Using a first stain to train a model to predict the region stained by a second stain

US11593656B2 · kind B2 · utility

Assignee

• AstraZeneca Computational Pathology GmbH · DE

Inventors

• Tobias Wiestler · Munich, DE
• Simon Lanzmich · Munich, DE
• Nicolas Brieu · Munich, DE
• Guenter Schmidt · Leun, DE
• Moritz Widmaier · Munich, DE

Key dates

• Filing date: Dec 14, 2018
• Grant date: Feb 28, 2023
• Expiry date: Oct 8, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06T2207/30024
• WIPO field: Measurement
• WIPO sector: Instruments

Abstract

A convolutional neural network predicts which regions of a tissue slice would be stained by a first stain by training a model to identify those regions based only on tissue stained by a second stain. Thereafter the first stain need not be used to mark cancerous regions on other tissue slices that are stained with the second stain. The training slice is stained with a first immunohistochemical stain and a second counterstain. A target region of an image of the training slice is identified using image analysis based on the first stain. A set of parameters for associated mathematical operations are optimized to train the model to classify pixels of the image as belonging to the target region based on the second stain but not on the first stain. The trained parameters are stored in a database and applied to other images of tissue not stained with the first stain.