The Laryngoscope8 dataset includes 3057 laryngeal images collected from head and neck surgery procedures in the department of otorhinolaryngology of the sixth medical center of PLA general hospital. A total of 1950 individuals are involved in our Laryngoscope8 dataset, from which 1297 individuals had only one laryngeal image, and the remaining 653 individuals collected two or more laryngeal images. The images in the dataset were classified into eight categories by the otolaryngologists in the hospital. The eight categories include: Reinke‘s Edema, Glottic Cancer, Granuloma, Vocal Cord Leukoplakia, Vocal Cord Cyst, Vocal Cord Nodules, Vocal Cord Polyps and Normal.
Here's an example of how the data looks：
Eight labels of Laryngoscope image: (1) Reinke‘s Edema; (2) Glottic Cancer; (3) Granuloma; (4) Vocal Cord Leukoplakia ; (5) Vocal Cord Cyst; (6) Vocal Cord Nodules; (7) Vocal Cord Polyps; (8) Normal.
Clink download(extraction password:soda) to download the whole data.
@article{YIN2021207,
title = {Laryngoscope8: Laryngeal image dataset and classification of laryngeal disease based on attention mechanism},
journal = {Pattern Recognition Letters},
volume = {150},
pages = {207-213},
year = {2021},
issn = {0167-8655},
doi = {https://doi.org/10.1016/j.patrec.2021.06.034},
author = {Li Yin and Yang Liu and Mingtao Pei and Jinrang Li and Mukun Wu and Yuanyuan Jia},
}