Method of storing and retrieving digital pathology analysis results
Inventors
Bredno, Joerg • Lorsakul, Auranuch
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12716825-B2
Publication Date
2026-08-25
Expiration Date
Abstract
The present disclosure is directed, among other things, to automated systems and methods for analyzing, storing, and/or retrieving information associated with biological objects having irregular shapes. In some embodiments, the systems and methods partition an input image into a plurality of sub-regions based on localized colors, textures, and/or intensities in the input image, wherein each sub-region represents biologically meaningful data.
Core Innovation
The invention provides a method for analyzing an image of a biological sample stained with at least one stain. A portion of the image corresponds to a first characteristic of the at least one stain, and a portion of the image corresponds to a second characteristic of the at least one stain. The method derives a set of feature metrics from the image, where the set of feature metrics is derived based on cell nuclei depicted by the image.
The invention segments the image into a plurality of sub-regions, where each sub-region defines a region of interest within the image. At least one sub-region is a superpixel corresponding to pixels of the image associated with the first characteristic, and at least one sub-region is a superpixel corresponding to pixels of the image associated with the second characteristic. One or more representational objects are determined for each sub-region of the plurality of sub-regions.
For each sub-region, the invention associates coordinates for each representational object of the one or more representational objects determined for the respective sub-region with a subset of the feature metrics. The subset of the feature metrics is derived from a portion of the image that corresponds to the respective sub-region. The same core operations are implemented in a system with one or more processors and in a non-transitory computer-readable medium.
Claims Coverage
The document includes three independent claims: a method, a system, and a non-transitory computer-readable medium. Each independent claim covers the same core pipeline with an emphasis on cell-nuclei-based feature metrics, superpixel-based sub-region segmentation tied to first and second stain characteristics, per-sub-region representational objects, and association of object coordinates with subset feature metrics; therefore the claims present the same number of core inventive features across the different claim formats.
Stained image with first and second stain characteristics
Accessing an image of a biological sample stained with at least one stain, wherein a portion of the image corresponds to a first characteristic of the at least one stain and a portion of the image corresponds to a second characteristic of the at least one stain.
Cell-nuclei-based feature metrics
Deriving a set of feature metrics from the image, wherein the set of feature metrics is derived based on cell nuclei depicted by the image.
Superpixel sub-regions tied to stain characteristics
Segmenting the image into a plurality of sub-regions, wherein each sub-region of the plurality of sub-regions defines a region of interest within the image, wherein at least one sub-region of the plurality of sub-regions is a superpixel corresponding to pixels of the image associated with the first characteristic, and wherein at least one sub-region of the plurality of sub-regions is a superpixel corresponding to pixels of the image associated with the second characteristic.
Representational objects per sub-region
Determining one or more representational objects for each sub-region of the plurality of sub-regions.
Coordinates linked to subset feature metrics
Associating, for each sub-region of the plurality of sub-regions, coordinates for each representational object of the one or more representational objects determined for the respective sub-region and a subset of feature metrics of the set of feature metrics, the subset of feature metrics derived from a portion of the image that corresponds to the respective sub-region.
Across the independent claims, the inventive scope is directed to a stained-image analysis pipeline that derives cell-nuclei-based feature metrics, segments the image into sub-regions including superpixels aligned with first and second stain characteristics, determines representational objects per sub-region, and associates per-object coordinates with subset feature metrics derived from the corresponding sub-region.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?