Use of artificial intelligence in oral radiology: a multicenter cross-sectional study in Egypt

Abstract

AI-driven tools augment dentists’ capabilities by assisting to identify essential anatomical structures and improving the precision of diagnosing health conditions. However, the effective application of AI depends on dentists’ awareness and willingness to utilize these technologies proficiently. The Egyptian government is actively promoting the integration of AI into healthcare, yet there remains a notable gap in research concerning Egyptian dentists’ knowledge and attitudes towards AI in oral radiology. This study aimed to bridge this gap by assessing Egyptian dentists’ understanding of AI and identifying the barriers to its clinical implementation. A cross-sectional survey was conducted online among Egyptian dentists. The questionnaire assessed their knowledge, attitudes, and perceived challenges related to AI in oral radiology. Dentists were recruited through academic institutions, professional organizations, and social media platforms. To identify the factors influencing respondents’ knowledge of AI, a logistic regression analysis was applied. Of the 399 participants, 50.3% reported being familiar with AI. Only 16.3% actively used AI in their clinical practice. A significant number (43.9%) learned about AI through self-study, whereas 14.8% attended conferences and workshops to gain knowledge. A strong majority (86%) supported the idea of integrating AI into dental education, 78% believed that AI could have a significant impact on oral radiology, and only 17% thought that AI could completely replace oral radiologists. This study identified several key barriers: limited knowledge (69.9%), lack of training opportunities (73%), and financial limitations (69.4%). The multivariable logistic regression analysis showed that attitudes toward AI were not significant predictors of AI knowledge. Participants working solely as practitioners had lower AI knowledge than those involved in policy or decision-making (OR = 0.44, 95% CI: 0.21–0.91, p = 0.026). Using multiple AI learning methods significantly improved knowledge compared to self-learning alone (OR = 3.72, 95% CI: 1.58–8.88, p = 0.008) and was more effective than any single method. Working in institutions with a clear AI implementation strategy was also associated with higher AI knowledge (OR = 3.31, 95% CI: 1.31–8.45, p = 0.012), while strategies still in development showed no significant benefit (AUC = 0.8). Egyptian dentists recognize the potential benefits of AI in oral radiology but face certain challenges, such as gaps in knowledge, training, and lack of strategic AI initiatives within their institutions. This study was registered at ClinicalTrials.gov with identifier NCT06908603.

Ähnliche Arbeiten