Published October 3, 2026 | Version V1.0

Deep-Fish-Bone: Brightfield Microscopy Dataset for Multi-Label Segmentation of Craniovertebral Skeletal Structures in Zebrafish Larvae

  • 1. Laboratory for Organogenesis and Regeneration (LOR), GIGA Institute, University of Liège, Belgium
  • 2. Department of Medical and Translational Biology, Umeå University, Sweden
  • 3. Montefiore Institute (Department of Computer Science and Electrical Engineering), University of Liège, Belgium

Description

This dataset contains 190 brightfield microscopy annotation records from 34 zebrafish larvae, with pixel-level annotations of 25 developing craniovertebral skeletal structures. Each record includes the microscopy image, individual binary masks for all 25 anatomical structures, and a supporting full-body mask. The dataset includes wild-type (WT), heterozygous (HET), and homozygous (HOM) specimens and is accompanied by metadata linking records to their biological fish identity and genotype.

Files

Deep-Fish-Bone_Dataset_v1.0.zip

Files (1.2 GB)

Name Size
md5:c9aa1ac1ad37ee27d12aaa62fa2a43ee
1.2 GB Preview Download

Additional details

Funding

European Union
BIOMEDAQU 766347

Software