AutoQC-Bench: a diffusion model and benchmark for automatic quality control in high-throughput microscopy
Reliable biomedical imaging demands rigorous quality control, yet high-throughput microscopy remains prone to diverse artifacts. We present AutoQC-Bench, a software based on a reconstruction-driven diffusion model flagging abnormal images without prior knowledge, and along with a benchmark of 8000 images capturing common quality issues. The software outperforms existing methods, generalizes across modalities, and supports large-scale bioimaging studies. The software and benchmark are openly shared to advance robust microscopy quality control.
Authors
- Matthias Gunzer (ORCID: https://orcid.org/0000-0002-5534-6055)
- Jianxu Chen (ORCID: https://orcid.org/0000-0002-8500-1357)
- Justin Sonneck (ORCID: https://orcid.org/0000-0002-1640-3045)
- Anja Hasenberg (ORCID: https://orcid.org/0000-0002-4468-6361)
- Zixuan Pan
- Yiyu Shi
- Dennis Nagel (ORCID: https://orcid.org/0009-0004-2911-9301)
Institutions
- University of Notre Dame (US)
- Leibniz Institute for Analytical Sciences - ISAS (DE)
- University of Duisburg-Essen (DE)
- Ruhr University Bochum (DE)
Publication Details
- Journal
- npj Imaging
- Published
- 2025-11-07
- DOI
- https://doi.org/10.1038/s44303-025-00117-8
- Citations
- 1
- Primary Topic
- Cell Image Analysis Techniques
- Type
- article
- Field-Weighted Citation Impact
- 1.63
Funders
- Deutsche Forschungsgemeinschaft
- Ministerium für Kultur und Wissenschaft des Landes Nordrhein-Westfalen