Impact of Artificial Intelligence-Assisted Radiographic Caries Detection on Clinical Decision-Making and Patient Outcomes: A Systematic Review
Fitri Yustika Sya’bani1, Titiek Berniyanti2*, Dina Izzatul Mila Azka1
Abstract
Dental caries remains a major global health problem, with approximately 2.24 billion people worldwide affected by untreated caries in permanent teeth in 2021. This persistent burden highlights the importance of accurate detection and appropriate management of carious lesions. Artificial intelligence (AI) is increasingly investigated for caries detection on dental radiographs, yet improved diagnostic performance does not necessarily translate into appropriate treatment decisions or better patient outcomes. This systematic review evaluated evidence linking AI-assisted radiographic caries assessment with treatment decisions, treatment intensity, clinical consequences, and economic outcomes. The method used is a Systematic Literature Review (SLR) approach guided by PRISMA 2020, with searches carried out in PubMed, Scopus, Web of Science, and SpringerLink. Eligible primary studies applied AI to dental radiographs and linked caries assessment to treatment decisions, treatment recommendations, treatment intensity or appropriateness, clinical consequences, or economic outcomes. Eleven studies were included from 998 records, comprising one cluster-randomized crossover trial, four economic/modeling evaluations, two diagnostic or ex vivo decision studies, and four treatment-planning or decision-support studies. AI assistance improved sensitivity and AUC but also altered treatment decisions, including increased invasive treatment for some lesions. Economic models generally favored AI when early lesions received non-restorative management, whereas trial-derived decisions showed no economic advantage. Treatment-recommendation systems demonstrated technical feasibility but lacked evidence of improved clinical appropriateness or patient outcomes. Overall, AI may influence treatment pathways, but clinical utility evidence remains less mature than diagnostic evidence. Future studies should evaluate AI using prospective designs, independent validation, lesion-specific management pathways, and patient-important outcomes.
Keywords
artificial intelligence; dental caries; radiography; clinical decision-making; treatment planning
Cite This Article
Sya’bani, F. Y., Berniyanti, T., Azka, D. I. M. (2026). Impact of Artificial Intelligence-Assisted Radiographic Caries Detection on Clinical Decision-Making and Patient Outcomes: A Systematic Review. International Journal of Scientific Advances (IJSCIA), Volume 7| Issue 5: Sep – Oct 2026, Pages 860-868. URL: https://www.ijscia.com/wp-content/uploads/2026/10/Volume7-Issue5-Sep-Oct-No.1107-860-868.pdf
Volume 7 | Issue 5: Sep – Oct 2026
