Modified U-Net for automatic brain tumor regions segmentation

Keerati Kaewrak; John Soraghan; Gaetano Di Caterina; Derek Grose

Modified U-Net for automatic brain tumor regions segmentation

Keerati Kaewrak, John Soraghan, Gaetano Di Caterina, Derek Grose

Electronic And Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

5 Citations (Scopus)

36 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 language	English
Number of pages	5
Publication status	Published - 5 Sept 2019
Event	27th European Signal Processing Conference - A Coruna, Spain Duration: 2 Sept 2019 → 6 Sept 2019 http://eusipco2019.org/

Conference

Conference	27th European Signal Processing Conference
Abbreviated title	EUSIPCO 2019
Country/Territory	Spain
City	A Coruna
Period	2/09/19 → 6/09/19
Internet address	http://eusipco2019.org/

Keywords

segmentation
u-net
tumor

Access to Document

Kaewrak-etal-Eusipco-2019-Modified-u-net-automatic-brain-tumor-regions-segmentationAccepted author manuscript, 604 KB

Cite this

@conference{cb67f5c746ab463680545a59c6700125,

title = "Modified U-Net for automatic brain tumor regions segmentation",

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.",

keywords = "segmentation, u-net, tumor",

author = "Keerati Kaewrak and John Soraghan and {Di Caterina}, Gaetano and Derek Grose",

year = "2019",

month = sep,

day = "5",

language = "English",

note = "27th European Signal Processing Conference, EUSIPCO 2019 ; Conference date: 02-09-2019 Through 06-09-2019",

url = "http://eusipco2019.org/",

}

TY - CONF

T1 - Modified U-Net for automatic brain tumor regions segmentation

AU - Kaewrak, Keerati

AU - Soraghan, John

AU - Di Caterina, Gaetano

AU - Grose, Derek

PY - 2019/9/5

Y1 - 2019/9/5

N2 - 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.

AB - 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.

KW - segmentation

KW - u-net

KW - tumor

UR - http://eusipco2019.org/

M3 - Paper

T2 - 27th European Signal Processing Conference

Y2 - 2 September 2019 through 6 September 2019

ER -

Modified U-Net for automatic brain tumor regions segmentation

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this