Diagnostic Performance of Ai vs Junior Dentists

OBJECTIVES * Compare diagnosis of Periapical lesions \& Proximal caries by AI and junior dentists * Evaluate Validity of a deep learning based convolutional neural network (CNN) for detecting proximal caries lesions, periapical lesions compared to junior dentists.

Inclusion Criteria: * Bitewings obtained from patients treated at Jordan University Hospital. Only bitewings of permanent maxillary and mandibular teeth, with the crowns of both dental arch being detectable. Coronal part of the tooth had to be in toto visible on the radiograph and that no structure is projected on the examined tooth, teeth with overlap free projection selected * Periapical Radiographs: patients over 18 years of age that were clear and displayed the full anatomical crown and root apex. Exclusion Criteria: * Radiographs of deciduous teeth or mixed dentition, those with severe noise or haziness, radiographs of partially edentulous areas, teeth exhibiting abnormal crown or root anatomy, and any radiographs that were difficult to assess. * Overlapping images, radiographs with distortions, noise and shadows, and low quality radiographs * Radiographs excluded if at least 2-3 examiners did not agree.