Artificial intelligence systems in dental shade-matching: A systematic review

Area of Science:

• Dentistry
• Artificial Intelligence
• Computer Science

Background:

• Artificial intelligence (AI) applications are expanding in dentistry, particularly for dental shade-matching in restorative procedures.
• A systematic review was conducted to evaluate the accuracy of AI in predicting dental shades.

Approach:

• A comprehensive search of multiple databases (PubMed, Scopus, Cochrane Library, Google Scholar) and manual searches was performed.
• Included studies were observational or interventional, published in English, and focused on AI-based dental shade-matching in restorative dentistry.
• Study quality was assessed using the Joanna Briggs Institute Critical Appraisal Checklist.

Key Points:

• Fifteen articles published between 2008 and March 2023 met the inclusion criteria.
• AI algorithms evaluated included fuzzy logic, neural networks, support vector machines, and deep learning (e.g., YOLO).
• The decision tree regression model achieved the highest accuracy (99.7%) for leucite-based dental ceramics, followed by fuzzy logic (99.62%) and support vector machines (97%).

Conclusions:

• The accuracy of AI dental shade-matching is influenced by lighting conditions, shade-matching devices, color space models, and the specific AI algorithm used.
• Knowledge-based systems and neural networks demonstrate superior accuracy in predicting dental shades.