CODE: Annotated Intraoral Images for Oral Mucosal Health Research

704 participants were prospectively recruited for this high-quality image dataset of oral mucosa conditions, ranging from normal tissue to Oral Potentially Malignant Disorders and Oral Cancers. Dr Madan Kumar P D and collaborators collected standardized intraoral photographs from eight standard sites per participant under a Hub-and-Spoke model in India. The dataset, last updated in April 2026, includes expert annotations and extensive metadata for each image.

Use Cases

• Train image classification models to detect oral mucosal conditions based on annotated intraoral images.
• Develop segmentation models for specific oral anatomical sites based on the eight standard photographed regions.
• Benchmark model performance on a dataset with a spectrum of conditions from normal mucosa to oral cancer.
• Study the effects of standardized imaging protocols on diagnostic model consistency.
• Create educational tools for dental students using expert-annotated clinical examples.

Strengths

• Dataset size is 16.1 GB, indicating a substantial collection of high-resolution images.
• Includes 704 participants with images covering a spectrum of oral mucosal conditions.
• Images were captured using a copyrighted Standard Operating Procedure to ensure standardization and reproducibility.
• Each image is accompanied by expert annotations and extensive metadata.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect geographic bias inherent to its collection from spoke centers in India.

Provenance

Source

Ragas Dental College and Hospital, Chennai, served as the Hub for data collection under an ICMR-funded study.

Collection Method

Intraoral images were captured using standardized mobile phone photography protocols at an approximate distance of 4–5 cm, with retraction aids for visualization.

Time Range

null

Freshness

Last updated 2026-04-13 10:20:06; freshness should be verified.

Geography

Data was collected from spoke centers across different regions of India.