The skin serves as a physical and chemical barrier, effectively protecting us against the external environment. The Baumann Skin Type Indicator (BSTI) classifies skin into 16 types based on traits such as dry/oily (DO), sensitive/resistant (SR), pigmented/nonpigmented (PN), and wrinkle-prone/tight (WT). Traditional assessments are time-consuming and challenging as they require the involvement of experts. While deep learning has been widely used in skin disease classification, its application in skin type classification, particularly using multimodal data, remains largely unexplored. To address this, we propose an improved Inception-v3 model incorporating transfer learning, based on the four-dimensional classification of the Baumann Skin Type Index (BSTI), which demonstrates outstanding accuracy. The dataset used in this study includes non-invasive physiological indicators, BSTI questionnaires, and skin images captured under various light sources. By comparing performance across different light sources, regions of interest (ROI), and baseline models, the improved Inception-v3 model achieved the best results, with accuracy reaching 91.11% in DO, 81.13% in SR, 91.72% in PN, and 74.9% in WT, demonstrating its effectiveness in skin type classification. This study surpasses traditional classification methods and previous similar research, offering a new, objective approach to measuring comprehensive skin types using multimodal and multi-light-source data.

Background and Objective: Skin tone correction is an essential focus within dermatology and cosmetology, particularly in achieving a balanced and even facial appearance. The application of complementary color theory in skin tone correction remains predominantly subjective, relying on individual user experiences rather than systematic and quantitative assessments. This study aims to evaluate the applicability of complementary color theory among young Chinese individuals and develop predictive models for personalized skin tone correction. Methods: Sixteen young Chinese female participants aged 20–25 were recruited. Standardized facial images were captured using the VISIA‐CR system under standardized lighting conditions, both before and after the application of six color‐correcting primers (orange, pink, blue, white, purple, and green). Four facial regions of interest (ROIs), defined as the forehead, under‐eye circles, cheeks, and near‐nose, were analyzed. Five colorimetric indices ( L *, a *, b *, ITA°, and Hab°) were quantified across each ROI. State‐of‐the‐art machine learning regression models were developed to predict post‐application ITA° and Hab° values based on pre‐application skin tone and primer characteristics. Results: Under‐eye circles exhibited the darkest and most yellowish‐red skin tone compared to other regions. Complementary color primers demonstrated statistically significant improvements in ITA° and Hab° values across all ROIs. Pink primers were most effective for under‐eye dark circles, while purple, pink, and blue primers resulted in greater improvements in overall skin tone. LightGBM and XGBoost regression models demonstrated superior performance, with R ² values reaching 0.824 for ITA° and 0.850 for Hab°. Conclusion: This study robustly validates the efficacy of complementary color primers in skin tone correction among young Chinese individuals. The integration of machine learning offers a robust framework for personalized cosmetic recommendations, paving the way for innovative and data‐driven advancements in skincare and makeup applications.

Ceramides constitute approximately 50% of the lipid mass in the stratum corneum and are essential for epidermal permeability barrier function. In acne vulgaris, barrier disruption is accompanied by altered ceramide composition, yet the molecular mechanisms that connect acne-associated signals to specific changes in ceramide metabolism remain incompletely understood. This review integrates evidence from lipidomic profiling, enzymatic pathway characterization, and inflammatory signaling studies to examine how ceramide biosynthesis and turnover are modulated in acne-prone skin. Particular attention is directed to the elongation of very-long-chain fatty acids (VLCFAs) and their incorporation into ceramides by ceramide synthase 3 (CerS3). In cultured keratinocytes, Cutibacterium acnes (C. acnes)-derived short-chain fatty acids (SCFAs) have been reported to suppress CerS3 and 3-ketodihydrosphingosine reductase (KDSR) through peroxisome proliferator-activated receptor α (PPARα), limiting ultra-long-chain fatty acid (ULCFA) supply. We propose a working model in which SCFAs act via PPARα to upregulate ELOVL-family elongases while suppressing CerS3 and KDSR, reducing ULCFA-ceramides and compromising barrier integrity. The CerS3–KDSR node is a candidate regulatory point coupling microbial metabolites to barrier impairment and inflammation. However, this model rests largely on in vitro evidence and requires in vivo validation. Clarifying these regulatory relationships may inform the development of barrier-restorative strategies as adjuncts to conventional acne therapy.