Quality metrics for automatic evaluation of dual ish images
AssigneeVentana Medical Systems, Inc.
InventorsAnindya Sarkar, Jim Martin, Jeffrey Atchison
Life of this patent
Estimating the term…
Abstract
The present disclosure is directed to a computer system designed to (i) receive a series of images as input; (ii) compute a number of metrics derived from focus features and color separation features within the images; and (iii) evaluate the metrics to return (a) an identification of the most suitable z-layer in a z-stack, given a series of z-layer images in a z-stack; and/or (b) an identification of those image tiles that are more suitable for cellular based scoring by a medical professional, given a series of image tiles from an area of interest of a whole slide scan.
Claims28
An image analysis method for determining one of a plurality of z-layers in a z-stack of digital images to be used for computing tissue-related or cell-related scores, the z-stack of digital images depicting different layers of a single tissue sample located on a slide, the tissue sample having been stained at least with a first stain having a first color and a second stain having a second color, the first and second colors differing from each other, the z-stack images having been acquired by varying the focus of an optical sensor, the method being implemented in an image analysis system and comprising:
- for each of the z-layers, computing focus metrics, each focus metric being an indicator of focus quality of the z-layer and being derived from image data of said z-layer;
- for each of the z-layers, computing color separation metrics, each color separation metric being an indicator of color separation quality of the z-layer and being derived from image data of said z-layer;
- and evaluating the focus metrics and the color separation metrics and determining the one of the z-layers to be used for computing tissue-related or cell-related scores in dependence on the focus metrics and color separation metrics of each of the z-layers.
A non-transitory computer-readable medium storing computer-executable instructions, which, when executed by one or more processors, cause the one or more processors to:
- compute focus metrics and color separation metrics for each z-layer within a z-stack of images, each z-layer within the z-stack of images corresponding to an image of a tissue sample, the focus metrics comprising a focus quality score for each z-layer, and the the color separation metrics comprising a color separation quality score for each z-layer, and wherein the focus quality score for each z-layer and the color separation quality score for each z-layer are independently computed within empirically determined color spaces optimized for signals of an in situ hybridization assay applied to the tissue sample, and
- evaluate the focus metrics and color separation metrics to determine a most suitable z-layer within the z-stack.
A system for the automated evaluation of image tiles derived from a whole slide scan, the imaging system comprising:
- (i) one or more processors, and (ii) a memory coupled to the processor, the memory to store computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
- (a) computing a plurality of focus features and a plurality of color separation features for each individual image tile;
- (b) deriving a focus quality score for each individual tile from the plurality of focus features and deriving a color separation quality score for each individual tile from the plurality of color separation features, wherein the focus quality scores and color separation quality score are computed only for those tiles having at least one dot corresponding to a first in situ hybridization signal and at least one dot corresponding to a second in situ hybridization signal;
- and
- (c) identifying digital image tiles more suitable for downstream processing based on the focus quality score and the color separation quality score.
Dependencies are inferred from the claim text.