RT Journal Article
SR Electronic
T1 A deep learning framework for nucleus segmentation using image style transfer
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 580605
DO 10.1101/580605
A1 Reka Hollandi
A1 Abel Szkalisity
A1 Timea Toth
A1 Ervin Tasnadi
A1 Csaba Molnar
A1 Botond Mathe
A1 Istvan Grexa
A1 Jozsef Molnar
A1 Arpad Balind
A1 Mate Gorbe
A1 Maria Kovacs
A1 Ede Migh
A1 Allen Goodman
A1 Tamas Balassa
A1 Krisztian Koos
A1 Wenyu Wang
A1 Norbert Bara
A1 Ferenc Kovacs
A1 Lassi Paavolainen
A1 Tivadar Danka
A1 Andras Kriston
A1 Anne E. Carpenter
A1 Kevin Smith
A1 Peter Horvath
YR 2019
UL http://biorxiv.org/content/early/2019/03/17/580605.abstract
AB Single cell segmentation is typically one of the first and most crucial tasks of image-based cellular analysis. We present a deep learning approach aiming towards a truly general method for localizing nuclei across a diverse range of assays and light microscopy modalities. We outperform the 739 methods submitted to the 2018 Data Science Bowl on images representing a variety of realistic conditions, some of which were not represented in the training data. The key to our approach is to adapt our model to unseen and unlabeled data using image style transfer to generate augmented training samples. This allows the model to recognize nuclei in new and different experiments without requiring expert annotations.