The Rise of Data-Driven Microscopy powered by Machine Learning

Authors: Leonor Morgado, Estibaliz Gómez-de-Mariscal, Hannah S Heil, Ricardo Henriques
Technologies: BioImage Model Zoo, DL4MicEverywhere, NanoJ, NanoJ-Fluidics and ZeroCostDL4Mic
Preprint published in arXiv, January 2024

The manuscript by Morgado et al. explores the integration of machine learning techniques into microscopy for enhanced data-driven analysis. Data-driven microscopy uses real-time data analysis to optimize imaging conditions and extract meaningful information through automated feedback loops. Machine learning algorithms, such as classification, segmentation, tracking, and reconstruction, are employed to perform tasks without explicit programming. The manuscript provides examples of applications in event-driven approaches and super-resolution microscopy, demonstrating the benefits of data-driven microscopy in balancing population-level behavioral monitoring and targeted high-resolution data collection.