US 12,450,315 B2
Method and system for performing prediction work on target image
In Wan Yoo, Seoul (KR)
Assigned to LUNIT INC., Seoul (KR)
Filed by LUNIT INC., Seoul (KR)
Filed on Feb. 15, 2022, as Appl. No. 17/671,936.
Application 17/671,936 is a continuation of application No. PCT/KR2021/015734, filed on Nov. 2, 2021.
Claims priority of application No. 10-2020-0144411 (KR), filed on Nov. 2, 2020.
Prior Publication US 2022/0172009 A1, Jun. 2, 2022
Int. Cl. G06F 18/2415 (2023.01); G06F 18/2431 (2023.01); G06T 7/11 (2017.01)
CPC G06F 18/2415 (2023.01) [G06F 18/2431 (2023.01); G06T 7/11 (2017.01); G06T 2207/20021 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A method for performing a prediction work on a whole slide image (WSI), the method comprising:
dividing the WSI into a plurality of sub-images such that a portion of a first image included in the plurality of sub-images overlaps with a portion of a second image adjacent to the first image, wherein the WSI corresponds to a pathology slide generated by staining human tissue obtained from a human body;
separately generating individual prediction results for individual pixels among a plurality of pixels included in each of the plurality of divided sub-images;
separately applying individual weights to the individual prediction results for the plurality individual pixels such that weights applied to first prediction results for first pixels in the plurality of pixels are larger than weights applied to second prediction results for second pixels in the plurality of pixels, wherein distances of the first pixels to a center in each of the plurality of divided sub-images are closer than distances of the second pixels to the center;
separately merging the individual prediction results applied with the individual weights for the individual pixels that are included in an overlapping portion of the first image and the second image; and
separately generating final prediction results for the individual pixels based on the merged prediction results for the individual pixels and final weight values for the individual pixels, wherein a final weight value for an individual pixel corresponds to a sum of weights applied to the individual pixel for each of the sub-images.