US 11,893,731 B2 · Group sparsity model for image unmixing

Abstract

Systems and methods described herein relate, among other things, to unmixing more than three stains, while preserving the biological constraints of the biomarkers. Unlimited numbers of markers may be unmixed from a limited-channel image, such as an RGB image, without adding any mathematical complicity to the model. Known co-localization information of different biomarkers within the same tissue section enables defining fixed upper bounds for the number of stains at one pixel. A group sparsity model may be leveraged to explicitly model the fractions of stain contributions from the co-localized biomarkers into one group to yield a least squares solution within the group. A sparse solution may be obtained among the groups to ensure that only a small number of groups with a total number of stains being less than the upper bound are activated.

Claims17

Dependent claims

A method comprising:

accessing an image depicting a biological sample stained with a set of biomarkers;
accessing a reference color matrix, wherein the reference color matrix includes co-localization data for the set of biomarkers, wherein the co-localization data defines a plurality of groups of biomarkers within the set of biomarkers, each group of biomarkers of the plurality of groups of biomarkers including a subset of biomarkers of the set of biomarkers, wherein a first biomarker and a second biomarker of a subset of biomarkers of a first group of biomarkers of the plurality of groups of biomarkers represent a first common biological structure;
unmixing, based on the co-localization data of the reference color matrix, the image to generate a result that identifies, for one or more pixels of the image, a sparsity model that represents a first estimated fraction of contribution of the first biomarker and a second estimated fraction of contribution of the second biomarker;
and
outputting the result.

2 more elements

A system comprising one or more processors and at least one memory, the at least one memory storing instructions which, when executed by the one or more processors, cause the system to perform one or more operations comprising:

accessing an image depicting a biological sample stained with a set of biomarkers;
accessing a reference color matrix, wherein the reference color matrix includes co-localization data for the set of biomarkers, wherein the co-localization data defines a plurality of groups of biomarkers within the set of biomarkers, each group of biomarkers of the plurality of groups of biomarkers including a subset of biomarkers of the set of biomarkers, wherein a first biomarker and a second biomarker of a subset of biomarkers of a first group of biomarkers of the plurality of groups of biomarkers represent a first common biological structure;
unmixing, based on the co-localization data of the reference color matrix, the image to generate a result that identifies, for one or more pixels of the image, a sparsity model that represents a first estimated fraction of contribution of the first biomarker and a second estimated fraction of contribution of the second biomarker;
and
outputting the result.

2 more elements

A non-transitory computer-readable medium comprising instructions which, when executed by one or more processors of a computing system, cause the computing system to perform one or more operations comprising:

accessing an image depicting a biological sample stained with a set of biomarkers;
accessing a reference color matrix, wherein the reference color matrix includes co-localization data for the set of biomarkers, wherein the co-localization data defines a plurality of groups of biomarkers within the set of biomarkers, each group of biomarkers of the plurality of groups of biomarkers including a subset of biomarkers of the set of biomarkers, wherein a first biomarker and a second biomarker of a subset of biomarkers of a first group of biomarkers of the plurality of groups of biomarkers represent a first common biological structure;
unmixing, based on the co-localization data of the reference color matrix, the image to generate a result that identifies, for one or more pixels of the image, a sparsity model that represents a first estimated fraction of contribution of the first biomarker and a second estimated fraction of contribution of the second biomarker;
and
outputting the result.

2 more elements

Dependencies are inferred from the claim text.

Group sparsity model for image unmixing

Life of this patent

Group sparsity model for image unmixing

Life of this patent

Abstract

Claims17