Unsupervised Domain Adaptation for Calcification Classification in Mammography (arXiv 2507.06826)

Forum topic · 小凯 · 2026-07-09

Summary

A 2025 arXiv paper (2507.06826) by Xuan Liu, Derek L. Nguyen, and Emily C. Barre proposes a calcification classification framework to improve malignant versus benign breast disease classification across multi-site mammography datasets. The framework combines two components: an unsupervised domain adaptation module based on style transfer models (AdaIN and CycleGAN) that generates vendor-specific and technique-specific training samples without additional annotations, and a supervised classification module using Swin Transformer V2 as the backbone. The method was evaluated on three datasets: cross-validation on OPTIMAM (UK NHS; n=2994), with external validation on EMBE (Emory University; n=125) and the Duke Calcification Dataset v1 (n=788), covering multiple vendors including full-field digital mammography and synthetic 2D images from digital breast tomosynthesis. Domain adaptation improved cross-site performance, raising AUC from 0.68 to 0.72 on EMBED and from 0.68 to 0.73 on the Duke dataset, showing that unsupervised domain adaptation can reduce domain shift and improve generalization for calcification classification across multi-site mammography data.

Paper Overview

Field: Computer Vision Authors: Xuan Liu, Derek L. Nguyen, Emily C. Barre Published: 2025-07-09 arXiv: 2507.06826

Abstract

Deep learning-based computer-aided diagnosis (CAD) systems have shown strong performance in breast cancer diagnosis, particularly for classification tasks in mammography. However, domain shifts across multi-site datasets remain a challenge, especially when models are applied to unseen domains.

In this work, the authors propose a calcification classification framework to improve malignant versus benign breast disease classification across multi-site mammography datasets. The framework consists of two components:

1. Unsupervised domain adaptation module based on style transfer models (AdaIN and CycleGAN) to generate vendor-specific and technique-specific training samples without additional annotations. 2. Supervised classification module using Swin Transformer V2 as the backbone.

Evaluation

The method was evaluated on three datasets:

OPTIMAM (UK National Health Service; n=2994) — used for cross-validation
EMBED (Emory University; n=125) — external validation
Duke Calcification Dataset v1 (n=788) — external validation

These datasets cover multiple vendors, including full-field digital mammography and synthetic 2D images derived from digital breast tomosynthesis.

Results

The proposed framework improved cross-site performance:

EMBED: AUC 0.68 → 0.72
Duke Calcification Dataset: AUC 0.68 → 0.73

These findings suggest that domain adaptation can reduce domain shift and improve the generalization of calcification classification across multi-site datasets.

---

*Auto-collected on 2026-07-09.*