by University of Washington researchersdevelopedDeep learning software, Omnipose, can help solve the challenge of identifying a variety of tiny bacteria in microscope images.Research PapersPublished in the journal Nature Methods.Omnipose The source code is hosted on GitHub under a non-commercial license. The researchers found that Omnipose, trained on a large database of bacterial images, performed well in characterizing and quantifying myriad bacteria in mixed microbial cultures and eliminated some of the errors that its predecessors may have had. Furthermore, due to the differences in the optical properties of different bacteria, Omnipose excels in overcoming identification problems. Omnipose’s high performance on a variety of cell morphologies and patterns could potentially unlock information from previously inaccessible microscope images, which could be a game-changer for biological image analysis, the researchers say.
