Three-dimensional microscopic quantification of in vivo healthy epidermis based on line-field confocal optical coherence tomography (LC-OCT) assisted by artificial intelligence

Line-field confocal optical coherence tomography (LC-OCT) is an imaging technique based on a combination of reflectance confocal microscopy and optical coherence tomography, allowing three-dimensional (3D) imaging of skin in vivo with an isotropic spatial resolution of about 1.3 micron and up to 400 microns in depth. Cellular-resolution 3D images obtained with LC-OCT offer a considerable amount of information for description and quantification of the upper layers of in vivo skin using morphological metrics, which can be critical for better understanding the skin changes leading to aging or some pathologies. This study introduces metrics for the quantification of the epidermis, and uses them to describe the variability of healthy epidermis between different body sites. These metrics include the stratum corneum thickness, the undulation of the dermal-epidermal junction (DEJ), and the quantification of the keratinocyte network. In order to generate relevant metrics over entire 3D images, an artificial intelligence approach was applied to automate the calculation of the metrics. We were able to quantify the epidermis of eight volunteers on seven body areas on the head, the upper limbs and the trunk. Epidermal thicknesses and DEJ undulation variations were observed between different body sites. The cheek presented the thinnest stratum corneum the least undulated DEJ, while the back of the hand presented the thickest stratum corneum and the back the most undulated DEJ. The process of keratinocyte maturation was evidenced in vivo. These 3D in vivo quantifications open the door in clinical practice to diagnose and monitor pathologies for which the epidermis is impaired.