Copyright Dorian Kauffmann CNRS France BioImaging
The ICAR team at LIRMM is involved in the release of Light My Cells, a large-scale open-access database described in the journal Scientific Data. This foundational resource aims to accelerate the development of artificial intelligence methods capable of predicting fluorescence images from transmitted-light images, paving the way for less invasive microscopy approaches.
Fluorescence microscopy is now an essential tool for observing cellular structures, but it relies on labeling methods that are often cumbersome. By offering an alternative based on “in silico labeling,” Light My Cells helps overcome these limitations by leveraging advanced machine learning techniques.
Developed as part of the France-BioImaging national infrastructure, this dataset comprises 2,574 acquisitions—representing nearly 57,000 2D images—from 30 studies conducted at 8 imaging centers. Thanks to the richness and diversity of the data it provides, it serves as a reference foundation for training, evaluating, and comparing new approaches in biomedical imaging.
At LIRMM, this work was led by Dorian Kauffmann, under the supervision of Emmanuel Faure and Guillaume Gay, within the ICAR team, in collaboration with a broad national consortium.
With Light My Cells, the LIRMM reaffirms its commitment to developing innovative methods at the intersection of artificial intelligence and biological imaging, with the goal of advancing our understanding of living organisms.
MORE DETAILS : https://www.ebi.ac.uk/biostudies/BioImages/studies/S-BIAD1047
CONTACT : Emmanuel FAURE
