Precision of a microscopy confocal laser oral scanner: A pilot study

Aim. The aim of this study was to evaluate the precision
of an intraoral impression device based on microscopy
confocal laser scanning technology.
Methods. One plaster abutment of a natural teeth
was selected as a test object. In order to achieve an
adequate statistical power, 40 scans has to be obtained
with a standardized procedure. Scanner was placed in a
fixed position and the abutment was fixed on a holder:
each scan started from the same position and angulation;
the abutment completed four turns at 0°, 30°, 60°
and 90°. Every four scans the scanner was cooled down
for 5 minutes, shut down and restarted. Scans were imported
in an engineering software and superimposed
with a best-fit algorithm first and then with a global
registration function. They were trimmed with planes
intersected on the models, in order to obtain the same
surface extension. The same 3-dimensional grid was
created and loaded on each 3D scan. The software created
control points defined by x, y, z coordinates were
the grid intersected 3D-models. Coordinates were exported,
converted in their absolute values and summed
together for comparison.
Results. 44 scans were needed in order to obtain 40
good scans. Four scans were discarded before saving
the file because of macroscopical flaws. During model
analysis, one scan has to be discarded because one hole,
undetected during the scanning process, was found: 39
scans were then analyzed. The 3-dimensional grid intersected
the models in 69 locations (L1-L69). The mean
range and the mean SD measured for each location were
0,078 mm (min 0,008 mm, max 0,253 mm; SD 0,056mm)
and 0,017 mm (min 0,002 mm, max 0,050 mm).
Conclusion. To our knowledge this is the first study
that evaluated defined locations between scans instead
of an overall 3D comparison of the volumes; furthermore
it is the first study that has good statistical power.
Within the limits of this pilot study, the tested device
was considered accurate for a clinical practice.