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 (DL) methods are increasingly recognised as powerful analytical tools for microscopy. Their potential to out-perform conventional image processing pipelines is now well established (1, 2). Despite the enthusiasm and innovations fuelled by DL technology, the need to access powerful and compatible resources, install multiple computational tools and modify code instructions to train neural networks all lead to an accessibility barrier that novice users often find difficult to cross. Here, we present ZeroCostDL4Mic, an entry-level teaching and deployment DL platform which considerably simplifies access and use of DL for microscopy. This is achieved by exploiting computational resources provided by Google Colab, a free, cloud-based service accessible through a web browser. ZeroCostDL4Mic allows researchers with little or no coding expertise to quickly train, assess and use popular DL networks. In parallel, it guides researchers to improve their understanding and to experiment with optimising DL parameters and network architectures. Altogether, ZeroCostDL4Mic accelerates the uptake of DL for new users and promotes their capacity to use increasingly complex DL networks.

No abstract