A large-scale multicenter breast cancer DCE-MRI benchmark dataset with expert segmentations

TL;DR

This paper introduces a comprehensive, expert-annotated multicenter breast cancer MRI dataset with over 1500 cases, including clinical data and baseline models, to advance AI research in diagnosis and treatment.

Contribution

It provides a large-scale, fully annotated breast cancer MRI dataset with expert corrections, clinical variables, and baseline models, filling a critical gap for AI development.

Findings

The dataset includes 1506 cases with expert annotations.

Baseline nnU-Net model trained on the dataset.

Includes 49 clinical and demographic variables.

Abstract

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.