Objective Rosacea Erythema Severity Assessment with Multimodal AI Framework

A retrospective cohort of 1,001 patients with rosacea was analyzed using standardized VISIA® imaging. The dataset supports an AI framework integrating handcrafted image-derived tabular features with deep learning representations for objective Clinician’s Erythema Assessment grading. It was created by Jun Zhang and last updated in May 2026.

Use Cases

• Training machine learning models for objective rosacea severity grading based on image-derived tabular features.
• Benchmarking multimodal AI frameworks against expert dermatologist consensus assessments.
• Developing interpretable AI tools for clinical dermatology and teledermatology applications.

Strengths

• Dataset is based on a cohort of 1,001 patients, providing a substantial sample size.
• Reference standard established by unanimous consensus from three expert dermatologists.
• The described AI model achieved a Macro-AUC of 0.902 and a Quadratic Weighted Kappa of 0.800.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect bias inherent to the single-source, retrospective collection method.

Provenance

Source

figshare

Collection Method

Retrospective analysis of patient imaging data with expert consensus labeling.

Freshness

Last updated 2026-05-11 05:26:03; freshness should be verified.