Interpretable Automatic Rosacea Detection with Whitened Cosine Similarity

2025-04-10

AI Summary

To address the dual challenges of insufficient accuracy and poor clinical interpretability in automated rosacea detection, this paper proposes an interpretable discriminative method based on whitened cosine similarity. The method extracts deep features using ResNet, applies a whitening transformation to the feature space to eliminate redundant correlations, and then computes whitened cosine similarities between each test sample and the class-mean embeddings of rosacea and normal skin. This enables high-confidence classification with transparent decision logic. Our key contribution is the first introduction of a whitened cosine similarity metric framework, which achieves statistically significant performance gains over conventional deep learning and statistical baselines on unseen test sets (p < 0.01). Crucially, the approach provides intuitive, traceable decision evidence—enhancing clinician–patient communication—while maintaining diagnostic accuracy. This work bridges the gap between algorithmic performance and clinical utility, facilitating early rosacea screening and improving public health awareness.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Transparent, Interpretable, Explainable ML
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Responsible Web: Algorithmic accountability and transparency on the web
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

According to the National Rosacea Society, approximately sixteen million Americans suffer from rosacea, a common skin condition that causes flushing or long-term redness on a person's face. To increase rosacea awareness and to better assist physicians to make diagnosis on this disease, we propose an interpretable automatic rosacea detection method based on whitened cosine similarity in this paper. The contributions of the proposed methods are three-fold. First, the proposed method can automatically distinguish patients suffering from rosacea from people who are clean of this disease with a significantly higher accuracy than other methods in unseen test data, including both classical deep learning and statistical methods. Second, the proposed method addresses the interpretability issue by measuring the similarity between the test sample and the means of two classes, namely the rosacea class versus the normal class, which allows both medical professionals and patients to understand and trust the results. And finally, the proposed methods will not only help increase awareness of rosacea in the general population, but will also help remind patients who suffer from this disease of possible early treatment, as rosacea is more treatable in its early stages. The code and data are available at https://github.com/chengyuyang-njit/ICCRD-2025. The code and data are available at https://github.com/chengyuyang-njit/ICCRD-2025.

Problem

• Detect rosacea automatically with high accuracy
• Improve interpretability using class similarity measures
• Increase awareness and enable early rosacea treatment

Innovation

• Uses whitened cosine similarity for detection
• Improves accuracy over traditional methods
• Enhances interpretability with class similarity measures