Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LUMINA: A Multi-Vendor Mammography Benchmark with Energy Harmonization Protocol

Hongyi Pan, Gorkem Durak, Halil Ertugrul Aktas, Andrea M. Bejar, Baver Tutun, Emre Uysal, Ezgi Bulbul, Mehmet Fatih Dogan, Berrin Erok, Berna Akkus Yildirim, Sukru Mehmet Erturk, Ulas Bagci

Abstract

Publicly available full-field digital mammography (FFDM) datasets remain limited in size, clinical annotations, and vendor diversity, hindering the development of robust models. We introduce LUMINA, a curated, multi-vendor FFDM dataset that explicitly encodes acquisition energy and vendor metadata to capture clinically relevant appearance variations often overlooked in existing benchmarks. This dataset contains 1824 images from 468 patients (960 benign, 864 malignant), with pathology-confirmed labels, BI-RADS assessments, and breast-density annotations. LUMINA spans six acquisition systems and includes both high- and low-energy imaging styles, enabling systematic analysis of vendor- and energy-induced domain shifts. To address these variations, we propose a foreground-only pixel-space alignment method (''energy harmonization'') that maps images to a low-energy reference while preserving lesion morphology. We benchmark CNN and transformer models on three clinically relevant tasks: diagnosis (benign vs. malignant), BI-RADS classification, and density estimation. Two-view models consistently outperform single-view models. EfficientNet-B0 achieves an AUC of 93.54% for diagnosis, while Swin-T achieves the best macro-AUC of 89.43% for density prediction. Harmonization improves performance across architectures and produces more localized Grad-CAM responses. Overall, LUMINA provides (1) a vendor-diverse benchmark and (2) a model-agnostic harmonization framework for reliable and deployable mammography AI.

LUMINA: A Multi-Vendor Mammography Benchmark with Energy Harmonization Protocol

Abstract

Publicly available full-field digital mammography (FFDM) datasets remain limited in size, clinical annotations, and vendor diversity, hindering the development of robust models. We introduce LUMINA, a curated, multi-vendor FFDM dataset that explicitly encodes acquisition energy and vendor metadata to capture clinically relevant appearance variations often overlooked in existing benchmarks. This dataset contains 1824 images from 468 patients (960 benign, 864 malignant), with pathology-confirmed labels, BI-RADS assessments, and breast-density annotations. LUMINA spans six acquisition systems and includes both high- and low-energy imaging styles, enabling systematic analysis of vendor- and energy-induced domain shifts. To address these variations, we propose a foreground-only pixel-space alignment method (''energy harmonization'') that maps images to a low-energy reference while preserving lesion morphology. We benchmark CNN and transformer models on three clinically relevant tasks: diagnosis (benign vs. malignant), BI-RADS classification, and density estimation. Two-view models consistently outperform single-view models. EfficientNet-B0 achieves an AUC of 93.54% for diagnosis, while Swin-T achieves the best macro-AUC of 89.43% for density prediction. Harmonization improves performance across architectures and produces more localized Grad-CAM responses. Overall, LUMINA provides (1) a vendor-diverse benchmark and (2) a model-agnostic harmonization framework for reliable and deployable mammography AI.
Paper Structure (13 sections, 8 equations, 10 figures, 10 tables, 1 algorithm)

This paper contains 13 sections, 8 equations, 10 figures, 10 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Works
  3. LUMINA Dataset Introduction
  4. Evaluation Protocol
  5. Pixel-Space Harmonization
  6. Deep Learning Approaches for Multi-View Mammography Classification
  7. Experimental Results
  8. Experimental Setup
  9. Benchmark Results
  10. Ablation Study And Discussion
  11. Conclusion
  12. Data and Code
  13. Acknowledgment

Figures (10)

  • Figure 1: Representative benign and malignant mammograms.
  • Figure 2: Age, BI-RADS, and breast distribution.
  • Figure 3: LUMINA pipeline. The EfficientNet-B0 backbone can be replaced by other backbones (ResNet-50, DenseNet-121, and Swin-T). Two‑view shared‑backbone reached accuracy comparable to independent‑backbone with 48% less parameters (Table \ref{['tab:pathology ablation']}) and outperformed four‑view variants (Table \ref{['tab:pathology']}).
  • Figure 4: Background influence. Standard histogram matching is degraded by large black background regions, whereas foreground-only histogram matching remains unaffected.
  • Figure 5: t-SNE visualization for diagnosis task. Higher resolution increases benign/malignant separation in the latent space.
  • ...and 5 more figures