A U-Net based multi-scale feature extraction for liver tumour segmentation in CT images

Ming Gong, John Soraghan, Gaetano Di Caterina, Derek Grose

Research output: Contribution to conferencePaperpeer-review

2 Downloads (Pure)

Abstract

A new method for automatic liver tumour segmentation from computed tomography (CT) scans based on deep neural network is presented. Two cascaded deep convolu-tional neural networks are used to segment the CT image of the abdominal cavity. The first U-net is used for coarse segmentation to obtain the approximate position of the liver and tumour. Using this as a prediction the original image is cropped to reduce its size in order to increase the segmentation accuracy. The second modified U-net is employed for accurate segmentation of the actual liver tumours. Residual modules and dense connections are added to U-net to help the network train faster while pro-ducing more accurate results. In addition, multi-dimensional information fusion is introduced to make the network more comprehensive in restoring information. The Liver Tumour Segmentation (LiTs) dataset is used to evaluate the relative segmenta-tion performance obtaining an average dice score of 0.665 based our method.
Original languageEnglish
Number of pages8
Publication statusPublished - 22 Aug 2021
Event10th International Conference on Communications, Signal Processing, and Systems - , China
Duration: 21 Aug 202122 Aug 2021
http://www.cspsconf.org/

Conference

Conference10th International Conference on Communications, Signal Processing, and Systems
Country/TerritoryChina
Period21/08/2122/08/21
Internet address

Keywords

  • CT data
  • liver tumour
  • U-Net
  • multi-scale feature fusion

Fingerprint

Dive into the research topics of 'A U-Net based multi-scale feature extraction for liver tumour segmentation in CT images'. Together they form a unique fingerprint.

Cite this