Modified U-Net for automatic brain tumor regions segmentation

Kaewrak, Keerati and Soraghan, John and Di Caterina, Gaetano and Grose, Derek (2019) Modified U-Net for automatic brain tumor regions segmentation. In: 27th European Signal Processing Conference, 2019-09-02 - 2019-09-06.

[thumbnail of Kaewrak-etal-Eusipco-2019-Modified-u-net-automatic-brain-tumor-regions-segmentation]
Preview
 Text (Kaewrak-etal-Eusipco-2019-Modified-u-net-automatic-brain-tumor-regions-segmentation)
Kaewrak_etal_Eusipco_2019_Modified_u_net_automatic_brain_tumor_regions_segmentation.pdf
Accepted Author Manuscript
Download (618kB)| Preview

    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.

    ORCID iDs

    Kaewrak, Keerati, Soraghan, John ORCID logoORCID: https://orcid.org/0000-0003-4418-7391, Di Caterina, Gaetano ORCID logoORCID: https://orcid.org/0000-0002-7256-0897 and Grose, Derek;
    CORE (COnnecting REpositories)