Title:
DEEP LEARNING-BASED METHOD FOR HIGH-RESOLUTION CONVERSION OF PATHOLOGY SLIDE IMAGE, AND COMPUTING SYSTEM FOR PERFORMING SAME
Document Type and Number:
WIPO Patent Application WO/2022/103122
Kind Code:
A1
Abstract:
Disclosed are a method for converting resolution of a partial area of a pathology slide image into high resolution by using a high-resolution conversion technology based on super resolution using deep learning, and a computing system for performing same. An aspect of the present invention provides a high-resolution conversion method performed by a computing system including a super resolution neural network pre-learned for conversion of an input original image into a high-resolution image, the method comprising the steps of: specifying a high-resolution conversion area which corresponds to a partial area to be subject to high-resolution conversion, among all areas of an original pathology slide image; inputting an extended area to the super resolution neural network to acquire a high-resolution image corresponding to the high-resolution conversion area, wherein the extended area is obtained by extending the high-resolution conversion area upward, downward, leftward, and rightward by p pixels (here, p corresponds to an integer equal to or greater than 1); and outputting the acquired high-resolution image.
Inventors:
PAIK IN YOUNG (KR)
MUN UI GEO (KR)
JUNG CHAE YOON (KR)
KWAK TAE YEONG (KR)
KIM SUN WOO (KR)
MUN UI GEO (KR)
JUNG CHAE YOON (KR)
KWAK TAE YEONG (KR)
KIM SUN WOO (KR)
Application Number:
PCT/KR2021/016249
Publication Date:
May 19, 2022
Filing Date:
November 09, 2021
Export Citation:
Assignee:
DEEP BIO INC (KR)
International Classes:
G06T3/40; G06N3/08; G06T1/60
Foreign References:
KR20160068647A | 2016-06-15 | |||
KR20180066983A | 2018-06-20 | |||
JP2020141908A | 2020-09-10 | |||
KR102254755B1 | 2021-05-21 |
Other References:
ALBAWI SAAD; MOHAMMED TAREQ ABED; AL-ZAWI SAAD: "Understanding of a convolutional neural network", 2017 INTERNATIONAL CONFERENCE ON ENGINEERING AND TECHNOLOGY (ICET), IEEE, 21 August 2017 (2017-08-21), pages 1 - 6, XP033327277, DOI: 10.1109/ICEngTechnol.2017.8308186
Attorney, Agent or Firm:
SHIM, Choong Sup (KR)
Download PDF: