Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
APPARATUS FOR CLASSIFYING PATHOLOGICAL DIAGNOSIS OF MEDICAL IMAGE, AND PATHOLOGICAL DIAGNOSIS SYSTEM USING SAME
Document Type and Number:
WIPO Patent Application WO/2017/022882
Kind Code:
A1
Abstract:
The present invention relates to an apparatus for classifying a pathological diagnosis of a medical image and a pathological diagnosis system using the same. Provided are an apparatus for classifying a pathological diagnosis of a medical image and a pathological diagnosis system using the same, the apparatus comprising: a feature extracting unit for extracting feature data associated with an input image, using a feature extraction variable; a feature vector transforming unit for transforming the extracted feature data into a feature vector, using a vector transformation variable; and a vector classifying unit for outputting a classification result of a pathological diagnosis associated with the input image by classifying the feature vector using a classification variable, wherein the feature extracting unit, the feature vector transforming unit, and the vector classifying unit are learned through: a first tagged image having a first tag indicating whether a lesion exists, and pathological diagnosis information; a second tagged image having the first tag and a second tag indicating location information of a lesion; and an image having no tag information.

Inventors:
KIM HYO-EUN (KR)
HWANG SANG-HEUM (KR)
PAEK SEUNG-WOOK (KR)
LEE JUNG-IN (KR)
JANG MIN-HONG (KR)
YOO DONG-GEUN (KR)
PAENG KYUNG-HYUN (KR)
PARK SUNG-GYUN (KR)
Application Number:
PCT/KR2015/009447
Publication Date:
February 09, 2017
Filing Date:
September 08, 2015
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
LUNIT INC (KR)
International Classes:
G06F19/00; G06N20/10; G06V10/774
Foreign References:
KR20140018748A2014-02-13
KR20140042531A2014-04-07
KR20140063288A2014-05-27
KR20130012297A2013-02-04
KR20130079863A2013-07-11
Attorney, Agent or Firm:
MI PATENT & LAW FIRM (KR)
Download PDF: