US 12,249,073 B2
Systems and methods for mesothelioma feature detection and enhanced prognosis or response to treatment
Gilles Wainrib, Pantin (FR); Thomas Clozel, New York, NY (US); Pierre Courtiol, Paris (FR); Charles Maussion, Paris (FR); Jean-Yves Blay, Lyons (FR); and Françoise Galateau Salle, Caen (FR)
Assigned to OWKIN, INC., New York, NY (US); OWKIN FRANCE SAS, Paris (FR); and CENTRE LÉON BÉRARD, Lyons (FR)
Filed by Owkin Inc., New York, NY (US); Owkin France SAS, Paris (FR); and Centre Léon Bérard, Lyons (FR)
Filed on Dec. 30, 2022, as Appl. No. 18/091,988.
Application 18/091,988 is a continuation of application No. 17/185,924, filed on Feb. 25, 2021, granted, now 11,544,851.
Application 17/185,924 is a continuation of application No. PCT/IB2020/056030, filed on Jun. 25, 2020.
Claims priority of application No. 19305839 (EP), filed on Jun. 25, 2019.
Prior Publication US 2023/0306598 A1, Sep. 28, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/00 (2017.01); G06N 3/08 (2023.01); G06T 7/11 (2017.01); G06T 7/194 (2017.01); G16H 30/20 (2018.01)
CPC G06T 7/0014 (2013.01) [G06N 3/08 (2013.01); G06T 7/11 (2017.01); G06T 7/194 (2017.01); G16H 30/20 (2018.01); 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)] 21 Claims
OG exemplary drawing
 
1. A machine readable medium for determining the prognosis of a subject known or suspected to have mesothelioma having executable instructions to cause one or more processing units to:
determine a set of discriminative features in a biopsy image using a processing of extracting a plurality of feature vectors from the biopsy image by applying a first convolutional neural network, wherein each of the features of the plurality of feature vectors represents local descriptors of the biopsy image and the discriminative features include a combination of low survival features and/or high survival features, the low survival features being mesothelioma features associated with a prognosis of low survival duration and the high survival features being mesothelioma features associated with a prognosis of high survival duration;
classify the biopsy image using at least the set of discriminative features and a classification model, wherein the classification model is trained using a training set of known mesothelioma images and each of the known mesothelioma images has a corresponding label with survival duration information; and
determine the prognosis of the subject based on at least the classification of the biopsy image, wherein the prognosis is an indication of a survival duration of the subject.