We are excited to announce the publication of a new dataset designed for the study of gene expression in wound bed tissues, complemented by visual information. Its descriptor is now available in "Scientific Data".
Skin wounds are a common part of life, and while many of us rely on simple remedies, complex cases often require medical intervention and tailored treatment. Accelerating healing and optimizing therapy depend on choosing the right intervention for each stage of this dynamic process.
The dataset includes annotated photographs of wounds, offering valuable material for developing computer vision models in medical image analysis. Researchers are already working on training neural networks to automatically recognize wound stages.
In addition to images, the dataset provides transcriptomic data from the wound tissues, carefully annotated with healing time points. What makes this dataset unique is its high temporal resolution: 15 time points spanning 21 days of healing. Early stages are captured daily, while later stages are sampled every 2–3 days, reflecting the natural progression from inflammation to tissue regeneration.
This dataset offers researchers an unprecedented opportunity to:
- Explore gene expression dynamics during healing with fine temporal detail.
- Train and validate machine learning algorithms on real, annotated multimodal data.
- Advance the development of tools for precision wound care.