US 11,989,882 B2
Histological image analysis
Ole Johan Skrede, Oslo (NO); Tarjei Sveinsgjerd Hveem, Oslo (NO); John Robert Maddison, Crowborough (GB); Havard Emil Greger Danielsen, Oslo (NO); and Knut Liestol, Oslo (NO)
Assigned to OSLO UNIVERSITETSSYKEHUS, Oslo (NO)
Filed by OSLO UNIVERSITETSSYKEHUS, Oslo (NO)
Filed on May 13, 2022, as Appl. No. 17/743,642.
Application 17/743,642 is a continuation of application No. 16/763,860, granted, now 11,361,442, previously published as PCT/EP2018/080828, filed on Nov. 9, 2018.
Claims priority of application No. 1718970 (GB), filed on Nov. 16, 2017.
Prior Publication US 2022/0270258 A1, Aug. 25, 2022
Int. Cl. G06T 7/00 (2017.01)
CPC G06T 7/0014 (2013.01) [G06T 2207/10024 (2013.01); G06T 2207/10056 (2013.01); G06T 2207/20021 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30024 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented analysis method for histological image analysis, the method comprising:
obtaining, by the computer, a test histological image that has been stained with a marker;
dividing the test histological image into a plurality of tiles;
evaluating each of the plurality of tiles using a trained machine-learning algorithm to deliver a plurality of scores corresponding to the plurality of tiles;
comparing a representation of the plurality of scores with a threshold value; and
outputting a representation of the scores as a single predicted outcome value.