This question led us to search publicly available dental imaging datasets comprehensively. AI's potential in dentistry is undeniable, but its progress relies heavily on large, well-annotated datasets for training, testing, and validation. Dental imaging data has been underrepresented—until now.
Our recent research identified 16 unique dental imaging datasets, including panoramic radiographs, cone beam computed tomography (CBCT), and intraoral photographs, all critical for AI model development. These datasets are available from various sources, including Kaggle, GitHub, and Zenodo, but gaps remain, particularly around ethical reporting and dataset licensing. This area needs improvement to ensure the development of fair and robust AI tools.
We included
The global distribution of datasets and the countries contributing the most data are China, Iran, and the United States.
The datasets' key characteristics incluiding image types, licensing details, metadata and ethical considerations.
Our recent research identified 16 unique dental imaging datasets, including panoramic radiographs, cone beam computed tomography (CBCT), and intraoral photographs, all critical for AI model development. These datasets are available from various sources, including Kaggle, GitHub, and Zenodo, but gaps remain, particularly around ethical reporting and dataset licensing. This area needs improvement to ensure the development of fair and robust AI tools.
We included
The global distribution of datasets and the countries contributing the most data are China, Iran, and the United States.
The datasets' key characteristics incluiding image types, licensing details, metadata and ethical considerations.
We are proud to contribute to this field and grateful for the support of the ITU/WHO/WIPO Global Initiative on Artificial Intelligence for Health-Dental Diagnostics and Digital Dentistry. The paper includes all the dental data needed for researchers and developers to develop or test AI models, and a GitHub repository is available for easy access.
This paper is part of a special issue to be published in the Journal of Dental Research in December this year, so stay tuned for more!. You can access the full paper here:
Uribe, S.E., Issa, J., Sohrabniya, F., Denny, A., Kim, N.N., Dayo, A.F., Chaurasia, A., Sofi-Mahmudi, A., Büttner, M., & Schwendicke, F. (2024). Publicly available dental image datasets for artificial intelligence. Journal of Dental Research, 1-10. https://doi.org/10.1177/00220345241272052.
Comments
Post a Comment