US FDA Authorizes ‘Paige Prostate’, A Clinical-Grade AI Solution For Prostate Cancer Detection

The US Food and Drug Administration has approved an AI-based software to help medical specialists (pathologists) who analyze body tissues to find suspect spots for cancer as an adjunct (supplement) to reviewing digitally scanned slide images from prostate biopsies (tissue removed from the body). Paige Prostate, an AI-based software developed specifically for prostate biopsies is able to identify areas of concern with the highest likelihood and send them off for review by pathologists.

Prostate cancer is a kind of cancer that begins in the prostate gland. Aside from non-melanoma skin cancer, Prostate cancer is the most common type of cancer among men in the US, according to CDC. 

Paige Prostate can be used with slide images that have been scanned and digitized. A slide image viewer can then be used to view the digitized slide image.

The FDA analyzed data from a clinical investigation in which 16 pathologists looked at 527 slide pictures of prostate biopsies that were digitized using a scanner (171 cancer and 356 benign). For each slide image, each pathologist conducted two assessments: one without Paige Prostate’s help (unassisted read) and one with Paige Prostate’s help (aided read) (assisted read). Paige Prostate enhanced cancer detection on individual slide photos by 7.3 percent on average when compared to pathologists’ unassisted reads for the complete slide images of individual biopsies, with no effect on the read of benign slide images, according to the clinical research.

The use of this device as an adjunct and professional evaluation by a qualified pathologist who has considered patient history, among other relevant clinical information will reduce the risk for false negative or false positive results. The FDA reviewed the device under the De Novo premarket review process, which is a new regulatory pathway for low-to-moderate-risk devices.


Other Source:

Press Release:

🐝 Join the Fastest Growing AI Research Newsletter Read by Researchers from Google + NVIDIA + Meta + Stanford + MIT + Microsoft and many others...