Skip to content
Preprint

BreastMammo and DenseMammo: Benchmarks for Mammography Domain Generalization

Aug 2026 · 0 citations · 31 references
Engineering Computer Science

TL;DR

A domain generalization framework that utilizes a foreground-only histogram matching protocol to resolve the domain shift issue arising from disparate clinical sources is proposed, significantly outperforming prominent domain generalization paradigms, including MixStyle and Discrete-Fourier-Transform-based frameworks.

Abstract

Breast density classification is a critical component of breast cancer risk assessment, yet AI models often struggle to generalize across clinical sites due to vendor-specific acquisition styles. In this work, we introduce two new datasets, BreastMammo and DenseMammo, to facilitate robust multi-view mammography research. We propose a domain generalization framework that utilizes a foreground-only histogram matching protocol to resolve the domain shift issue arising from disparate clinical sources. Internal evaluation using a 5-fold cross-validation protocol demonstrates the efficacy of our approach, with the Swin Transformer backbone achieving a peak AUC of 98.32% for density classification. External evaluation on the TNMammo and LUMINA datasets demonstrates that the proposed approach consistently reduces domain shift, significantly outperforming prominent domain generalization paradigms, including MixStyle and Discrete-Fourier-Transform-based frameworks.

View source

Similar papers

#artificial intelligence Open access Sep 2026

Feature Space Guidance for Breast Cancer Classification in DCE-MRI

The proposed framework substantially outperforms finetuned radiology foundation models and prior methods, improving mean AUROC by nearly 8 points and balanced accuracy by 4 points over the strongest baseline, further demonstrating its effectiveness for robust breast cancer classification.

Benjamin Hamm, Y. Kirchhoff, M. Rokuss et al. · 1 citation
Open access Aug 2026

Transfer Learning Model for Breast Cancer Detection Using Mammograms from Low-Resource Settings

The study shows that transfer learning acts as a strong base for AI-assisted mammography triage systems in low-resource environments when patient level assessment continues throughout system development.

Zinat Abdulkadiri, Muhammad A. Suleiman, Joshua Abah · 0 citations
Conference Aug 2026

Adaptive Edge-Aware Breast Cancer Classification from Mammograms using Explainable Density-Guided Filtering with Transformer-Enhanced Deep Unfolding Network and Hybrid Optimization

Breast cancer (BC) remains one of the foremost causes of cancer-related mortality among women globally, and early detection through mammography screening is essential for improving patient survival outcomes. Although recent deep learning frameworks have demonstrated encouraging performance in automated mammogram classi...

P. Revathi, B. Vigneshwaran · 0 citations
Open access Aug 2026

Breast Cancer Classification in Ultrasound Images Using Two-Phase EfficientNetB7 Transfer Learning

A systematic but innovative two phases transfer learning’sbased deep learning classification framework is developed using popular EfficientNetB7 architecture architecture for breast cancer classification which encompasses formalisation into components like data augmentation, progressive fine-tuning and adaptive learnin...

M. Kavya, G. Thirupati · 0 citations
Open access Sep 2026

BCSeg-IRUNet: A Hybrid Inception–Residual Encoder–Decoder for Mammographic Mass Segmentation

Breast cancer accounted for approximately 2.4 million new cases and 694,000 deaths worldwide in 2024. Deep learning approaches, particularly encoder–decoder architectures, have been investigated for mammographic mass segmentation. However, evaluations are commonly centered on benchmarked datasets, which include only an...

Fabian Cienfuegos-Caraveo, A. Guzmán-Pando, G. Ramírez-Alonso et al. · 0 citations
Open access Oct 2026

The Reproducibility Gap in Mammographic CAD: Quantifying Patient-Level Leakage Effects on CBIS-DDSM

This study conducted a systematic comparison of three evaluation protocols on the CBIS-DDSM benchmark using EfficientNetB3 under strictly identical architecture, hyperparameters, and augmentation strategies, establishing a methodologically valid benchmark for mammographic mass classification on CBIS-DDSM.

Chaima Elmejgari, Younes Nadir, M. Qbadou · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.