Analysis of Correlation between Electron Diffraction Images of Corneocytes and Skin Barrier Properties by Multilayer Perceptron and Convolutional Neural Network

• Published in: Sensors and materials 2024-06, Vol.36 (6), p.2557
• Main Authors: Takahashi, Yoshifumi, Iida, Yusuke, Nakazawa, Hiromitsu
• Format: Article
• Language: English
• Subjects: Anisotropy; Artificial neural networks; Correlation; Electron diffraction; Image acquisition; Lipids; Moisture content; Multilayer perceptrons; Parameters; Skin; Water loss
• ISSN: 0914-4935; 2435-0869
• DOI: 10.18494/SAM5013

Skin parameters such as transepidermal water loss (TEWL) and water content are important information in relation to the skin barrier function of the human body. It was reported that the structures of corneocytes and intercellular lipids are important for the skin barrier. In a recent study, it has been found that there is a correlation between the packing structure of intercellular lipids and TEWL. However, there have been no studies that focused on the correlation between the structures of individual cells and skin parameters. On the other hand, the advances in sensor technology have made it possible to acquire high-resolution 2D electron diffraction (ED) images. Thus, we attempted to examine the relationship between the 2D ED images of corneocytes and the TEWL or water content values, which is difficult with the rule-based analysis, by introducing a deep learning model. Our results showed that the highest prediction accuracy of 13.92 ± 0.57% as the error rate is achieved for water content with a diffraction image rather than with 1D ED profiles, which suggests that spatial anisotropy in a 2D image may contribute to the skin barrier function.