Journal of dentistry · 2026
Digital and artificial intelligence-based screening for erosive tooth wear.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
O'Toole S, Gerhardt Mdn, Kuralt M, Hallberg N, Kang J, Austin R, Bartlett D
Аннотация
BACKGROUND: Accurate detection of Erosive Tooth Wear (ETW) is essential for early intervention and monitoring. The Basic Erosive Wear Examination (BEWE) is frequently used clinically, yet reproducibility is limited. Digital intraoral scans and Artificial Intelligence (AI) may offer increased sensitivity, reduced bias, and improved diagnostic reliability.
METHODS: This cross-sectional clinical study (ISRCTN16797270) recruited 61 dentate adults with mild, moderate and severe ETW, with analysis of 1600 teeth (4797 surfaces). Participants received a clinical BEWE and dentine-exposure examination by Examiner 1, followed by an intraoral scan (TRIOS 5, 3Shape A/S, Denmark). After a 2-week washout, Examiner 1, Examiner 2 and an AI assessment tool independently assessed BEWE and dentine on each scan. Twelve patients (n = 24 scans) reflecting mild, moderate and severe wear were reassessed by Examiners and the AI to determine intra-examiner repeatability. Sensitivity, specificity, and Intraclass Correlation Coefficients (ICC) were calculated at surface, tooth and patient-level.
RESULTS: On-scan assessments recorded increased BEWE scores >2 (37%) than clinical examination (22.3%). On-scan assessment-clinical agreement was good (ICC 0.73-0.78). AI-clinical agreement was moderate at surface level with the AI scoring more wear than clinical examination (ICC=0.65). Tooth-level sensitivity/specificity of on-scan versus clinical scoring was 0.98/0.56 respectively. AI-clinical sensitivity/specificity was 0.85/0.68. For dentine-exposure detection, sensitivity/specificity exceeded 0.80; AI achieved 0.82/0.90. Intra-examiner and inter-examiner ICC's were 0.73-0.78 and 0.49 respectively while AI demonstrated perfect repeatability (ICC=1.00).
CONCLUSION: AI-derived assessments demonstrated perfect repeatability alongside comparable sensitivity/specificity with clinical assessment and improved dentine exposure assessment. Digital and AI-derived wear assessments represent promising adjuncts for earlier diagnosis and improved monitoring of ETW.
CLINICAL SIGNIFICANCE: Digital intraoral scanning combined with artificial intelligence improves the sensitivity and diagnostic consistency of identifying tooth wear. This has the potential to position them as the gold standard for wear assessment and, at the minimum, to be promising adjuncts to traditional BEWE scoring in clinical and research settings.
METHODS: This cross-sectional clinical study (ISRCTN16797270) recruited 61 dentate adults with mild, moderate and severe ETW, with analysis of 1600 teeth (4797 surfaces). Participants received a clinical BEWE and dentine-exposure examination by Examiner 1, followed by an intraoral scan (TRIOS 5, 3Shape A/S, Denmark). After a 2-week washout, Examiner 1, Examiner 2 and an AI assessment tool independently assessed BEWE and dentine on each scan. Twelve patients (n = 24 scans) reflecting mild, moderate and severe wear were reassessed by Examiners and the AI to determine intra-examiner repeatability. Sensitivity, specificity, and Intraclass Correlation Coefficients (ICC) were calculated at surface, tooth and patient-level.
RESULTS: On-scan assessments recorded increased BEWE scores >2 (37%) than clinical examination (22.3%). On-scan assessment-clinical agreement was good (ICC 0.73-0.78). AI-clinical agreement was moderate at surface level with the AI scoring more wear than clinical examination (ICC=0.65). Tooth-level sensitivity/specificity of on-scan versus clinical scoring was 0.98/0.56 respectively. AI-clinical sensitivity/specificity was 0.85/0.68. For dentine-exposure detection, sensitivity/specificity exceeded 0.80; AI achieved 0.82/0.90. Intra-examiner and inter-examiner ICC's were 0.73-0.78 and 0.49 respectively while AI demonstrated perfect repeatability (ICC=1.00).
CONCLUSION: AI-derived assessments demonstrated perfect repeatability alongside comparable sensitivity/specificity with clinical assessment and improved dentine exposure assessment. Digital and AI-derived wear assessments represent promising adjuncts for earlier diagnosis and improved monitoring of ETW.
CLINICAL SIGNIFICANCE: Digital intraoral scanning combined with artificial intelligence improves the sensitivity and diagnostic consistency of identifying tooth wear. This has the potential to position them as the gold standard for wear assessment and, at the minimum, to be promising adjuncts to traditional BEWE scoring in clinical and research settings.