Melasma is a frequent pigmentary disorder caused by abnormal melanin deposits in the skin. In vivo reflectance confocal microscopy (RCM) is a repetitive imaging tool that provides real-time images of the skin at nearly histological resolution. As melanin is the strongest endogenous contrast in human skin, pigmentary disorders are the most suitable candidates for RCM examination but RCM features of melasma have never been reported. This study investigates the pilot use of RCM in melasma to provide a set of well-described morphological criteria with histological correlations. RCM images were acquired from melasma skin and compared to adjacent control skin in 26 patients. Skin biopsies were obtained from eight patients. In the epidermis, RCM showed in all patients a significant increase in hyperrefractile cobblestoning cells. These cells corresponded to hyperpigmented basal keratinocytes in histology. In six patients, dendritic cells corresponding to activated melanocytes were also found in the epidermis. In the dermis, RCM identified in nine patients plump bright cells corresponding to melanophages. Interestingly, for a given patient, the topographic distribution of melanophages in melasma lesions was very heterogeneous. RCM also showed a significant increase in solar elastosis and blood vessels in the dermis. RCM is a non-invasive technique that detects pigmentary changes in melasma at a cellular level resolution. Therefore, RCM provides an innovative way to classify melasma by pigment changes.

Melasma is a commonly acquired hyperpigmentary disorder of the face, but its pathogenesis is poorly understood and its treatment remains challenging. We conducted a comparative histological study on lesional and perilesional normal skin to clarify the histological nature of melasma. Significantly, higher amounts of melanin and of melanogenesis-associated proteins were observed in the epidermis of lesional skin, and the mRNA level of tyrosinase-related protein 1 was higher in lesional skin, indicating regulation at the mRNA level. However, melanocyte numbers were comparable between lesional and perilesional skin. A transcriptomic study was undertaken to identify genes involved in the pathology of melasma. A total of 279 genes were found to be differentially expressed in lesional and perilesional skin. As was expected, the mRNA levels of a number of known melanogenesis-associated genes, such as tyrosinase, were found to be elevated in lesional skin. Bioinformatics analysis revealed that the most lipid metabolism-associated genes were downregulated in lesional skin, and this finding was supported by an impaired barrier function in melasma. Interestingly, a subset of Wnt signaling modulators, including Wnt inhibitory factor 1, secreted frizzled-related protein 2, and Wnt5a, were also found to be upregulated in lesional skin. Immunohistochemistry confirmed the higher expression of these factors in melasma lesions.

In this paper, we compare two different approaches for semiautomatic detection of skin hyper-pigmentation on multispectral images. These two methods are support vector machine (SVM) and blind source separation. To apply SVM, a dimension reduction method adapted to data classification is proposed. It allows to improve the quality of SVM classification as well as to have reasonable computation time. For the blind source separation approach we show that, using independent component analysis, it is possible to extract a relevant cartography of skin pigmentation.

We demonstrate convergent validation to clinical assessment and repeatability of 3D skin imaging in atrophic acne scarring. Image analysis is straightforward and can be integrated into an automated workflow.

No abstract