Determining the reliability of diagnosis and treatment using artificial intelligence software with panoramic radiographs

• DOI: 10.5624/isd.20230109
• Publication Date: 2023-08-07T23:51:21Z
• Authors: Kaan Orhan; Ceren Aktuna Belgin; David Manulis; Maria Golitsyna; Seval Bayrak; Secil Aksoy; Alex Sanders; Merve Önder; Matvey Ezhov; Mamat Shamshiev; Maxim Gusarev; Vladislav Shlenskii

The objective of this study was to evaluate the accuracy and effectiveness an artificial intelligence (AI) program in identifying dental conditions using panoramic radiographs (PRs), as well assess appropriateness its treatment recommendations.PRs from 100 patients (representing 4497 teeth) with known clinical examination findings were randomly selected a university database. Three dentomaxillofacial radiologists Diagnocat AI software evaluated these PRs. evaluations focused on various treatments, including canal filling, caries, cast post core, calculus, fillings, furcation lesions, implants, lack interproximal tooth contact, open margins, overhangs, periapical periodontal bone loss, short voids root overfillings, pontics, fragments, impacted teeth, crowns, missing healthy teeth.The demonstrated almost perfect agreement (exceeding 0.81) most assessments when compared ground truth. sensitivity very high (above 0.8) for evaluation implants. However, low assessment pontic canal, overhangs.Despite limitations study, synthesized data suggest that AI-based decision support systems can serve valuable tool detecting conditions, used PR applications.