Characterisation of mineral loss as a function of depth using confocal laser scanning microscopy to study erosive lesions in enamel: A novel non-destructive image processing model

[Display omitted] •Novel image processing application for monitoring erosive lesion progression.•Fluorescent volume depth profiles distinguished differences between lesion types.•Captured erosion as complex, concurrent surface stripping and sub-surface depletion.•No steady state of erosion was shown to exist. The aim was to develop a novel image processing protocol for confocal laser scanning microscopy (CLSM) to study mineral distribution within erosive lesions as a function of depth. Polished bovine enamel samples (n = 80) were divided into groups (8/group) with similar mean surface microhardness (SMH) values. Samples underwent erosion (1 % citric acid pH3.8) for 1,5,10,15, or 30 min, with or without stirring giving 10 treatment groups in a 2*5 factorial design. SMH was used to measure erosive softening. Profilometry was used to measure bulk tissue loss. Samples were then stained with rhodamine-B (0.1 mM, 24 h) and imaged using CLSM. Image processing was used to measure fluorescence volume (FV) as a function of depth for each image. The data from reference images were subtracted from post-erosive data to determine changes in fluorescent volume (ΔFV) as a function of depth. 2-way ANOVA and linear regression analysis were used where applicable. Surface softening and bulk tissue loss increased with acid erosion duration with or without stirring. Stirring significantly increased net softening at each time point; specimens underwent significantly more bulk tissue loss (P