Scholarlink

Dental caries represents a significant oral health concern and is experiencing a rapid rise in prevalence across all age groups. The integration of artificial intelligence has facilitated the development of many datasets and approaches to support dentists in tooth decay diagnosis. However, most existing datasets primarily utilize X-ray images such as periapical, bitewing, or panoramic films, while standardized intraoral photographic datasets for dental caries diagnosis remain limited. This study develops an intraoral dental images dataset captured using smartphone devices for machine learning applications, and explores the use of a fuzzy knowledge graph model as an novel approach to support dental caries diagnosis. The experimental results demonstrate that the fuzzy knowledge graph model exhibits significant potential and suitability for intraoral dental image datasets in supporting dental caries diagnosis, with accuracy rates consistently above 60%, ranging from a minimum of 60.95% to a maximum of 65.09%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Dataset of Intraoral Images Using Machine Learning for Dental Caries Diagnosis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Nguyen Duc Hoang,
  • Vu Manh Tuan,
  • Nguyen Hong Tan,
  • Nguyen Thi Giang,
  • Truong Quoc Bao,
  • Tran Manh Tuan

摘要

Dental caries represents a significant oral health concern and is experiencing a rapid rise in prevalence across all age groups. The integration of artificial intelligence has facilitated the development of many datasets and approaches to support dentists in tooth decay diagnosis. However, most existing datasets primarily utilize X-ray images such as periapical, bitewing, or panoramic films, while standardized intraoral photographic datasets for dental caries diagnosis remain limited. This study develops an intraoral dental images dataset captured using smartphone devices for machine learning applications, and explores the use of a fuzzy knowledge graph model as an novel approach to support dental caries diagnosis. The experimental results demonstrate that the fuzzy knowledge graph model exhibits significant potential and suitability for intraoral dental image datasets in supporting dental caries diagnosis, with accuracy rates consistently above 60%, ranging from a minimum of 60.95% to a maximum of 65.09%.