Searcharxiv⌕ Search

arXiv subjects

Linus Schilpp

Publications and source records attributed to Linus Schilpp.

3 recordsLinked to original sources

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

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.

cs.CV↗

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

Imaging Flow Cytometry (IFC) generates large multispectral datasets, yet the need for expert annotations bottlenecks the development of supervised deep learning models. While Self-Supervised Learning (SSL) offers a solution, its application to IFC remains fragmented and lacks systematic benchmarking. To address this, we present IFC-SSL-Bench, a modular framework for evaluating SSL in IFC. We benchmark six SSL approaches across four backbones on two heterogeneous datasets. By assessing downstream classification performance across annotation regimes of the training sets (1%-100%), we quantify annotation efficiency and identify optimal configurations for high-dimensional biological imaging. SSL matches the performance of fully supervised baselines trained on 100% of annotated data while using as little as 5% of annotations for morphological classification, with advantages in low-data regimes below 5%. Across both datasets, the contrastive approach MoCo-v2 paired with ConvNeXt backbones emerged as the top-performing configuration, with ConvNeXtV1-T best for immunological synapse classification and ConvNeXtV2-T for morphological classification. We identify a clear pattern in transfer dynamics: ImageNet initialization benefits morphological classification, consistent with its visual similarity to natural images, whereas random initialization better captures the more domain-specific immunological synapse classification. SSL elevates class-specific F1-scores by over 35 percentage points for structurally complex classes at 1% of annotated data. We synthesize these insights into a unified set of guidelines for deploying self-supervised workflows in high-throughput IFC, and release the benchmark as a resource for future work.

cs.CV↗

napari-TurboBox: Interactive Multi-Viewer Bounding Box Annotation for AI-Guided Volumetric Bioimaging

Promptable segmentation models turn one 3D bounding box into an instance mask. Box placement therefore limits how fast microscopy volumes with thousands of objects are annotated. Each box must be checked in several synchronized viewers, but napari-bbox shares no state between viewers. An edit must update every viewer and slows down with more boxes. napari-TurboBox instead holds one shared copy of the boxes for all viewers. In a four-viewer layout, napari-TurboBox needs a few milliseconds per mouse move to update the three slice viewers up to $10^4$ boxes. The same update in napari-bbox leaves the 33 ms budget beyond about $10^3$ boxes and takes 159 ms at $10^4$. Boxes stay axis-aligned and export to COCO and YOLO as per-slice 2D boxes. napari-TurboBox is a pure-Python plugin for napari 0.6.6 to 0.9. The code is available at https://github.com/N-Friederich/napari-turbobox under the Apache 2.0 licence, and version 0.1.0 is archived at https://doi.org/10.5281/zenodo.23090828 on Zenodo.

eess.IV↗