Interpretable deep learning of label-free live cell images uncovers functional hallmarks of highly-metastatic melanoma
A. Zaritsky, A. R. Jamieson, E. S. Welf, A. Nevarez, J. Cillay, U. Eskiocak, B. L. Cantarel, G. Danuser
This paper uses an adversarial auto-encoding deep learning model to classify metastatic efficiency in melanoma cells from label-free live cell images and identifies pseudopodial extensions and light scattering as functional hallmarks.
bioRxiv · May 15, 2020Cell Biology