Automatic brain tumour regions segmentation using modified U-Net

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

Research output: Contribution to journalArticlepeer-review

Abstract

Early diagnosis is an important key for brain tumour patients' survival. The segmentation of the tumour regions is done manually by the experts and it is time-consuming. In this work, we present a novel network architecture that automatically segments the whole tumour regions and intra-tumour structures (edema, enhancing tumour, necrotic and non-enhancing tumour). We evaluated the results using dice similarity coefficient and obtained promising results.
Original languageEnglish
Pages (from-to)45-48
Number of pages4
JournalAcademic Journal for Thai Researchers in Europe
Volume1
Issue number1
DOIs
Publication statusPublished - 31 Dec 2020

Keywords

  • automatic segmentation
  • brain tumour
  • U-net

Fingerprint Dive into the research topics of 'Automatic brain tumour regions segmentation using modified U-Net'. Together they form a unique fingerprint.

Cite this