BreastDCEDL_AMBL: Curated Breast DCE-MRI Dataset with lesion segmentation, malignant and bening

Abstract

BreastDCEDL_AMBL DCE MRI Dataset 88 patients 132 lesions 78 Malignant lesions 54 Bening lesions Full segmentation Files: BreastDCEDL_AMBL_paitents.csv - paitent's table BrestDCEDL_AMBL_dce.tar.gz - 3D niftii files of dynamic MRI scans sus_tum.tar.gz - 3D niftii files of lesion segmentation, malignant and bening segformer_best_epoch130_0.5786.pth - trained model weights If you find the data usefull, pleadr cite: @misc{fridman2025transformerclassificationbreastlesions, title={Transformer Classification of Breast Lesions: The BreastDCEDL_AMBL Benchmark Dataset and 0.92 AUC Baseline}, author={Naomi Fridman and Anat Goldstein}, year={2025}, eprint={2509.26440}, archivePrefix={arXiv}, primaryClass={cs.AI}, url={https://arxiv.org/abs/2509.26440}, }