Abstract
Microscopy is one of the highest-impact innovations in the history of science. The recent development of Super-Resolution Microscopy has further enable the unprecedented capacity to observe cellular behaviour at near molecular-scale. However, the way we setup microscopes is still based on educated guesses by researchers with some help from empirical criteria. Furthermore, the quantification and interpretation of images, generating models on cellular behaviour is decoupled of the imaging procedure itself. This project aims take the first steps to develop artificial intelligence via deep reinforcement learning and convolution neural networks, guiding microscopes to autonomously observe cells while mathematically describing behaviour.
References
1. Gustafsson, N., Culley, S., Ashdown, G., Owen, D. M., Pereira, P. M. & Henriques, R. Fast live-cell conventional fluorophore nanoscopy with ImageJ through super-resolution radial fluctuations. Nat. Commun. 7, 12471 (2016).
2. Culley, S., Albrecht, D., Jacobs, C., Pereira, P. M., Leterrier, C., Mercer, J. & Henriques, R. Quantitative mapping and minimization of super-resolution optical imaging artifacts. Nat. Methods (2018). doi:10.1038/nmeth.4605
3. Weigert, M. et al. Content-Aware Image Restoration: Pushing the Limits of Fluorescence Microscopy. bioRxiv 236463 (2017). doi:10.1101/236463