Exportar Publicação

Sarwar, F., Garrido, N., Sebastião, P. & Silveira, M. (2025). Enhanced multiple instance learning for breast cancer detection in mammography: Adaptive patching, advanced pooling, and deep supervision. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. IEEE. https://doi.org/10.1109/EMBC58623.2025.11254317

F. Sarwar et al.,  "Enhanced multiple instance learning for breast cancer detection in mammography: Adaptive patching, advanced pooling, and deep supervision", in Proc. of the Annu. Int. Conf. of the IEEE Engineering in Medicine and Biology Society, EMBS, Copenhagen, Denmark, IEEE, 2025

@inproceedings{sarwar2025_1790072726525,
	author = "Sarwar, F. and Garrido, N. and Sebastião, P. and Silveira, M.",
	title = "Enhanced multiple instance learning for breast cancer detection in mammography: Adaptive patching, advanced pooling, and deep supervision",
	booktitle = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",
	year = "2025",
	editor = "",
	volume = "",
	number = "",
	series = "",
	doi = "10.1109/EMBC58623.2025.11254317",
	publisher = "IEEE",
	address = "Copenhagen, Denmark",
	organization = "Eng Med Biol Soc",
	url = "https://ieeexplore.ieee.org/document/11254317"
}

TY  - CPAPER
TI  - Enhanced multiple instance learning for breast cancer detection in mammography: Adaptive patching, advanced pooling, and deep supervision
T2  - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
AU  - Sarwar, F.
AU  - Garrido, N.
AU  - Sebastião, P.
AU  - Silveira, M.
PY  - 2025
SN  - 2375-7477
DO  - 10.1109/EMBC58623.2025.11254317
CY  - Copenhagen, Denmark
UR  - https://ieeexplore.ieee.org/document/11254317
AB  - This paper addresses the challenge of weakly supervised learning for breast cancer detection in mammography by introducing an Enhanced Embedded Space MI-Net model with deep supervision. The framework integrated adaptive patch creation, convolution feature extraction, and pooling methods -max, mean, log-sum-expo, attention, and gated attention pooling - evaluated in three MIL models, Instance Space mi-Net, Embedded Space MI-Net and Enhanced Embedded Space MI-Net. A key contribution is the incorporation of deep supervision, improving feature learning across network layers and enhancing bag-level classification performance. Experimental results on the CBIS / DDSM dataset demonstrate that the Enhanced MI-Net model achieves the highest AUC of 86% with attention pooling. This work addresses the gap in leveraging MIL techniques for high-resolution medical imaging without requiring detailed annotations, offering a robust and scalable solution for breast cancer detection.Clinical Relevance - This study highlights the potential of MIL-based models with attention pooling to accurately detect breast cancer in mammographic images without requiring detailed ROI annotations, offering a scalable and efficient diagnostic tool for clinical practice.
ER  -