Deep learning computer network excels at verification of breast cancer biopsy slides

Researchers have established a deep learning computer network that is highly precise and accurate in verifying whether invasive forms of breast cancer are present in whole biopsy slides.

A research team supervised by Case Western Reserve University published results of their research in Scientific Reports, detailing their deep learning computer network approach.

The research first involved training the network by downloading 400 biopsy images from several hospitals and then presenting the network with 200 images from The Cancer Genome Atlas and University Hospitals Cleveland Medical Center. Deep learning computer network excels at verification of breast cancer biopsy slides.

In accordance to Anant Madabushi, professor of biomedical engineering at Case Western Reserve and co-author of the study, the network scored 100% precision in determining the presence or absence of cancer on whole slides.

“This is a research with 600 patients, so it is fairly robust,” claims Madabushi, who also directs Case Western Reserve’s Center of Computational Imaging and Personalized Diagnostics. “And there were many human-machine comparisons done.”

In fact, compared with the analyses of 4 pathologists, the machine was more consistent and accurate, Madabushi asserts.

“Pathologists are highly busy, and we are talking about microscopic-level detail in these tissue slides. So, obviously, for them to go in and pick out every cell of cancer wasn’t tenable. There just was not enough time for them to be capable to sit down and manually do that,” adds Madabushi. “The network initiated to get more sophisticated, more granular and more accurate than the pathologists.”

Previous month, the Food and Drug Administration approved the marketing of the Philips IntelliSite Pathology Solution, the first whole slide imaging (WSI) system that enables review and interpretation of digital surgical pathology slides prepared from biopsied tissue. The system enables pathologists to read tissue slides digitally to make diagnoses, instead of looking straightly at a tissue sample mounted on a glass slide under a conventional light microscope.

In accordance to Madabushi, this is the first time the FDA has permitted the marketing of a WSI system for these purposes, which he says is a huge milestone for pathology. “A pathologist can look at an image of a slide on their computer monitor, and that is equivalent to the pathologist looking at a slide under their microscope,” he points out. “That means digital pathology—the digitization of slides—can now be utilized for primary diagnosis by a pathologist. That is a game changer.”

He considers that as pathologists increasingly adopt digital pathology there will be “an even greater need for software and analytics like the one we released in this paper.” Finally, Madabushi emphasizes that the FDA’s clearance of the Philips system “opens the door to an entire market for the analysis of digital pathology slide images.”