Explainable Artificial Intelligence in Dentistry: A Systematic Review of Its Trust and Translation
Abstract
Introduction and aims: Explainable artificial intelligence (XAI) is a set of methods and processes that make the decisions of artificial intelligence (AI) models understandable to those who are not conversant with the technology. This "black box" nature of complex AI models appears to be a primary barrier to their clinical adoption in health sciences, including dentistry. XAI is being touted as a solution to build clinician trust. This review critically assesses whether current dental XAI research is methodologically rigorous enough to substantiate claims of enhanced trustworthiness.
Methods: This systematic review followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines, searching PubMed, IEEE Xplore, medRxiv, and Ovid for dental XAI studies (2015-2025). We assessed the risk of bias and applicability using QUADAS-2 and PROBAST.
Results: Nineteen of the 100 identified studies met the inclusion criteria. Although these studies used diverse XAI techniques, including image-based saliency methods (eg, Grad-CAM), feature attribution approaches (eg, SHAP), and local approximation methods (eg, LIME) across various dental specialties, quality assessment exposed significant limitations. Most (14 of 19) exhibited a high risk of bias, driven by small retrospective datasets, lack of external validation, and weak reference standards. Interestingly, only 1 study has evaluated the impact of XAI on human understanding.
Conclusion: Current dental XAI…
How strong is the evidence?
Evidence Hub reads this as a systematic review, done in people. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid.
Read the paper
Related studies
- Knowledge Management Perspective of Generative Artificial Intelligence
- Navigating Generative Artificial Intelligence Promises and Perils for Knowledge and Creative Work
- Characterizing the evolving research on enterprise content management
- Large language models and multimodal foundation models for precision oncology
- From Generative AI to Extended Reality: Multidisciplinary Perspectives on the Challenges, Opportunities, and Future of Educational Computing
- Generative Artificial Intelligence: A New Engine for Advancing Environmental Science and Engineering
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.