Covering Scientific & Technical AI | Wednesday, December 11, 2024

Paige, AI for Pathology Startup, Attracts New Funding 

via Shutterstock

Medical applications for AI and machine learning have so far yielded mixed results. While proving useful in areas like medical imaging, where tools can help reduce confirmation bias, the technology has fallen short in other applications like drug discovery.

“Clinical-grade” computational pathology, an emerging approach that combines multiple data sources and models to generate diagnostic inferences, is attracting investors who view the framework as a way of developing new pathology workflows, especially for treating cancer patients.

Among the early entrants is Paige.ai, which this week completed a $70 million funding round that included $15 million in additional investment from Goldman Sachs Merchant Banking Division. The unit has so far invested $20 million in Paige.

Founded in 2017 by Thomas Fuchs and colleagues at New York Memorial Sloan Kettering Cancer Center, the startups’ name is an acronym for Pathology-AI-Guidance-Engine. Paige said its algorithms have been trained using expert diagnoses along with thousands of diagnostic images.

The resulting AI suite is designed as a “clinical decision support system” running on a GPU-backed engine with more than 10 petaflops of capacity. That infrastructure is aimed at computational pathology applications that are time-consuming.

Paige and its investors are betting the field of computational pathology is ripe for what the startup calls “augmented clinical analysis” that combines proprietary data sets with machine learning. The startup claims to be the first company to receive the U.S. Food & Drug Administration’s “breakthrough therapy” designation for an AI tool used to diagnose cancers.

The Memorial Sloan Kettering spin-out has so far raised more than $95 million in a pair of funding rounds. Other investors include Healthcare Venture Partners.
Paige said it would use its new venture funding to advance its pathology workflows while working with bioscience and pharmaceutical firms “to create custom diagnostic and clinical trial solutions….”

Last November, Paige announced prostate cancer detection tool along with an AI-based “pathology viewer.” The tool dubbed CE Mark addresses increased case volumes and a growing shortage of qualified pathologists.

The company noted that prostate cancer is among the most difficult and time-consuming biopsies for pathologists to review. The Paige framework is designed to streamline diagnostic workflows to improve detection.

 

About the author: George Leopold

George Leopold has written about science and technology for more than 30 years, focusing on electronics and aerospace technology. He previously served as executive editor of Electronic Engineering Times. Leopold is the author of "Calculated Risk: The Supersonic Life and Times of Gus Grissom" (Purdue University Press, 2016).

AIwire