Democratising deep learning for microscopy with ZeroCostDL4Mic

Deep Learning (DL) methods are powerful analytical tools for microscopy and can outperform conventional image processing pipelines. Despite the enthusiasm and innovations fuelled by DL technology, the need to access powerful and compatible resources to train DL networks leads to an accessibility bar...

Full description

Saved in:

Bibliographic Details
Published in:	Nature communications Vol. 12; no. 1; p. 2276
Main Authors:	von Chamier, Lucas, Laine, Romain F., Jukkala, Johanna, Spahn, Christoph, Krentzel, Daniel, Nehme, Elias, Lerche, Martina, Hernández-Pérez, Sara, Mattila, Pieta K., Karinou, Eleni, Holden, Séamus, Solak, Ahmet Can, Krull, Alexander, Buchholz, Tim-Oliver, Jones, Martin L., Royer, Loïc A., Leterrier, Christophe, Shechtman, Yoav, Jug, Florian, Heilemann, Mike, Jacquemet, Guillaume, Henriques, Ricardo
Format:	Journal Article
Language:	English
Published:	London Nature Publishing Group UK 15-04-2021 Nature Publishing Group Nature Portfolio
Subjects:	13/56 14/63 631/114/1305 631/114/1564 631/80/2373 Accessibility Animals Biological activity Cell Line, Tumor Cloud Computing Computer applications Datasets as Topic Deep Learning Humanities and Social Sciences Humans Image processing Image Processing, Computer-Assisted - methods Image segmentation Life Sciences Microscopy Microscopy - methods Model testing multidisciplinary Neurobiology Neurons and Cognition Noise reduction Object recognition Optimization Primary Cell Culture Rats Science Science (multidisciplinary) Software Image processing Cellular imaging Machine learning
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Deep Learning (DL) methods are powerful analytical tools for microscopy and can outperform conventional image processing pipelines. Despite the enthusiasm and innovations fuelled by DL technology, the need to access powerful and compatible resources to train DL networks leads to an accessibility barrier that novice users often find difficult to overcome. Here, we present ZeroCostDL4Mic, an entry-level platform simplifying DL access by leveraging the free, cloud-based computational resources of Google Colab. ZeroCostDL4Mic allows researchers with no coding expertise to train and apply key DL networks to perform tasks including segmentation (using U-Net and StarDist), object detection (using YOLOv2), denoising (using CARE and Noise2Void), super-resolution microscopy (using Deep-STORM), and image-to-image translation (using Label-free prediction - fnet, pix2pix and CycleGAN). Importantly, we provide suitable quantitative tools for each network to evaluate model performance, allowing model optimisation. We demonstrate the application of the platform to study multiple biological processes. Deep learning methods show great promise for the analysis of microscopy images but there is currently an accessibility barrier to many users. Here the authors report a convenient entry-level deep learning platform that can be used at no cost: ZeroCostDL4Mic.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	2041-1723 2041-1723
DOI:	10.1038/s41467-021-22518-0