US 12,307,655 B2
Cell image analysis method and cell image analysis device
Ayako Akazawa, Kyoto (JP)
Assigned to SHIMADZU CORPORATION, Kyoto (JP)
Filed by Shimadzu Corporation, Kyoto (JP)
Filed on Sep. 28, 2020, as Appl. No. 17/035,042.
Claims priority of application No. JP2019-187865 (JP), filed on Oct. 11, 2019.
Prior Publication US 2021/0110536 A1, Apr. 15, 2021
Int. Cl. G06T 7/00 (2017.01); G01N 15/14 (2006.01); G01N 15/1429 (2024.01); G01N 15/1433 (2024.01); G06T 7/11 (2017.01); G06T 7/136 (2017.01); G01N 15/01 (2024.01); G01N 15/10 (2006.01); G01N 15/1434 (2024.01)
CPC G06T 7/0012 (2013.01) [G01N 15/1429 (2013.01); G01N 15/1433 (2024.01); G01N 15/1468 (2013.01); G06T 7/11 (2017.01); G06T 7/136 (2017.01); G01N 15/01 (2024.01); G01N 2015/1006 (2013.01); G01N 2015/1454 (2013.01); G01N 2015/1477 (2013.01); G06T 2207/10056 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30024 (2013.01); G06T 2207/30104 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A cell image analysis method comprising:
obtaining a first phase image of a first cell having a cell organelle based on hologram data acquired by holographic microscopy;
staining the first cell;
obtaining a stained cell organelle image of the first cell acquired by microscopy;
generating a first learned model using training data including the first phase image as an input image of machine learning and the stained cell organelle image as a ground truth image; and
inputting a second phase image of a second cell into the first learned model to obtain an estimated cell area image,
wherein the first phase image and the second phase image are not stained.