DBT-DINO: Towards Foundation model based analysis of Digital Breast Tomosynthesis
Authors:
Felix J. Dorfner,
Manon A. Dorster,
Ryan Connolly,
Oscar Gentilhomme,
Edward Gibbs,
Steven Graham,
Seth Wander,
Thomas Schultz,
Manisha Bahl,
Dania Daye,
Albert E. Kim,
Christopher P. Bridge
Abstract:
Foundation models have shown promise in medical imaging but remain underexplored for three-dimensional imaging modalities. No foundation model currently exists for Digital Breast Tomosynthesis (DBT), despite its use for breast cancer screening.
To develop and evaluate a foundation model for DBT (DBT-DINO) across multiple clinical tasks and assess the impact of domain-specific pre-training.
Sel…
▽ More
Foundation models have shown promise in medical imaging but remain underexplored for three-dimensional imaging modalities. No foundation model currently exists for Digital Breast Tomosynthesis (DBT), despite its use for breast cancer screening.
To develop and evaluate a foundation model for DBT (DBT-DINO) across multiple clinical tasks and assess the impact of domain-specific pre-training.
Self-supervised pre-training was performed using the DINOv2 methodology on over 25 million 2D slices from 487,975 DBT volumes from 27,990 patients. Three downstream tasks were evaluated: (1) breast density classification using 5,000 screening exams; (2) 5-year risk of developing breast cancer using 106,417 screening exams; and (3) lesion detection using 393 annotated volumes.
For breast density classification, DBT-DINO achieved an accuracy of 0.79 (95\% CI: 0.76--0.81), outperforming both the MetaAI DINOv2 baseline (0.73, 95\% CI: 0.70--0.76, p<.001) and DenseNet-121 (0.74, 95\% CI: 0.71--0.76, p<.001). For 5-year breast cancer risk prediction, DBT-DINO achieved an AUROC of 0.78 (95\% CI: 0.76--0.80) compared to DINOv2's 0.76 (95\% CI: 0.74--0.78, p=.57). For lesion detection, DINOv2 achieved a higher average sensitivity of 0.67 (95\% CI: 0.60--0.74) compared to DBT-DINO with 0.62 (95\% CI: 0.53--0.71, p=.60). DBT-DINO demonstrated better performance on cancerous lesions specifically with a detection rate of 78.8\% compared to Dinov2's 77.3\%.
Using a dataset of unprecedented size, we developed DBT-DINO, the first foundation model for DBT. DBT-DINO demonstrated strong performance on breast density classification and cancer risk prediction. However, domain-specific pre-training showed variable benefits on the detection task, with ImageNet baseline outperforming DBT-DINO on general lesion detection, indicating that localized detection tasks require further methodological development.
△ Less
Submitted 15 December, 2025;
originally announced December 2025.
The R-index: A universal metric for evaluating OAM content and mode purity in optical fields
Authors:
Monika Bahl,
Georgios M. Koutentakis,
Mikhail Maslov,
Tom Jungnickel,
Timo Gassen,
Mikhail Lemeshko,
Oliver H. Heckl
Abstract:
Despite its pivotal role in optical manipulation, high-capacity communications, and quantum information, a general measure of orbital angular momentum (OAM) in structured light remains elusive. In optical fields, where multiple vortices coexist, the local nature of vortex OAM and the absence of a common rotation axis make the total OAM of the field difficult to quantify. Here, we introduce the R-i…
▽ More
Despite its pivotal role in optical manipulation, high-capacity communications, and quantum information, a general measure of orbital angular momentum (OAM) in structured light remains elusive. In optical fields, where multiple vortices coexist, the local nature of vortex OAM and the absence of a common rotation axis make the total OAM of the field difficult to quantify. Here, we introduce the R-index, a metric that captures the intrinsic OAM content of any structured optical field, from pure Laguerre-Gaussian modes to arbitrary multi-vortex superpositions. Not only does this metric quantify the total OAM, it also assesses field purity, providing insight into the fidelity and robustness of the OAM generation. By unifying OAM characterization into a single figure of merit, the R-index enables direct comparison across diverse beam profiles and facilitates the identification of optimal configurations for both foundational studies and applied technologies.
△ Less
Submitted 18 August, 2025;
originally announced August 2025.