IADR Abstract Archives
High-Precision Surface Metrology of sub-80 µm Enamel Lesions Using Intra-Oral Scanning
Objectives: To validate a surface-registration and surface-subtraction metrology algorithm for depth (µm) quantification on natural enamel surfaces using intra-oral scanning (IOS).
Methods: A software algorithm combining iterative-closest-point (ICP) surface-registration and surface-subtraction was developed for surface metrology of natural enamel.
Firstly, software errors (%) for measuring a simulated 80 μm deep lesion were quantified for surface-registration (with and without reference areas) and surface-registration plus surface-subtraction, using softgauges (digital measurement standards for testing the numerical correctness of metrology software). Softgauges were created using spreadsheet software by taking a non-contacting laser profilometry (NCLP) scan of a natural enamel sample and mathematically creating an 80 μm deep simulated lesion on the surface.
Secondly, the precision (μm) of repeated (n=5) scans of a natural enamel sample was quantified for the IOS in comparison to the NCLP using surface-registration software.
Finally, the IOS’s depth measurement threshold was quantified using the combined surface-registration and surface-subtraction algorithm for measurement of lesions with depths, 11 µm to 81 µm, in natural enamel samples (n=14).
Results: Firstly, the software errors for measuring the 80 µm softguage using surface-registration alone ranged from a 29.75% underestimation of the lesion depth without reference areas to 2.69% underestimation with reference areas. When surface-registration and surface-subtraction were combined, these errors reduced: from a 0.05% underestimation without reference areas to finally a 0.00% software error using reference areas.
Secondly, the mean (SD) precision of repeated surface-registration using surface-registration software was 1.2 (1.0) μm for the IOS scans compared to 0.1 (0.2) μm for the NCLP scans.
Finally, the depth measurement threshold of IOS was determined to be 73 μm whereas the NCLP successfully quantified all lesion depths from 11 µm onwards.
Conclusions: This study validated a novel surface metrology algorithm combining surface-registration and surface subtraction. Intra-oral scanning has a 73 μm threshold for measuring depths in natural enamel.
Methods: A software algorithm combining iterative-closest-point (ICP) surface-registration and surface-subtraction was developed for surface metrology of natural enamel.
Firstly, software errors (%) for measuring a simulated 80 μm deep lesion were quantified for surface-registration (with and without reference areas) and surface-registration plus surface-subtraction, using softgauges (digital measurement standards for testing the numerical correctness of metrology software). Softgauges were created using spreadsheet software by taking a non-contacting laser profilometry (NCLP) scan of a natural enamel sample and mathematically creating an 80 μm deep simulated lesion on the surface.
Secondly, the precision (μm) of repeated (n=5) scans of a natural enamel sample was quantified for the IOS in comparison to the NCLP using surface-registration software.
Finally, the IOS’s depth measurement threshold was quantified using the combined surface-registration and surface-subtraction algorithm for measurement of lesions with depths, 11 µm to 81 µm, in natural enamel samples (n=14).
Results: Firstly, the software errors for measuring the 80 µm softguage using surface-registration alone ranged from a 29.75% underestimation of the lesion depth without reference areas to 2.69% underestimation with reference areas. When surface-registration and surface-subtraction were combined, these errors reduced: from a 0.05% underestimation without reference areas to finally a 0.00% software error using reference areas.
Secondly, the mean (SD) precision of repeated surface-registration using surface-registration software was 1.2 (1.0) μm for the IOS scans compared to 0.1 (0.2) μm for the NCLP scans.
Finally, the depth measurement threshold of IOS was determined to be 73 μm whereas the NCLP successfully quantified all lesion depths from 11 µm onwards.
Conclusions: This study validated a novel surface metrology algorithm combining surface-registration and surface subtraction. Intra-oral scanning has a 73 μm threshold for measuring depths in natural enamel.