32,043 brightfield microscopy images sliced from 4K resolution sources, featuring cell-level bounding box annotations. The collection is partitioned into 27,161 training, 3,302 validation, and 1,580 test examples for biological imaging tasks.
Use Cases
- Train unconditional diffusion models to generate synthetic brightfield microscopy images
- Develop cell detection algorithms using the provided bounding box labels
- Evaluate object detection performance on high-resolution sliced microscopy data using the 1,580 test examples
Strengths
- 32,043 total image examples across train, validation, and test splits
- Images are derived from high-fidelity 4K resolution brightfield microscopy
- Includes bounding box annotations for individual cell detection