Abstract
Current research in breast cancer Magnetic Resonance Imaging (MRI),especially with Artificial Intelligence (AI), faces challenges due to the lackof expert segmentations. To address this, we introduce the MAMA-MIA dataset,comprising 1506 multi-center dynamic contrast-enhanced MRI cases with expertsegmentations of primary tumors and non-mass enhancement areas. These caseswere sourced from four publicly available collections in The Cancer ImagingArchive (TCIA). Initially, we trained a deep learning model to automaticallysegment the cases, generating preliminary segmentations that significantlyreduced expert segmentation time. Sixteen experts, averaging 9 years ofexperience in breast cancer, then corrected these segmentations, resulting inthe final expert segmentations. Additionally, two radiologists conducted avisual inspection of the automatic segmentations to support future qualitycontrol studies. Alongside the expert segmentations, we provide 49 harmonizeddemographic and clinical variables and the pretrained weights of the well-knownnnUNet architecture trained using the DCE-MRI full-images and expertsegmentations. This dataset aims to accelerate the development and benchmarkingof deep learning models and foster innovation in breast cancer diagnostics andtreatment planning.