acia-workflows: Automated Single-cell Imaging Analysis for Scalable and Deep Learning-based Live-cell Imaging Analysis Workflows

Johannes Seiffarth, Richard D. Paul, Katharina Nöh

Keywords: Deep-learning single-cell analysis, Cell segmentation & tracking, Analysis workflows, Microfluidic live-cell imaging, Spatio-temporal phenotyping, Jupyter Notebooks

TL;DR: Acia-workflows is an open-source platform that integrates deep learning–based cell segmentation and tracking tools into modular, reproducible, and user-friendly Jupyter Notebook workflows for automated, high-throughput live-cell imaging analysis.

Abstract: Live-cell imaging technology enables detailed spatio-temporal characterization of living cells at single-cell resolution, which is critical for advancing research across the life sciences, from biomedical applications to bioprocessing. High-throughput setups with tens to hundreds of parallel cell cultivations offer the potential for robust and reproducible insights. However, these insights are hidden within hundreds of GBs of time-lapse imaging data recorded per experiment. Recent advances in state-of-the-art deep learning methods for cell segmentation and tracking now enable the automated analysis of such data volumes, offering unprecedented opportunities to study single-cell dynamics systematically. The next key challenge, however, lies in integrating these powerful tools into accessible, flexible, and user-friendly workflows that support routine application in biological research. In this work, we present **acia-workflows** a platform that combines three key components: (1) the automated live-cell imaging analysis (acia) Python library for modular design of image analysis pipelines supporting eight deep learning segmentation and tracking approaches, (2) the design of workflows that assemble the sequential image analysis pipeline, software dependencies, documentation, and visualizations into a single Jupyter Notebook leading to accessible, reproducible and scalable analysis workflows, (3) a collection of application workflows that demonstrate the analysis and customization capabilities in real-world applications. In particular, we present a subset of three application workflows investigating various types of microfluidic live-cell imaging experiments ranging from growth rate comparisons to a precise, minute-resolution quantitative analysis of the response dynamics of individual cells to changing oxygen conditions. Our extensive collection of more than ten application workflows is open source and publicly available at https://github.com/JuBiotech/acia-workflows.

Submission Number: 22