Automated detection of oral lesions

Abstract

The accurate diagnosis of oral lesions, such as apical periodontitis, is a prerequisite of effective endodontic therapy but remains a significant clinical challenge due to the subjective nature of radiographic interpretation and the potential for human error. Artificial intelligence (AI), particularly its subfields of machine learning and deep learning, has emerged as a transformative technology with the potential to overcome these limitations. This chapter synthesizes the scientific literature on the application of AI for the detection, assessment, and diagnosis of periapical and oral lesions from dental images, illustrated with clinical case studies. The evidence demonstrates that AI models, especially deep learning algorithms, can identify periapical radiolucencies as effectively as clinical specialists and have achieved quantitative accuracy rates as high as 92.8% in detection tasks. Furthermore, AI systems have shown promise in classifying the seriousness of lesions and differentiating between pathologies, such as periapical cysts and granulomas, which has direct implications for treatment planning.