AI-driven Remote Facial Skin Hydration and TEWL Assessment from Selfie Images: A Systematic Solution

2025-09-07

AI Summary

Conventional assessment of key skin barrier function metrics—stratum corneum hydration (SH) and transepidermal water loss (TEWL)—requires specialized, clinic-based instrumentation, limiting accessibility for general users. Method: This work proposes the first non-invasive, remote estimation framework leveraging only standard selfie facial images. We introduce Skin-Prior Adaptive ViT, an end-to-end vision transformer regression model that integrates facial anatomical priors and contrastive learning to enhance robustness under limited physiological labeling. Crucially, we propose a symmetric contrastive regularization mechanism to mitigate model bias induced by severe label imbalance in SH/TEWL distributions. Contribution/Results: Evaluated on real-world data, our method achieves high-accuracy estimation of both SH and TEWL (mean absolute percentage error <8.2%), significantly outperforming existing baselines. This work establishes a novel paradigm for lightweight, smartphone-deployable, and population-scale intelligent skin health monitoring.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Intelligent Robots: State Estimation
• Machine Learning: Privacy

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Skin health and disease resistance are closely linked to the skin barrier function, which protects against environmental factors and water loss. Two key physiological indicators can quantitatively represent this barrier function: skin hydration (SH) and trans-epidermal water loss (TEWL). Measurement of SH and TEWL is valuable for the public to monitor skin conditions regularly, diagnose dermatological issues, and personalize their skincare regimens. However, these measurements are not easily accessible to general users unless they visit a dermatology clinic with specialized instruments. To tackle this problem, we propose a systematic solution to estimate SH and TEWL from selfie facial images remotely with smartphones. Our solution encompasses multiple stages, including SH/TEWL data collection, data preprocessing, and formulating a novel Skin-Prior Adaptive Vision Transformer model for SH/TEWL regression. Through experiments, we identified the annotation imbalance of the SH/TEWL data and proposed a symmetric-based contrastive regularization to reduce the model bias due to the imbalance effectively. This work is the first study to explore skin assessment from selfie facial images without physical measurements. It bridges the gap between computer vision and skin care research, enabling AI-driven accessible skin analysis for broader real-world applications.

Problem

• Assessing skin hydration and TEWL remotely from selfie images
• Eliminating need for specialized clinical instruments for skin measurements
• Addressing data annotation imbalance in AI-based skin assessment models

Innovation

• AI-driven remote skin assessment from selfies
• Skin-Prior Adaptive Vision Transformer model
• Symmetric contrastive regularization reduces bias