Artificial Intelligence (AI) research in breast cancer Magnetic Resonance Imaging (MRI) faces challenges due to limited expert-labeled segmentations. To address this, we present a multicenter dataset of 1506 pre-treatment T1-weighted dynamic contrast-enhanced MRI cases, including expert annotations of primary tumors and non-mass-enhanced regions. The dataset integrates imaging data from four collections in The Cancer Imaging Archive (TCIA), where only 163 cases with expert segmentations were initially available. To facilitate the annotation process, a deep learning model was trained to produce preliminary segmentations for the remaining cases. These were subsequently corrected and verified by 16 breast cancer experts (averaging 9 years of experience), creating a fully annotated dataset. Additionally, the dataset includes 49 harmonized clinical and demographic variables, as well as pre-trained weights for a baseline nnU-Net model trained on the annotated data. This resource addresses a critical gap in publicly available breast cancer datasets, enabling the development, validation, and benchmarking of advanced deep learning models, thus driving progress in breast cancer diagnostics, treatment response prediction, and personalized care.
View on arXiv@article{garrucho2025_2406.13844,
title={ A large-scale multicenter breast cancer DCE-MRI benchmark dataset with expert segmentations },
author={ Lidia Garrucho and Kaisar Kushibar and Claire-Anne Reidel and Smriti Joshi and Richard Osuala and Apostolia Tsirikoglou and Maciej Bobowicz and Javier del Riego and Alessandro Catanese and Katarzyna Gwoździewicz and Maria-Laura Cosaka and Pasant M. Abo-Elhoda and Sara W. Tantawy and Shorouq S. Sakrana and Norhan O. Shawky-Abdelfatah and Amr Muhammad Abdo-Salem and Androniki Kozana and Eugen Divjak and Gordana Ivanac and Katerina Nikiforaki and Michail E. Klontzas and Rosa García-Dosdá and Meltem Gulsun-Akpinar and Oğuz Lafcı and Ritse Mann and Carlos Martín-Isla and Fred Prior and Kostas Marias and Martijn P.A. Starmans and Fredrik Strand and Oliver Díaz and Laura Igual and Karim Lekadir },
journal={arXiv preprint arXiv:2406.13844},
year={ 2025 }
}