A multimodal dental dataset facilitating machine learning research and clinic services

Area of Science:

• Dentistry
• Medical Imaging
• Machine Learning

Background:

• Oral diseases impact billions globally, with limited resources hindering access to care.
• Machine learning in dentistry offers potential for improved services and cost reduction.
• Existing dental datasets lack volume, multimodal data, and diversity, impeding ML development.

Purpose of the Study:

• To introduce a novel, comprehensive dental dataset to address limitations in current resources.
• To facilitate advancements in imaging-based machine learning for oral healthcare.

Main Methods:

• A new dental dataset was curated, including data from 169 patients.
• The dataset incorporates three common dental imaging modalities.
• Images capture a variety of oral cavity health conditions.

Main Results:

• The dataset provides a larger, more diverse collection of dental images than previously available.
• It includes multimodal imaging data, crucial for complex ML models.
• The dataset encompasses varied oral health conditions, enhancing its utility.

Conclusions:

• The proposed dataset is poised to significantly benefit research in oral medical services.
• It will support the development of applications like 3D reconstruction, diagnosis assistance, image translation, and segmentation.
• This resource has the potential to drive innovation in dental AI and improve patient outcomes.