BioCosMe: Lip-based Cosmetics with Colorimetric Biosensors for Salivary Analysis using Deep Learning

Sun S.; Ku Y.-H.; Unsihuay N.; Florez O.; Suk J.Y.; Yetisen A.K.; Vega K. · ISWC 2024 - Proceedings of the 2024 ACM International Symposium on Wearable Computers (2024) · DOI: 10.1145/3675095.3676610

Resumen

This paper introduces three lip-based product biosensors as novel form factors for health monitoring that display pH levels through color variation. Using the unique properties of lip-based products such as different colors, easy application and reapplication on lips, interaction with saliva, we aim to provide an always-Available and non-invasive access to information typically obtained through lab analysis. This paper presents our skin-safe fabrication processes and technical evaluations of a lipstick, lip tint, and lip gloss. We created a mobile app with a Convolutional Neural Network (CNN) model to detect the pH levels. Our dataset involved six users, eight different lighting conditions, three cameras and seven pH levels. The results showed improved detection of pH variations compared to traditional methods. A user study with 11 participants was conducted to evaluate usability. This approach offers a convenient and unexplored form factor for monitoring biochemical information, blending self-expression with health awareness.

Instituciones: Pontificia Universidad Católica del Perú