France BioImaging’s Light My Cells 2 Challenge! 

After the success of Light My Cells in 2024, France-BioImaging is launching Light My Cells 2, an international  machine learning competition for fluorescence prediction from transmitted light microscopy images, also known as in-silico labeling.

In the 2024 Light My Cells challenge,participants predicted fluorescence images of four cellular structures —nucleus, mitochondria, tubulin, and actin— from label-free transmitted-light microscopy images acquired using bright-field (BF), phase-contrast (PC), or differential interference contrast (DIC) modalities. The resulting 2D dataset has been published in Scientific Data (Nature Portfolio), providing an open benchmark for in-silico labeling.

The 2026 edition introduces: 

• additional training data 
• a new task to predict fluorescence maximum-intensity projection (MIPs) from transmitted-light image stacks. It encourages participants to explore novel architectures, loss functions, and metadata-aware training strategies.

How you can get involved

With Light My Cells 2, participants are invited to:

• Predict fluorescence images of four cellular structures (nucleus, mitochondria, tubulin and actin) from label-free transmitted-light microscopy images (BF, PC, DIC)
• Reconstruct fluorescence Maximum-Intensity Projections (MIPs) from transmitted-light image stacks

Key dates:

• September 21^st, 2026: validation phase opens — participants can test their algorithms and the submission process, with up to five test submissions.
• December 1^st–13^th, 2026: final evaluation phase — participants will have one final submission
• December 18^th, 2026: results announcement

Join the challenge 

Ready to take on the challenge? Registration is now open and the training database is already available.

Discover Light My Cells 2  here: https://lightmycells2.grand-challenge.org/lightmycells2/

Not a participant? Share the challenge with colleagues working in microscopy, computer vision, or machine learning!

Article contributed by France-BioImaging