Abstract
Segmentation of breast MRI in a longitudinal high-risk cohort is challenging due to considerable variability in image quality and artefacts, imaging protocols, scanner types and field strengths, and not least anatomy. Automated segmentation of the whole breast is a prerequisite for quantitative tissue characterization in early detection as well as image-based risk modeling and predictive machine learning. We investigated the behavior of a 2D U-net architecture when being faced with a large set of error-prone training input and compared the results to the training on a smaller but human-corrected high-quality annotation set, both when being used for fine-tuning and training from scratch. Our dataset consists of a total of 876 pre-contrast axial T1 weighted MRI volumes from 166 subjects acquired between 2006 and 2021 in a longitudinal high-risk breast cancer screening cohort study. All images were previously segmented using a fully automated heuristic algorithm, resulting in error-prone segmentation masks, whichwere used in an initial ‘human-free’ experiment. We randomly separated on a per-subject level 102 volumes from 23 subjects, for which an expert radiologist manually corrected these segmentation masks, providing the basis for the additional two experiments. Our results indicate a subclass of input errors can be compensated for by the regularization capacity of standard deep convolutional neural networks while other errors are learnt or even newly introduced. In particular, tissue boundaries in axial directions were missed in our experiment. In the inner region, the median Dice coefficient of our fine-tuning U-net exceeded 98% at a reasonable robustness and consistency, which is promising given the simplicity of our approach. Future work will address efficient learning schemes, aiming at boosting the segmentation quality with minimal human input, and the boundary issue.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ma X, Wang J, Zheng X, et al. Automated fibroglandular tissue segmentation in breast MRI using generative adversarial networks. Phys Med Biol. 2020;65(10):105006.
Doran SJ, Hipwell JH, Denholm R, et al. Breast MRI segmentation for density estimation: do different methods give the same results and how much do differences matter? Med Phys. 2017;44(9):4573–92.
Niell BL, Abdalah M, Stringfield O, et al. Quantitative measures of background parenchymal enhancement predict breast cancer risk. Am J Radiol. 2021;217(1):64–75.
Wang L, Platel B, Ivanovskaya T, et al. Fully automatic breast segmentation in 3D breast MRI. Proc. IEEE Int Symposium Biomed Imaging (ISBI). 2012:1024–27.
Pandey D, Yin X,Wang H, et al. Automatic and fast segmentation of breast region-of-interest (ROI) and density in MRIs. Heliyon. 2018;4(12):e01042.
Thakran S, Chatterjee S, Singhal M, et al. Automatic outer and inner breast tissue segmentation using multi-parametric MRI images of breast tumor patients. PLoS One. 2018;13(1):e0190348.
Giannini V, Vignati A, Morra L, et al. A fully automatic algorithm for segmentation of the breasts in DCE-MR images. Annu Int Conf IEEE Eng Med Biol Soc. 2010:3146–49.
Verburg E, Wolterink JM, Waard SN de, et al. Knowledge-based and deep learning-based automated chest wall segmentation in magnetic resonance images of extremely dense breasts. Med Phys. 2019;46(10):4405–16.
Fooladivanda A, Shokouhi SB, Ahmadinejad N. Localized-atlas-based segmentation of breast MRI in a decision-making framework. Australas Phys Eng Sci Med. 2017;40(1):69–84.
Lin M, Chen JH, Wang X, et al. Template-based automatic breast segmentation on MRI by excluding the chest region. Med Phys. 2013;40(12):122301.
Gubern-Mérida A, Kallenberg M, Martí R, Karssemeijer N. Segmentation of the pectoral muscle in breast MRI using atlas-based approaches. Proc. Med Image Comput Comput Assist Interv (MICCAI). Vol. 15. (Pt 2). Springer, 2012:371–78.
Wang Y, Morrell G, Heibrun ME, et al. 3D multi-parametric breast MRI segmentation using hierarchical support vector machine with coil sensitivity correction. Acad Radiol. 2013;20(2):137–47.
Zhang Y, Chan S, Chen JH, et al. Development of U-Net breast density segmentation method for Fat-Sat MR images using transfer learning based on Non-fat-sat model. J Digit Imaging. 2021;34(4):877–87.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 Der/die Autor(en), exklusiv lizenziert an Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature
About this paper
Cite this paper
Ambroladze, A. et al. (2023). CNN-based Whole Breast Segmentation in Longitudinal High-risk MRI Study. In: Deserno, T.M., Handels, H., Maier, A., Maier-Hein, K., Palm, C., Tolxdorff, T. (eds) Bildverarbeitung für die Medizin 2023. BVM 2023. Informatik aktuell. Springer Vieweg, Wiesbaden. https://doi.org/10.1007/978-3-658-41657-7_35
Download citation
DOI: https://doi.org/10.1007/978-3-658-41657-7_35
Published:
Publisher Name: Springer Vieweg, Wiesbaden
Print ISBN: 978-3-658-41656-0
Online ISBN: 978-3-658-41657-7
eBook Packages: Computer Science and Engineering (German Language)