Artificial intelligence (AI) is rapidly entering the dental field, with the potential to augment diagnosis and management. This chapter synthesizes current and emerging clinical applications of AI in dentistry and examines the ethical and practical challenges of adoption. Using a narrative review approach from an oral clinician’s perspective, we draw on reported uses across multiple dental specialties. A wide variety of emerging applications is described, including automated radiographic and histopathological interpretation, orthodontic treatment planning, and oral cancer detection and risk stratification. Reported benefits include improved diagnostic consistency, enhanced workflow efficiency, and potential improvements in access to care. However, major risks remain, including poor-quality or unrepresentative training data, algorithmic bias that may widen health inequalities, challenges to informed consent, unclear accountability and liability, and commercial conflicts of interest. Clinically safe implementation will require ethical-by-design development, diverse and high-quality datasets, transparent validation, ongoing auditing, and education that supports informed human oversight within a hybrid AI clinical model.

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

Artificial Intelligence in Dentistry and Oral Medicine: Ethical Issues and Challenges

  • Molly Harte,
  • Qingmei Joy Feng,
  • Barbara Carey,
  • Owen Addison,
  • Zhi Qin Tan,
  • Yunpeng Li,
  • Rui Albuquerque

摘要

Artificial intelligence (AI) is rapidly entering the dental field, with the potential to augment diagnosis and management. This chapter synthesizes current and emerging clinical applications of AI in dentistry and examines the ethical and practical challenges of adoption. Using a narrative review approach from an oral clinician’s perspective, we draw on reported uses across multiple dental specialties. A wide variety of emerging applications is described, including automated radiographic and histopathological interpretation, orthodontic treatment planning, and oral cancer detection and risk stratification. Reported benefits include improved diagnostic consistency, enhanced workflow efficiency, and potential improvements in access to care. However, major risks remain, including poor-quality or unrepresentative training data, algorithmic bias that may widen health inequalities, challenges to informed consent, unclear accountability and liability, and commercial conflicts of interest. Clinically safe implementation will require ethical-by-design development, diverse and high-quality datasets, transparent validation, ongoing auditing, and education that supports informed human oversight within a hybrid AI clinical model.