ラットカリエスの評価におけるデジタルアプローチの可能性

This study aimed to assess the sensitivity, specificity, and accuracy of a digital-based method for caries assessment in a rodent model, compared to a conventional modified-Keyes method.

108 mandible molars of Sprague Dawley rats were collected from a cariogenic caries rodent model, including both caries-free and carious teeth. Smooth surfaces were evaluated using digital photography, whereas sulcal lesions and volumetric quantification were assessed with micro-CT and the Amira software. In the conventional assessment, smooth surfaces were examined under a stereomicroscope with tactile probing, and sulcal surfaces were evaluated on stained, sectioned teeth. Sensitivity, specificity, and accuracy of the digital approaches were calculated, with the conventional method serving as the gold standard. Spearman correlation analysis was conducted between conventional enamel caries scores, combined digital (photography and micro-CT) enamel caries scores, and micro-CT quantified remaining enamel volume.

A total of 107 molars were available for smooth surfaces evaluation, and 98 molars were included for sulcal surfaces, tooth-level caries detection and enamel lesion assessments. For smooth surfaces, the digital photographic method demonstrated an accuracy of 88.0%, with a sensitivity of 88.8% and a specificity of 85.7%. In the evaluation of sulcal caries, Micro-CT analysis achieved an accuracy of 96.9%, with 100% sensitivity and 87.0% specificity. For tooth-level caries detection, the combined digital approaches of smooth surface photography and sulcal micro-CT slicing achieved 96.9% accuracy, 97.6% sensitivity, and 92.9% specificity.

This study demonstrates the efficacy of digital photography and micro-CT in the caries assessment of rodent models. The findings support the establishment of a standardized, non-destructive imaging protocol to enhance consistency in caries research.