cs.CVNov 15, 2026

Self-supervised learning for zebrafish embryo phenotype evaluation: Systematic comparison of algorithms, label-efficient classification, and transfer learning

Authors: Thomas Lautenschlager, Gaëlle Hayot, Quentin Brassart, Angelo Jovin Yamachui Sitcheu, Linus Schilpp, Tijana Geroski, Tetyana Goldanova, Stephan Allgeier, 

Organizations: Karlsruhe Institute of Technology, Institute for Automation and Applied Informatics, Eggenstein-Leopoldshafen, 76344, Germany · Karlsruhe Institute of Technology, Institute for Biological and Chemical Systems - Biological Information Processing, Eggenstein-Leopoldshafen, 76344, Germany · Institut de Génétique et de Biologie Moléculaire et Cellulaire, Illkirch CEDEX, 67400, France · Steinbeis Advanced Risk Technologies Institute Kragujevac, Kragujevac, 34000, Serbia · 

Abstract

Zebrafish embryos are a prominent model organism for image-based phenotyping and especially suited for large-scale imaging studies, for which machine learning (ML) models can automate evaluation. However, training ML models usually requires large labeled datasets, and differences in imaging conditions often impede reuse across laboratories. Self-supervised learning (SSL) addresses both constraints by learning transferable representations from unlabeled images. Yet central aspects like the choice of algorithm remain underexplored for domain-specific applications. To close this gap, we benchmark popular SSL algorithms on the public EmbryoNet dataset, explore label-efficient learning, and test transfer to an unseen setting. Using representations from our strongest model, a ResNet50 network trained with SwAV, a simple linear classifier reaches 86.4% accuracy and a non-linear classifier 87.1%, rivaling the fully supervised EmbryoNet model (87.1%). The SwAV model decisively outperforms supervised training from scratch when labels are scarce (70.2% vs. 45.7% accuracy with 5% of labels). On a three-class problem (normal, dead, falsely cropped) with a small, previously unseen zebrafish toxicity dataset, recorded in a different setting, our model reaches 81.3% accuracy without retraining, while EmbryoNet achieves 75.9%. Overall, self-supervised models for zebrafish embryos seem to mostly facilitate label-efficiency and transfer to new settings rather than improving supervised results.

Figures & tables

Explore similar work

CardsList
  1. Hierarchical Specialised Ensembles for Classification of Zebrafish Phenotypes Using the Selected Image Recognition Methods

    Jul 17, 2026Piotr S. Maciąg, Monika Maciąg, Magdalena MajdanHierarchical ClassificationMulti-Label Classification

  2. Automated Detection of Abnormalities in Zebrafish Development

    May 11, 2026Sarath Sivaprasad, Hui-Po Wang, Anna-Lisa Jäckel +4Spatiotemporal TransformersToxicity Detection

  3. IFC-SSL-Bench: Benchmarking Self-Supervised Learning for Annotation-Efficient Multispectral Imaging Flow Cytometry

    Oct 22, 2026Angelo Jovin Yamachui Sitcheu, Roxana-Teodora Fetescu, Thomas Lautenschlager +6Self-Supervised LearningSelf-Supervised Visual Representation Learning