Inter-Observer Processing and Measurement Error Are Low for 2D Dental Measurements on Shared microCT Scans

Area of Science:

• Paleontology
• Bioarchaeology
• Dental Anthropology

Background:

• Sharing micro-computed tomographic (μCT) scans enhances data accessibility and reduces the need for rescanning specimens.
• Potential for new errors exists when multiple researchers use the same μCT data (TIFF stacks or DICOMs).

Purpose of the Study:

• To investigate whether the use of identical μCT scans by different individuals yields comparable results.
• To assess inter-observer error in the reconstruction, slicing, and measurement of primate teeth using μCT data.

Main Methods:

• μCT scanning of worn and unworn *Cercocebus atys* upper molars at 22 μm resolution.
• Two individuals created 2D mesial slices from TIFF stacks; two individuals reconstructed worn teeth.
• Three researchers measured tooth shape, linear enamel thickness, average enamel thickness, and relative enamel thickness (AET and RET), calculating inter-observer percent error.

Main Results:

• Inter-observer error was generally low across all measurements.
• For unworn teeth, measurement error was primarily attributed to the individual performing the measurement.
• For worn teeth, the reconstruction process introduced statistically significant variation for AET and RET.

Conclusions:

• Different researchers can reliably use the same μCT scans for reconstruction, slicing, and measurement.
• Inter-observer error is generally low, though linear measurements are more susceptible to error.
• Worn teeth do not introduce additional sources of error compared to unworn teeth in μCT analysis.