US 12,333,780 B2
Migration system of learning model for cell image analysis and migration method of learning model for cell image analysis
Hiroaki Tsushima, Kyoto (JP); Ryuji Sawada, Kyoto (JP); Takeshi Ono, Kyoto (JP); and Shuhei Yamamoto, Kyoto (JP)
Assigned to SHIMADZU CORPORATION, Kyoto (JP)
Filed by SHIMADZU CORPORATION, Kyoto (JP)
Filed on Oct. 4, 2022, as Appl. No. 17/959,800.
Claims priority of application No. 2021-166341 (JP), filed on Oct. 8, 2021.
Prior Publication US 2023/0111880 A1, Apr. 13, 2023
Int. Cl. G06K 9/00 (2022.01); G06T 7/00 (2017.01); G06V 10/72 (2022.01); G06F 21/60 (2013.01)
CPC G06V 10/72 (2022.01) [G06T 7/0012 (2013.01); G06F 21/602 (2013.01); G06T 2207/10056 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30024 (2013.01)] 11 Claims
OG exemplary drawing
 
1. A migration system of a learning model for cell image analysis that migrates a learning model used for analyzing a cell image from a first learning device to a second learning device, wherein
the second learning device includes
a migration information input reception unit that receives an input of learning model migration information including first algorithm specification information for specifying a first algorithm of the learning model used for analyzing the cell image and a first parameter, which is output when the learning model is generated and is a learning parameter when the cell image is analyzed by using the learning model,
a second learning device storage unit that stores a second algorithm of the learning model used for analyzing the cell image,
an algorithm consistency determination unit that determines, based on second algorithm specification information for specifying the second algorithm stored in the second learning device storage unit and the first algorithm specification information, whether or not consistency is established for estimation results, which are obtained when the first parameter is used, between the first algorithm and the second algorithm,
a notification unit that makes a notification of whether or not the consistency is established between the first algorithm and the second algorithm, and
a learning model parameter setting unit that sets, when the consistency is established between the first algorithm and the second algorithm, the first parameter to be used together with the second algorithm.