Foundation Models in Dermatopathology: Skin Tissue Classification

TL;DR

This study evaluates foundation models UNI and Virchow2 for classifying dermatopathology whole-slide images, demonstrating that Virchow2's features improve classification accuracy and robustness in automated skin tissue diagnosis.

Contribution

It introduces the use of foundation models as feature extractors for WSI classification in dermatopathology, highlighting Virchow2's superior performance over UNI.

Findings

Virchow2 outperforms UNI in slide-level classification accuracy.

Logistic regression achieved 90% accuracy with Virchow2 features.

Data augmentation and normalization improve model robustness.

Abstract

The rapid generation of whole-slide images (WSIs) in dermatopathology necessitates automated methods for efficient processing and accurate classification. This study evaluates the performance of two foundation models, UNI and Virchow2, as feature extractors for classifying WSIs into three diagnostic categories: melanocytic, basaloid, and squamous lesions. Patch-level embeddings were aggregated into slide-level features using a mean-aggregation strategy and subsequently used to train multiple machine learning classifiers, including logistic regression, gradient-boosted trees, and random forest models. Performance was assessed using precision, recall, true positive rate, false positive rate, and the area under the receiver operating characteristic curve (AUROC) on the test set. Results demonstrate that patch-level features extracted using Virchow2 outperformed those extracted via UNI across most slide-level classifiers, with logistic regression achieving the highest accuracy (90%) for Virchow2, though the difference was not statistically significant. The study also explored data augmentation techniques and image normalization to enhance model robustness and generalizability. The mean-aggregation approach provided reliable slide-level feature representations. All experimental results and metrics were tracked and visualized using WandB.ai, facilitating reproducibility and interpretability. This research highlights the potential of foundation models for automated WSI classification, providing a scalable and effective approach for dermatopathological diagnosis while paving the way for future advancements in slide-level representation learning.