Martin Stumpe, Chief of AI at Tempus, shared on LinkedIn:

“We recently published a paper about predicting Microsatellite Instability (MSI) in prostate cancer from hematoxylin and eosin stained biopsy images.

This manuscript is the outcome of amazing work by a lot of people – huge congratulations to Kunal Nagpal, Isabelle (Qiyuan) Hu, Abbas Rizvi, Geoffrey Schau, Kshitij Ingale, Yoni Muller, Rachel Baits, Sebastian Pretzer, Aïcha Bentaieb, Abigail Gordhamer, Roberto Nussenzveig, Adam Cole, Matthew O. Leavitt, MD, Ryan Jones, Rohan Joshi, Nike Beaubier!

MSI is an important biomarker for immunotherapy, but due to its low prevalence in prostate cancer, molecular testing for MSI is not standard of care. Predicting MSI from routinely generated hematoxylin and eosin stained biopsy images could therefore facilitate the identification of patients who might benefit from immunotherapy.

However, as many of us are acutely aware, publishing a manuscript about an algorithm is only the first step on the long journey to real-world impact, especially in healthcare. That’s why it’s even more exciting that the machine learning model described in that paper is (an earlier version of) the model that powers Tempus’ recently launched first clinical digital pathology test.

This is a huge milestone and there are more people who contributed to making this happen beyond the paper. Special kudos here in particular to Sina Zomorrodian, Elizabeth Morency, Maggie Huston, Alain Silk, and Michael Carlson and team!
Learn more.”

Additional information.
Source: Martin Stumpe/LinkedIn