AI-AIDED VOLUMETRIC ROOT RESORPTION ASSESSMENT FOLLOWING PERSONALIZED FORCES IN ORTHODONTICS: PRELIMINARY RESULTS OF A RANDOMIZED CLINICAL TRIAL
- PMID: 40335201
- DOI: 10.1016/j.jebdp.2025.102095
AI-AIDED VOLUMETRIC ROOT RESORPTION ASSESSMENT FOLLOWING PERSONALIZED FORCES IN ORTHODONTICS: PRELIMINARY RESULTS OF A RANDOMIZED CLINICAL TRIAL
Abstract
Introduction: External apical root resorption (EARR) is an undesirable loss of hard tissues of the tooth root frequently affecting to the maxillary incisors. The magnitude of orthodontic forces is a major treatment-related factor associated with EARR occurrence in orthodontics. The primary aim of the present randomized clinical trial was (i) to quantify the impact of a sequence of personalized force archwires on EARR compared to the conventional standard of care and (ii) compare the 3D-quantification of EARR using two quantification methods (manual or automated AI-aided segmentation).
Material and methods: A superiority two arms-parallel-randomized clinical trial (RCT) was conducted to quantify the EARR of two regime forces [CONSORT-guidelines]. A total of 18/43 patients were randomly assigned [block-size: 4] to Control Group [Ni-Ti archwires sequence] or Experimental Group [selective individualized force archwires]. After 142 days sectorial CBCT were obtained; upper incisors were segmented manually and with AI and the volume/length of root quantified. Method error/descriptive statistics (mean; SD; range) and Student t-test were used to assess the differences between groups (Post hoc adjustment for confounders [95% CI; P < .05]).
Results: The total root volume loss detected by AI was 2.44 ± 6.59 mm3 / 2.42 ± 4.75 mm3 (P > .05) and the mean root length loss was 0.20± 0.23mm/0.42 ± 0.43 mm (P = .045) for control/test group, respectively. Despite length loss showed similar changes when it was quantified with both methods, manual and automatic segmentations (P > .05), differences are observed at volume loss. The results demonstrated greater volume loss detection with manual segmentation than with AI-aided segmentation at the global level, volume by thirds, and 4 mm from the apex. However, as we approached apically, the differences equalized and even diminished, resulting in a greater loss with automatic segmentation 1 mm from the apex in the EG (P = .011).
Conclusions: A non direct-force-dependent effect over EARR (6 months) was observed. Individualized force induces slightly higher root resorption at the apical third at 1-2 mm.
Keywords: AI; CBCT; RCT; Resorption.
Copyright © 2025 Elsevier Inc. All rights reserved.
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Miscellaneous
