Breast Cancer Classification in Deep Ultraviolet Fluorescence Images Using a Patch-Level Vision Transformer Framework

TL;DR

This paper presents a novel patch-level vision transformer framework for classifying breast cancer in deep ultraviolet fluorescence images, achieving high accuracy and interpretability in intraoperative margin assessment.

Contribution

It introduces a ViT-based classification method with saliency weighting for improved accuracy and interpretability in breast cancer detection using DUV fluorescence images.

Findings

Achieved 98.33% classification accuracy.

Outperformed conventional deep learning methods.

Enhanced interpretability with Grad-CAM++ saliency maps.

Abstract

Breast-conserving surgery (BCS) aims to completely remove malignant lesions while maximizing healthy tissue preservation. Intraoperative margin assessment is essential to achieve a balance between thorough cancer resection and tissue conservation. A deep ultraviolet fluorescence scanning microscope (DUV-FSM) enables rapid acquisition of whole surface images (WSIs) for excised tissue, providing contrast between malignant and normal tissues. However, breast cancer classification with DUV WSIs is challenged by high resolutions and complex histopathological features. This study introduces a DUV WSI classification framework using a patch-level vision transformer (ViT) model, capturing local and global features. Grad-CAM++ saliency weighting highlights relevant spatial regions, enhances result interpretability, and improves diagnostic accuracy for benign and malignant tissue classification. A comprehensive 5-fold cross-validation demonstrates the proposed approach significantly outperforms conventional deep learning methods, achieving a classification accuracy of 98.33%.