Open Oral–Maxillofacial Imaging Datasets: Progress, Gaps, and the Path Ahead

Adapted from npj | digital medicine, July 5, 2025

Open-access imaging datasets are essential for driving innovation in artificial intelligence (AI), clinical training, and collaborative research. A new systematic review, published in npj Digital Medicine (2025), provides the first large-scale analysis of openly available oral–maxillofacial (OMF) imaging datasets and their characteristics.

The authors identified 105 unique OMF imaging datasets containing a combined 437,538 images from 21 countries. Most datasets involved panoramic radiographs, cone-beam computed tomography (CBCT), and cephalometric images.

Fig. 3: Representative examples of image types included in the datasets. a Intraoral photograph of a patient missing two maxillary central incisors. b Periapical radiograph of the maxillary right posterior teeth and surrounding alveolar bone, displaying moderate horizontal bone loss and severe dental caries. c Panoramic radiograph providing a comprehensive view of the maxillary and mandibular teeth and jaw structure. d Lateral cephalometric radiograph illustrating a lateral perspective of the skull, teeth, and soft tissue profile. e Axial cone-beam computed tomography (CBCT) scan presenting detailed cross-sectional imaging. f Sagittal magnetic resonance imaging (MRI) scan presenting a sagittal view of craniofacial structures.

From: Characteristics, licensing, and ethical considerations of openly accessible oral-maxillofacial imaging datasets: a systematic review

Key Findings

• Annotations: Nearly 80% of datasets included annotations, but only 25.7% reported the qualifications of those who performed the annotations.

• Ethics Oversight: 83.8% of datasets did not disclose ethics approvals—raising questions about data sourcing and patient consent.

• Licensing: About 61.9% of datasets included reuse terms, most commonly Creative Commons licenses (especially CC BY). However, inconsistent or missing licenses remain a challenge.

• Modalities: Panoramic radiographs and CBCT dominated the datasets, with fewer MRI and intraoral image collections.

• Global Distribution: The datasets spanned 21 countries, highlighting international momentum, though access and standards varied widely.

Why It Matters

The review concludes that while open OMF imaging datasets are increasing, greater transparency and standardization are urgently needed. In particular, the authors call for:

• Clearer reporting of annotator expertise

• Disclosure of ethics approvals and patient consent processes

• Adoption of consistent reuse licenses to ensure legal and ethical data sharing

• Development of AI-specific consent frameworks for future dataset creation

Such improvements are essential for advancing AI tools that rely on imaging data to support diagnosis, surgical planning, and patient care.

Read the full article in npj Digital Medicine

Editor’s Note

Open datasets accelerate discovery, but they also raise important questions for surgeons and researchers. As AI continues to enter the specialty, understanding where the data come from and how they can be reused is critical. This review offers a roadmap for assessing datasets, while underscoring the ethical responsibility to ensure patient privacy, consent, and transparency.

FACE to FACE will continue to highlight work that connects innovation with practical considerations for the global OMF community.

---

Authors and Citations

Hao J; Nalley A; Yeung AWK; Tanaka R; Ai QYH; Lam WYH; Shan Z; Leung YY; AlHadidi A; Bornstein MM; Tsoi JKH; McGrath C; Hung KF. Characteristics, licensing, and ethical considerations of openly accessible oral-maxillofacial imaging datasets: a systematic review. npj Digital Medicine. 2025;8:412. doi:10.1038/s41746-025-01818-5. Open Access (CC BY-NC-ND 4.0). Link: nature.com/articles/s41746-025-01818-5. Key findings: 105 datasets; 437,538 images; 21 countries; ~80% annotated; 25.7% report annotator qualification; 83.8% no ethics disclosure; 61.9% specify reuse terms.