CRICVA Subtype Reclassification Metadata

CRICVA Subtype Classification Metadata This repository provides a refined and restructured version of the CRICVA Database (Ferreira et al., 2019), in which cytology images have been manually categorized into six diagnostic classes based on cytopathological review. 📁 File Structure The images have been organized into the following subdirectories, each corresponding to a diagnostic category: CRICVA_Refined/ ├── ASC-H/ ├── ASC-US/ ├── HSIL/ ├── LSIL/ ├── Negative/ ├── SCC/ Each subdirectory contains original images from the CRICVA Database. 🧠 Purpose This curated structure enables efficient supervised learning workflows, including training and evaluation of models for cervical cell classification. It also improves dataset readability and facilitates reproducibility by explicitly grouping images by diagnostic category. 📚 Citation Requirements If you use this metadata or classification structure, you must cite: The original dataset: @dataset{ferreira2019cricva, author = {Ferreira, Daniel S. and Ramalho, Geraldo L. B. and Torres, Débora and Tobias, Alessandra H. G. and Rezende, Mariana T. and Medeiros, Fátima N. S. and Bianchi, Andrea G. C. and Carneiro, Cláudia M. and Ushizima, Daniela M.}, title = {CRICVA Database}, year = 2019, publisher = {Mendeley Data}, version = {V1}, doi = {10.17632/bk45c9yxb9.1} } This metadata resource (Zenodo DOI), if cited or redistributed in publications. ⚠️ Licensing and Usage No image data is redistributed in this repository. This metadata refers to image filenames and class labels only. Redistribution of the CRICVA images themselves must follow the original licensing terms at Mendeley Data. 📝 License This metadata is shared under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license. ✉️ Contact For questions about the classification process or how to reproduce this structure, please contact the maintainers of this repository or the corresponding Zenodo entry.