Toward explainable AI approaches for breast imaging: adapting foundation models to diverse populations
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915632591667200 |
|---|---|
| author | Cavalcante, Guilherme J. Moreira, José Gabriel A. Nascimento, Gabriel A. B. do Dong, Vincent Nguyen, Alex Rêgo, Thaís G. do Malheiros, Yuri Filho, Telmo M. Silva Torrez, Carla R. Zeballos Gee, James C. McCarthy, Anne Marie Maidment, Andrew D. A. Barufaldi, Bruno |
| author_facet | Cavalcante, Guilherme J. Moreira, José Gabriel A. Nascimento, Gabriel A. B. do Dong, Vincent Nguyen, Alex Rêgo, Thaís G. do Malheiros, Yuri Filho, Telmo M. Silva Torrez, Carla R. Zeballos Gee, James C. McCarthy, Anne Marie Maidment, Andrew D. A. Barufaldi, Bruno |
| contents | Foundation models hold promise for specialized medical imaging tasks, though their effectiveness in breast imaging remains underexplored. This study leverages BiomedCLIP as a foundation model to address challenges in model generalization. BiomedCLIP was adapted for automated BI-RADS breast density classification using multi-modality mammographic data (synthesized 2D images, digital mammography, and digital breast tomosynthesis). Using 96,995 images, we compared single-modality (s2D only) and multi-modality training approaches, addressing class imbalance through weighted contrastive learning. Both approaches achieved similar accuracy (multi-modality: 0.74, single-modality: 0.73), with the multi-modality model offering broader applicability across different imaging modalities and higher AUC values consistently above 0.84 across BI-RADS categories. External validation on the RSNA and EMBED datasets showed strong generalization capabilities (AUC range: 0.80-0.93). GradCAM visualizations confirmed consistent and clinically relevant attention patterns, highlighting the models interpretability and robustness. This research underscores the potential of foundation models for breast imaging applications, paving the way for future extensions for diagnostic tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_17828 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Toward explainable AI approaches for breast imaging: adapting foundation models to diverse populations Cavalcante, Guilherme J. Moreira, José Gabriel A. Nascimento, Gabriel A. B. do Dong, Vincent Nguyen, Alex Rêgo, Thaís G. do Malheiros, Yuri Filho, Telmo M. Silva Torrez, Carla R. Zeballos Gee, James C. McCarthy, Anne Marie Maidment, Andrew D. A. Barufaldi, Bruno Computer Vision and Pattern Recognition Artificial Intelligence Foundation models hold promise for specialized medical imaging tasks, though their effectiveness in breast imaging remains underexplored. This study leverages BiomedCLIP as a foundation model to address challenges in model generalization. BiomedCLIP was adapted for automated BI-RADS breast density classification using multi-modality mammographic data (synthesized 2D images, digital mammography, and digital breast tomosynthesis). Using 96,995 images, we compared single-modality (s2D only) and multi-modality training approaches, addressing class imbalance through weighted contrastive learning. Both approaches achieved similar accuracy (multi-modality: 0.74, single-modality: 0.73), with the multi-modality model offering broader applicability across different imaging modalities and higher AUC values consistently above 0.84 across BI-RADS categories. External validation on the RSNA and EMBED datasets showed strong generalization capabilities (AUC range: 0.80-0.93). GradCAM visualizations confirmed consistent and clinically relevant attention patterns, highlighting the models interpretability and robustness. This research underscores the potential of foundation models for breast imaging applications, paving the way for future extensions for diagnostic tasks. |
| title | Toward explainable AI approaches for breast imaging: adapting foundation models to diverse populations |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2511.17828 |