Making digital tissue imaging better
Thursday, April 18, 2019 - 20:10
in Mathematics & Economics
A low-tech problem troubles the high-tech world of digital pathology imaging: There are no reliable standards for the quality of digitized tissue slides comprising the source material for computers reading and analyzing vast numbers of images. Poor-quality slides get mixed in with accurate slides, potentially confusing a computer program trying to learn what a cancerous cell looks like. Researchers are trying to fix this, sharing an open-source quality control standard.