Deep learning-based algorithms outperform veterinary pathologists in finding the mitotically most active region

Authors: M. Aubreville, C. A. Bertram, C. Marzahl, C. Gurtner, M. Dettwiler, A. Schmidt, F. Bartenschlager, S. Merz, M. Fragoso, O. Kershaw, R. Klopfleisch & A. Maier

Part of most tumor grading schemes is finding the area of a whole slide image where most cell division (mitosis) is occurring. We assumed that this would be a really challenging task and investigated in this paper, how human experts performed against a selection of deep neural networks. Our results have been published in SpringerNature’s Scientific Reports:

https://www.nature.com/articles/s41598-020-73246-2