Transfer learning for breast cancer malignancy classification based on dynamic contrast-enhanced MR images

Haarburger C, Langenberg P, Truhn D, Schneider H, Thüring J, Schrading S, Kuhl CK, Merhof D (2018)

Publication Type: Conference contribution

Publication year: 2018

Journal

Informatik aktuell Gesellschaft für Informatik (GI)

Publisher: Springer Science and Business Media Deutschland GmbH

Book Volume: 0

Pages Range: 216-221

Conference Proceedings Title: Informatik aktuell

Event location: Erlangen, DEU

ISBN: 9783540295945

DOI: 10.1007/978-3-662-56537-7_61

Abstract

In clinical contexts with very limited annotated data, such as breast cancer diagnosis, training state-of-the art deep neural networks is not feasible. As a solution, we transfer parameters of networks pretrained on natural RGB images to malignancy classification of breast lesions in dynamic contrast-enhanced MR images. Since DCE-MR images comprise several contrasts and timepoints, a direct finetuning of pretrained networks expecting three input channels is not possible. Based on the hypothesis that a subset of the acquired image data is sufficient for a computer-aided diagnosis, we provide an experimental comparison of all possible subsets of MR image contrasts and determine the best combination for malignancy classification. A subset of images acquired at three timepoints of dynamic T1-weighted images which closely corresponds to human interpretation performs best with an AUC of 0.839.

Involved external institutions

Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen DE Germany (DE) Universitätsklinikum Aachen (UKA) DE Germany (DE)

How to cite

APA:

Haarburger, C., Langenberg, P., Truhn, D., Schneider, H., Thüring, J., Schrading, S.,... Merhof, D. (2018). Transfer learning for breast cancer malignancy classification based on dynamic contrast-enhanced MR images. In Andreas Maier, Thomas M. Deserno, Heinz Handels, Klaus H. Maier-Hein, Christoph Palm, Thomas Tolxdorff (Eds.), Informatik aktuell (pp. 216-221). Erlangen, DEU: Springer Science and Business Media Deutschland GmbH.

MLA:

Haarburger, Christoph, et al. "Transfer learning for breast cancer malignancy classification based on dynamic contrast-enhanced MR images." Proceedings of the Workshop on Bildverarbeitung fur die Medizin, 2018, Erlangen, DEU Ed. Andreas Maier, Thomas M. Deserno, Heinz Handels, Klaus H. Maier-Hein, Christoph Palm, Thomas Tolxdorff, Springer Science and Business Media Deutschland GmbH, 2018. 216-221.

BibTeX: Download