The Challenge is now open! You can register to access the training data and start developing your algorithm.
Project objectives
The first Light My Cells Challenge (2024) established a large-scale, multi-site benchmark for evaluating in silico fluorescence prediction from transmitted-light unlabeled microscopy images under realistic imaging variability.
In Light My Cells 2 challenge aims to advance in-silico labeling methods for biological microscopy. Participants must predict 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.
It introduces additional training data and a new prediction task to reconstruct the fluorescence maximum-intensity projection (MIP) from transmitted-light image stacks. It encourages participants to explore novel architectures, loss functions, and metadata-aware training strategies.
Focus on tasks
Task 1
The main task is to predict the best-focused output-images of four fluorescently labelled organelles from label-free 2D transmitted light input-images.
Task 2
The second task is to predict the MIP output-images of four fluorescently labelled organelles from a stack of 2D input-images of label-free transmitted light.
