Multicenter Morphometric Analysis of Stratum Corneum Nanotexture for Skin Barrier Assessment

Jen-Hung Wang, Chia-Yu Chu, Felipe Colombelli, Ching-Wen Du, Maria Oberländer Christensen, Jorge Pereda, Ivone Jakasa, Sanja Kezic, Jacob Pontoppidan Thyssen, Edwin.Hwu, Gisele Miranda

Keywords: Atopic Dermatitis, Object Detection, Instance Segmentation, Morphometric Analysis, Stratum Corneum Nanotexture

TL;DR: We propose and validate a segmentation-based morphometric analysis pipeline on a multicenter stratum corneum nanotexture dataset to quantify skin barrier impairment in atopic dermatitis.

Abstract: Stratum corneum nanotexture (SCN) has emerged as a promising non-invasive biomarker for quantifying skin barrier impairment and the severity of inflammatory skin diseases such as atopic dermatitis (AD). In this multicenter study, we analyzed stratum corneum tape-strip samples from 90 patients with AD and 30 healthy controls recruited in Taiwan and Denmark, yielding a heterogeneous dataset of more than 2,000 SCN images. Participants were evenly stratified into four AD severity groups defined by the Eczema Area and Severity Index (EASI), enabling robust evaluation of SCN-derived metrics across the full spectrum of disease severity. Previous studies have primarily relied on count-based measures to quantify the density of circular nano-size objects (CNOs) in SCN images from single-center cohorts, without leveraging instance-level segmentation or comprehensive morphometric profiling. In this study, we propose and validate a segmentation-based SCN analysis pipeline that integrates YOLOv12 with Segment Anything Model 3 (SAM3) for accurate CNO delineation in a multicenter setting. This framework enables the extraction of detailed morphometric descriptors and facilitates systematic evaluation of SCN-derived biomarkers for quantitative skin barrier assessment in AD.

Primary Subject Area: Detection and Diagnosis

Secondary Subject Area: Segmentation

Registration Requirement: Yes

Reproducibility: https://github.com/mirandaresearchlab/SCN-SAM

Visa & Travel: No

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Midl Latex Submission Checklist: Ensure no LaTeX errors during compilation., Replace NNN with your OpenReview submission ID., Includes \documentclass{midl}, \jmlryear{2026}, \jmlrworkshop, \jmlrvolume, \editors, and correct \bibliography command., Did not override options of the hyperref package., Did not use the times package., Use the correct spelling and format, avoid Unicode characters, and use LaTeX equivalents instead., Any math in the title and abstract must be enclosed within $...$., Did not override the bibliography style defined in midl.cls and did not use \begin{thebibliography} directly to insert references., Avoid using \scalebox; use \resizebox when needed., Included all necessary figures and removed *unused* files in the zip archive., Removed special formatting, visual annotations, and highlights used during rebuttal., All special characters in the paper and .bib file use LaTeX commands (e.g., \'e for é)., No separate supplementary PDF uploads., Acknowledgements, references, and appendix must start after the main content.

Submission Number: 23