellules fiables pour marquer les cellules fiables détectées et détecte, sur la base de l'image de continuité cellulaire améliorée, des particules ou des cellules anormales pour marquer les particules ou les cellules anormales détectées ; et une unité de filtrage de cellules d'occlusion qui élimine, à partir des multiples cellules fiables étiquetées, une ou plusieurs cellules d'occlusion occluses par les particules marquées ou les cellules anormales marquées.
(KO)
본 발명의 일 실시예에 의하면, 골수 판독을 위해 관심영역을 자동으로 레이블링하고 검출할 수 있다. 이를 위해 본 발명의 일 실시예는, 골수 표본 슬라이드의 모션 제어 촬상으로 획득된 다수의 분할 이미지를 스티칭한 스티칭 이미지를 다운 스케일링하는 이미지 스케일링부; 스케일링된 스티칭 이미지의 RGB 컬러를 영상처리용 색공간으로 변환하는 컬러 변환부; 영상처리용 색공간으로 변환된 스티칭 이미지를 하이 쓰레스홀딩과 로우 쓰레스홀딩을 포함하는 다중 쓰레스홀딩을 적용하여 구분화하는 세그먼테이션부; 하이 쓰레스홀딩을 적용하여 구분화된 스티칭 이미지에 대하여 모폴로지 오프닝(morphology opening) 연산으로 세포 핵 이미지를 강화하고, 로우 쓰레스홀딩을 적용하여 구분화된 스티칭 이미지에 대하여 모폴로지 딜레이션(morphology dilation) 연산으로 세포 연속성 이미지를 강화하는 모폴로지 변환부; 강화된 세포 핵 이미지에 기반하여 다수의 신뢰성 있는 세포를 검출하여 레이블링하고, 강화된 세포 연속성 이미지에 기반하여 파티클 또는 비정상적 세포를 검출하여 레이블링하는 레이블링부; 및 레이블링된 다수의 신뢰성 있는 세포에서 레이블링된 파티클 또는 레이블링된 비정상적 세포와 폐색(occlusion)되는 1 이상의 폐색 세포를 제거하는 폐색 세포 필터링부를 포함하는 골수 판독을 위한 관심영역 레이블링 장치를 포함할 수 있다.
Also published as
Latest bibliographic data on file with the International Bureau