Regulatory Focus™ > News Articles > 2021 > 9 > FDA OKs new pathology AI software, launches AI-enabled device database

FDA OKs new pathology AI software, launches AI-enabled device database

Posted 22 September 2021 | By Kari Oakes 

FDA OKs new pathology AI software, launches AI-enabled device database

The US Food and Drug Administration (FDA) has authorized software designed to help pathologists detect prostate cancer in digitally scanned slides from prostate biopsies. Separately, the agency announced Wednesday that it has made available a list of devices that use artificial intelligence and machine learning.
 
The newly authorized software, called Paige Prostate, is the first artificial intelligence (AI) tool okayed by FDA for in vitro diagnostic use for prostate cancer detection in biopsies. The AI pathology tool is “designed to identify an area of interest on the prostate biopsy image with the highest likelihood of harboring cancer so it can be reviewed further by the pathologist if the area of concern has not been identified on initial review,” according to FDA’s Tuesday press release announcing the authorization of the software through the de novo pathway.
 
Paige Prostate, a product of the firm Paige.AI, is used by pathologists on digitized slide images. Authorization of the software was based on a study involving 16 pathologists who looked at 527 digitized slide images from prostate biopsies. Of these, 171 had cancerous tissue and 356 were benign.
 
Each pathologist completed an unassisted and an assisted read of each slide image. Cancer detection in the slides that contained cancer cells was improved by 7.3% with the help of Paige Prostate compared with unassisted reads. Use of the AI tool had no impact on pathologists’ reads of the slide images of benign prostate tissue.
 
FDA emphasized that use of Paige Prostate carries a risk of both false negative and false positive results; this risk can be mitigated when a pathologist qualified to read prostate biopsy slides reads the slides and includes relevant clinical information in diagnostic decision-making. Pathologists may also choose to use other diagnostic techniques on biopsy samples before reaching a diagnostic decision.
 
Since prostate cancer is usually diagnosed on the basis of multiple biopsy sites and slides, FDA did not look at Paige Prostate’s impact on the final patient diagnosis.
 
“Pathologists examine biopsies of tissue suspected for diseases, such as prostate cancer, every day. Identifying areas of concern on the biopsy image can help pathologists make a diagnosis that informs the appropriate treatment,” said Tim Stenzel, director of the Office of In Vitro Diagnostics and Radiological Health in FDA’s Center for Devices and Radiological Health, in the press release.
 
“The authorization of this AI-based software can help increase the number of identified prostate biopsy samples with cancerous tissue, which can ultimately save lives,” Stenzel added.
 
With this authorization, FDA is also establishing special controls “related to labeling and performance testing” for devices of this new type; subsequent devices of this type may be cleared through the 510(k) premarket process, using Paige Prostate as the predicate device.
 
 
New AI/ML-enabled database
On Wednesday, FDA also announced it has launched a database of medical devices that utilize AI and machine learning (AI/ML).
 
In the announcement, FDA noted that the agency has seen an increase in recent years in submissions for AI/ML-enabled devices for use in many therapeutic areas and coming in through all of its submission pathways.
 
Describing the list of AI/ML-enabled devices as “a resource to the public about these devices and the FDA’s work in this area,” FDA laid out its methodology for building the list: first, it combed public-facing FDA information about device approvals and authorizations. Next, it turned to other publicly available databases from such sources as Nature, STAT, and the American College of Radiology, among others. Finally, the agency referenced “publicly available materials published by the specific manufacturers.”
 
The list is a public resource, but not meant to be “exhaustive or comprehensive,” clarified FDA. Currently, the list has 343 entries and is current to mid-June 2021.
 
Prostate cancer AI press release
 
AI/ML-enabled device database
 

 

© 2021 Regulatory Affairs Professionals Society.

Regulatory Focus newsletters

All the biggest regulatory news and happenings.

Subscribe