Skin from various ethnic origins and aging: an in vivo cross-sectional multimodality imaging study.

BACKGROUND: Ethnic differences in skin structural features have not been thoroughly investigated, and the few reported studies are contradictory. Thus, we have carried out a set of in vivo measurements on the skin of about 400 volunteers from various ethnic origins living in the same environment. METHODS: Female subjects were distributed into four ethnic groups: African Americans, Mexicans, Caucasians, and Chinese. Inter- and intra-ethnic skin structural differences, according to age and anatomic site, were investigated using three non-invasive skin-imaging methods: ultrasound (US) at 25 and 150 MHz, and optical coherence tomography (OCT). RESULTS: The thickness of the skin is higher on the cheek compared with the dorsal and ventral forearm, with no ethnic or age-related specificity. We confirm that the sub-epidermal non-echogenic band is a sensitive marker of skin aging, and reveal for the first time that it is less pronounced in African Americans. From OCT images, we bring out evidence that the thickness of the dermal-epidermal junction (DEJ) decreased with age, and was higher in African Americans than in Caucasians. Finally, by comparing US images at 150 MHz with OCT images, we show that papillary dermis thickness can be measured and appears to be quite constant irrespective of age or ethnic group. CONCLUSION: Our study confirms that skin imaging is very attractive to further our knowledge of the morphology of skin from various ethnic origins. Regarding age effects, quantitative parameters have shown that they would be delayed in African Americans compared with all other ethnic populations.

Statistical analysis of sets of random walks: how to resolve their generating mechanism.

The analysis of experimental random walks aims at identifying the process(es) that generate(s) them. It is in general a difficult task, because statistical dispersion within an experimental set of random walks is a complex combination of the stochastic nature of the generating process, and the possibility to have more than one simple process. In this paper, we study by numerical simulations how the statistical distribution of various geometric descriptors such as the second, third and fourth order moments of two-dimensional random walks depends on the stochastic process that generates that set. From these observations, we derive a method to classify complex sets of random walks, and resolve the generating process(es) by the systematic comparison of experimental moment distributions with those numerically obtained for candidate processes. In particular, various processes such as Brownian diffusion combined with convection, noise, confinement, anisotropy, or intermittency, can be resolved by using high order moment distributions. In addition, finite-size effects are observed that are useful for treating short random walks. As an illustration, we describe how the present method can be used to study the motile behavior of epithelial microvilli. The present work should be of interest in biology for all possible types of single particle tracking experiments.