Skin Nanotexture Biometrics: A Novel Deep Learning Approach for Quantifying Atopic Dermatitis Severity Across Fitzpatrick Skin Types
Non classé | Scientific PublicationsSkin Nanotexture Biometrics: A Novel Deep Learning Approach for Quantifying Atopic Dermatitis Severity Across Fitzpatrick Skin Types
IFSCC 2025 Congress 2025, Full Paper (Abstract N° IFSCC2025-1771)
This study presents a new deep learning method for objectively assessing atopic dermatitis (AD) severity across different skin tones. Traditional diagnostic scales, such as SCORAD or EASI, often struggle to provide consistent results because they rely on visible redness, which varies with melanin content. Researchers applied the Effective Corneocyte Topographical Index (ECTI), a quantitative biomarker derived from nanoscale imaging of corneocytes (outer skin cells) using high-speed atomic force microscopy (HS-DAFM). Samples were collected from 90 participants in Taiwan and Denmark with diverse Fitzpatrick skin types (II–V). The AI model accurately identified nanoscale circular features (CNOs) linked to barrier dysfunction. Results showed that ECTI scores increased consistently with AD severity, independent of skin pigmentation. This confirms ECTI’s potential as a universal, noninvasive diagnostic tool to monitor skin barrier health and improve equitable dermatological care across all populations.
Other articles
Assessment of the Severity of Atopic Dermatitis Using Atomic Force Microscopy Analysis of Skin Tape Strips
Assessment of the Severity of Atopic Dermatitis Using Atomic Force Microscopy Analysis of Skin Tape StripsAllergy, 2025; 0:1-3This letter for the European Journal of Allergy and Clinical Immunology presents new results showing that atomic force microscopy (AFM) can...
From bits to biometrics: sustainable hacking of optical storage technologies for atomic force microscopy and medical applications
From bits to biometrics: sustainable hacking of optical storage technologies for atomic force microscopy and medical applicationsJapanese Journal of Applied Physics 64 040804, 2025This article highlights how everyday tech from CD and Blu-ray players can be "hacked"...
Stratum corneum nanotexture feature detection using deep-learning and spatial analysis: a non-invasive tool for skin barrier assessment
Stratum corneum nanotexture feature detection using deep-learning and spatial analysis: a non-invasive tool for skin barrier assessmentGigaScience, 1-10, September 2024Building on prior work, this study improves atopic dermatitis severity assessment by refining...