Related Experiment Video
Updated: Aug 6, 2026
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Quantitative CRFF-OCT Imaging Features for Characterization of Disease Activity in Non-Segmental Vitiligo: A Machine
Chau Yee Ng1,2,3,4, I-Ling Chen5, Yi-Ting Chen5
1Department of Dermatology, Chang Gung Memorial Hospital, Linkou Main Branch, Taoyuan, Taiwan.
Experimental Dermatology
|July 22, 2026
Summary
Cellular-resolution full-field optical coherence tomography (CRFF-OCT) combined with machine learning can differentiate active from stable vitiligo lesions. This imaging approach offers objective assessment for vitiligo disease activity and monitoring.
Area of Science:
- Dermatology
- Medical Imaging
- Artificial Intelligence
Background:
- Vitiligo is an autoimmune skin condition causing melanocyte loss and unpredictable activity.
- Assessing vitiligo disease activity, especially with subtle changes, is clinically challenging.
- Cellular-resolution full-field optical coherence tomography (CRFF-OCT) provides high-resolution, histology-like skin imaging.
Purpose of the Study:
- To evaluate the feasibility of using CRFF-OCT with machine learning for characterizing active and stable vitiligo lesions.
- To develop an imaging-based quantitative analysis for vitiligo activity assessment.
- To explore CRFF-OCT's potential in monitoring vitiligo progression.
Main Methods:
- Prospective enrollment of 50 non-segmental vitiligo patients.
- CRFF-OCT imaging of lesional, perilesional, and normal skin.
- Extraction of quantitative features (epidermal structure, DEJ morphology, reflectivity) and development of a machine learning-assisted computer-aided detection (CADe) framework.
Main Results:
- CRFF-OCT revealed distinct microstructural patterns in active versus stable vitiligo lesions.
- Significant differences observed in basal epidermal pigment-associated reflectivity and lesion-to-normal reflectivity ratios.
- Quantitative analysis showed significant variations in epidermal thickness, DEJ reflectivity, and reflectivity heterogeneity.
- The CADe framework achieved 80.6% accuracy in classifying lesion activity using a support vector machine model.
Conclusions:
- CRFF-OCT-based quantitative imaging is feasible for objective vitiligo lesion characterization.
- This technology supports potential applications in assessing vitiligo disease activity.
- CRFF-OCT shows promise for longitudinal monitoring of vitiligo patients.
