Improved Breast Mass Segmentation in Mammograms with Conditional Residual U-Net

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We explore the use of deep learning for breast mass segmentation in mammograms. By integrating the merits of residual learning and probabilistic graphical modelling with standard U-Net, we propose a new deep network, Conditional Residual U-Net (CRU-Net), to improve the U-Net segmentation performance. Benefiting from the advantage of probabilistic graphical modelling in the pixel-level labelling, and the structure insights of a deep residual network in the feature extraction, the CRU-Net provides excellent mass segmentation performance. Evaluations based on INbreast and DDSM-BCRP datasets demonstrate that the CRU-Net achieves the best mass segmentation performance compared to the state-of-art methodologies. Moreover, neither tedious pre-processing nor post-processing techniques are not required in our algorithm.
Original languageEnglish
Title of host publicationImage Analysis for Moving Organ, Breast, and Thoracic Images
PublisherSpringer
ISBN (Electronic)978-3-030-00946-5
ISBN (Print)978-3-030-00945-8
DOIs
Publication statusPublished - 16 Sep 2018
Event4th International Workshop, BIA 2018 -
Duration: 16 Sep 201820 Sep 2018
https://www.springerprofessional.de/en/image-analysis-for-moving-organ-breast-and-thoracic-images/16114584?tocPage=2#TOC

Conference

Conference4th International Workshop, BIA 2018
Period16/09/1820/09/18
Internet address

Fingerprint

Dive into the research topics of 'Improved Breast Mass Segmentation in Mammograms with Conditional Residual U-Net'. Together they form a unique fingerprint.

Cite this