Preprocessing
Overview
Preprocessing is an essential step to prepare your microscopy data for analysis in Celldetective. It includes both off-software and in-software methods to align, correct, and optimize your stacks for segmentation and downstream measurements.
Registration
Stage drift and shaking make still objects move across the frames, which misleads tracking and blurs time-averaged measurements. We highly recommend registering the movies before segmentation. Celldetective registers stacks by phase cross-correlation, directly in the Preprocessing module (see the table below). The stacks can also be aligned before being imported, with external tools like Fiji (ImageJ), which can correct rotations too.
See also
How to register stacks to register the stacks in Celldetective, and How to register stacks with Fiji for a step-by-step guide on using the customized Fiji macro for batch registration.
In-software preprocessing
The Preprocessing module lets you batch-correct stacks directly within Celldetective. Corrected stacks are saved with the prefix Corrected_. The corrections added to the list are applied in order: the first one reads the raw movie, the next ones correct the Corrected_ stack in place. Each step is recorded with its parameters in the log_preprocessing.txt file of the position.
Task |
Description |
Guide |
|---|---|---|
Background correction (Model Fit) |
Fits a 2D surface to the background, excluding cells. Best for fluorescence images. |
|
Background correction (Model Free) |
Computes a median background from multiple positions or timeframes. Best for brightfield/reflection microscopy. |
|
Channel offset correction |
Aligns channels that have pixel shifts between modalities. |
|
Stack registration |
Corrects the drift of the field over time, estimated on one channel and applied to all. |
For a full list of parameters (thresholds, tile options, etc.), see the Preprocessing Protocols Reference.
Bibliography
Schindelin, J., Arganda-Carreras, I., Frise, E. et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 9, 676–682 (2012). https://doi.org/10.1038/nmeth.2019