flp-r/breast-mri-segmentation-dataset

Dataset & Preprocessing

This repository contains the refined ground-truth vessel masks and registered breast DCE-MRI dataset as an attachment to the Master's Thesis: Characterization of Breast Tissue in Dynamic MR Scans.

🌐 The Data Source

Duke Breast Cancer MRI dataset (available on TCIA), used under the CC BY-NC 4.0 license.

⚙️ Image Registration

DCE-MRI scans were co-registered using the implementation adopted from MR breast co-registration and Segmentation .

🎯 Vessel GT Annotations

The dataset includes blood vessel segmentation masks. These masks underwent manual refinement. The baseline annotations were adopted from the study by Lew et al. (hosted on the Duke dataset).

📝 References

Duke Breast Cancer MRI dataset (CC BY-NC 4.0):

Saha, A., Harowicz, M. R., Grimm, L. J., Weng, J., Cain, E. H., Kim, C. E., Ghate, S. V., Walsh, R., & Mazurowski, M. A. (2021). Dynamic contrast-enhanced magnetic resonance images of breast cancer patients with tumor locations [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/TCIA.e3sv-re93

Image Registration:

JAKUBICEK, Roman; OUREDNICEK, Petr; ENGELE, Ilze; KUZMINSKIS, Maris; CERVENKOVA, Jana et al., 2026. Towards Improved Breast MRI Analysis: Co-registration and Segmentation Model for Oncological Applications. Online. IFMBE Proceedings. Roč. 2026, pp. 1045-1053. ISBN 9783032065308. ISSN 1680-0737. Available at: https://doi.org/10.1007/978-3-032-06531-5_94. [accessed 2026-05-18].

Baseline Vessel Annotations:

LEW, Christopher O.; HAROUNI, Majid; KIRKSEY, Ella R.; KANG, Elianne J.; DONG, Haoyu et al., 2024. A publicly available deep learning model and dataset for segmentation of breast, fibroglandular tissue, and vessels in breast MRI. Scientific Reports, vol. 14, issue 1. ISSN 2045-2322. Available at: https://doi.org/10.1038/s41598-024-54048-2.