SagivTech collaborates with Roche on the development of state of the art Deep Learning methods for Digital Pathology. Recently, we have published a paper together with researchers at Roche pRED on “Generalizing multistain immunohistochemistry tissue segmentation using one-shot color deconvolution deep neural networks“.
This paper is the result of a joint collaboration with Roche pRed, and Pathologists at Roche are now using this system.

In this paper we suggested a Deep Learning based solution for segmentation of whole slide images that generalizes over many different stain types (and not just the widely used H&E stain). The algorithm segments the image into “Background”, “Healthy Tissue”, “Tumor” and “Necrosis” parts. This task is a crucial task in drug development, prone to human error, and is extremely time consuming when done by humans.
We showed how using 1×1 convolutions allowed faster and better training on a dataset with considerable variability because of the stain types and demonstrated different visualization techniques for understanding the network and its outputs.

SagivTech also took part in Roche Digital Health innovation Lab. The main idea of the Digital Health Accelerator is to provide learning and exchange platform between startups and experts in this field of Digital Health. SagivTech was invited by Roche to join the Innovation Lab in order explore the use of Deep Learning for various problems in Digital Pathology.

More information can be found here.

This is how DeepREDai was born – a Digital Pathology viewer that offers Deep Learning modules to solve various problems such as Cell Detection, Staining normalization and Tissue Segmentation and Classification.
DeepREDai now collaborates with Roche and with the Cologne University Hospital in order to launch the most innovative viewer and Deep Learning capabilities in order to assist pathologists and enable better patient care.

Legal Disclaimer:
You understand that when using the Site you may be exposed to content from a variety of sources, and that SagivTech is not responsible for the accuracy, usefulness, safety or intellectual property rights of, or relating to, such content and that such content does not express SagivTech’s opinion or endorsement of any subject matter and should not be relied upon as such. SagivTech and its affiliates accept no responsibility for any consequences whatsoever arising from use of such content. You acknowledge that any use of the content is at your own risk.