Association of mandibular cortical erosion with chronic periodontitis, apical periodontitis, and tooth fracture in Japanese women aged 40 years and older: An artificial intelligence-based analysis

To investigate the association between mandibular inferior cortical erosion on dental panoramic radiographs (DPRs) and tooth loss and related dental pathologies in women aged 40 years and older, using a novel artificial intelligence (AI)-based assessment system. We evaluated 307 Japanese women (mean age ± SD, 66.0 ± 12.5 years) who underwent DPRs at a dental clinic. Mandibular inferior cortical morphology was classified as normal, mildly to moderately eroded, or severely eroded using an AI system for automated cortical analysis. Full-mouth dental parameters included the number of present teeth, periodontal pockets ≥ 4 mm, bleeding on probing (BOP), mobile teeth, apical periodontitis, tooth fractures, and decayed or treated teeth. Poisson regression analyses were performed after adjusting for age, body size, medical history, use of osteoporosis medication, and oral hygiene behavior. Increasing mandibular inferior cortical erosion was significantly associated with fewer premolars and molars (P for trend < 0.05) but not anterior teeth (P for trend = 0.573). The number of periodontal pockets, BOP sites, mobile teeth, and cases of apical periodontitis increased progressively as cortical morphology worsened (all P for trend < 0.001). The association with apical periodontitis remained significant even after additional adjustment for mobile teeth (P for trend < 0.001). Cortical erosion was also associated with tooth fractures (P for trend = 0.005). AI-based mandibular cortical assessment using DPRs is associated with tooth loss and multiple dental pathologies in women aged ≥ 40 years and may help identify individuals at risk of oral disease.