Paige gets CE mark for prostate cancer detection technology

Dive Brief: Paige received CE marks for its digital pathology system and an associated technology […]

Dive Brief:

  • Paige received CE marks for its digital pathology system and an associated technology for detecting prostate cancer, the company said Tuesday.
  • The European regulatory clearances position Paige to pitch software products designed to make it easier for pathologists to analyze biopsies for signs of cancer.
  • Paige is built on technology from Memorial Sloan Kettering Cancer Center, which has shown the approach could enable pathologists to safely exclude 75% of slides from manual review.

Dive Insight:

Digital technology has arrived relatively late in pathology. Philips won FDA clearance to sell digital pathology technology in the U.S. in 2017, putting the field well behind disciplines such as radiology. Digital pathology arrived earlier in Europe but there are signs the promised efficiency gains are yet to take effect. In the U.K., a survey found 97% of histopathology departments are understaffed.
Paige wants to ease the pressures on pathology departments by enabling them to get through more work without increasing their headcount. Artificial intelligence is key to Paige’s attempt to achieve that goal.
The CE marks cover digital pathology viewer Paige Insight and AI-based module Paige Prostate. Paige Insight enables pathologists to review and collaborate on digitized slides. By pairing that technology with Paige Prostate, Paige wants to help pathologists zero in on slides from cancer patients, thereby increasing throughput and diagnostic quality.
Paige co-founder Thomas Fuchs and his colleagues at MSK published a paper on the AI capabilities in Nature Medicine earlier this year. The study found the application of AI to biopsy slides would allow pathologists to automatically exclude more than 75% of slides without overlooking cases of prostate cancer.
“For pathologists who must operate in the increasingly complex, detailed and data-driven environment of cancer diagnostics, tools such as this will allow non-subspecialized pathologists to confidently and efficiently classify cancer with 100% sensitivity,” the authors wrote.
The AI was trained and originally tested on slides from MSK. However, the researchers also tested the AI on 12,000 prostate cancer slides captured at institutions other than MSK, enabling them to show whether the technology can work outside of the context in which it was trained. Diagnostic accuracy fell somewhat in the broader test set but the researchers found sensitivity remained high.
Paige’s affiliation with MSK recently came under a spotlight, with a report by The New York Times and ProPublica highlighting concerns pathologists and experts in nonprofit law and corporate governance have with the relationship between the two entities.

Original Article: (