An investigation of laser scanning techniques for quantifying changes in facial soft-tissue volume.

This paper describes a laser scanning technique to measure accurately changes in facial volume. Three potential sources of error were identified: the registration algorithm, the inherent accuracy of the scanner, and the natural minute-to-minute variability of the human face. In experiments performed to discover the relative magnitude and significance of each source of error, the performance of the registration algorithm was found to be limited by the accuracy of the data, being 1.6 cc, on average, measured over the entire face of a plaster model. Scan-to-scan variability in the shape of the human face was measured at 3.3 cc on average. We conclude that laser scanning is a simple, non-invasive, accurate, and reproducible means of assessing changes in facial volume.