acia#
Automated single-cell image analysis for 2D+t live-cell imaging.
acia turns a time-lapse microscopy file into quantitative single-cell results.
It gives you one API for reading ND2, CZI, OME-TIFF and folders of TIFFs; lazy,
numpy-style slicing over the (T, H, W, C) axes; physical units that travel with
the data from load to result; eight state-of-the-art segmentation and tracking
backends; and visualization from annotated videos to lineage trees.
from acia.segm.open import open_sequence
src = open_sequence("experiment.nd2").position(0)
src = src[::20, 256:768, 256:768] # every 20th frame, a 512x512 crop
src # interactive viewer, right in Jupyter
Although it is built with microfluidic live-cell imaging in mind, nothing in the library assumes cells — it works for any objects you can detect in images.
New here? Five short tutorials take you from opening your first file to a growth-rate curve. Every one runs in your browser on Colab — no install.
Task-shaped recipes: reading from SMB/S3/OMERO, slicing and calibration, units in the extracted tables, and scaling a notebook over hundreds of sequences.
Every module, class and function, generated from the source.
Instance or Contour? Tracklet graph or tracking graph? The vocabulary,
defined once.
Installation#
pip install acia
Optional features — ND2 and CZI readers, OMERO, remote shares, the interactive widgets, and the segmentation backends — live behind extras. See Installation, which also explains why the segmentation backends are mutually exclusive.
Applied examples#
For complete, published analyses built on acia — growth-rate quantification,
fluorescence co-culture characterization, single-cell response to oxygen
alternation, and scaling those across hundreds of sequences — see the companion
acia-workflows
collection. The tutorials here teach the library; those notebooks show it applied
to real experiments.