Modified U-Net for automatic brain tumor regions segmentation

Research output: Contribution to conferencePaper

8 Downloads (Pure)

Abstract

Novel deep learning based network architectures are investigated for advanced brain tumor image classification and segmentation. Variations in brain tumor characteristics together with limited labelled datasets represent significant challenges in automatic brain tumor segmentation. In this paper, we present a novel architecture based on the U-Net that incorporates both global and local feature extraction paths to improve the segmentation accuracy. The results included in the paper show superior performance of the novel segmentation for five tumor regions on the large BRATs 2018 dataset over other approaches.
Original languageEnglish
Number of pages5
Publication statusPublished - 5 Sep 2019
Event27th European Signal Processing Conference - A Coruna, Spain
Duration: 2 Sep 20196 Sep 2019
http://eusipco2019.org/

Conference

Conference27th European Signal Processing Conference
Abbreviated titleEUSIPCO 2019
CountrySpain
CityA Coruna
Period2/09/196/09/19
Internet address

Keywords

  • segmentation
  • u-net
  • tumor

Fingerprint Dive into the research topics of 'Modified U-Net for automatic brain tumor regions segmentation'. Together they form a unique fingerprint.

  • Cite this

    Kaewrak, K., Soraghan, J., Di Caterina, G., & Grose, D. (2019). Modified U-Net for automatic brain tumor regions segmentation. Paper presented at 27th European Signal Processing Conference, A Coruna, Spain.