Deep learning semantic segmentation for high-resolution medical volumes

Imad Eddine Toubal, Ye Duan, Deshan Yang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Automated semantic segmentation in the domain of medical imaging can enable a faster, more reliable, and more affordable clinical workflow. Fully convolutional networks (FCNs) have been heavily used in this area due to the level of success that they have achieved. In this work, we first leverage recent architectural innovations to make an initial segmentation: (i) spatial and channel-wise squeeze and excitation mechanism; (ii) a 3D U-Net++ network and deep supervision. Second, we use classical methods for refining the initial segmentation: (i) spatial normalization and (ii) local 3D refinement network applied to patches. Finally, we put our methods together in a novel segmentation pipeline. We train and evaluate our models and pipelines on a dataset of a 120 abdominal magnetic resonance - volumetric - images (MRIs). The goal is to segment five different organs of interest (ORI): liver, kidneys, stomach, duodenum, and large bowel. Our experiments show that we can generate high resolution segmentation of comparable quality to the state-of-the-art methods on low resolution without adding significant computational cost.

Original languageEnglish
Title of host publication2020 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728182438
DOIs
StatePublished - Oct 13 2020
Event2020 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2020 - Washington, United States
Duration: Oct 13 2020Oct 15 2020

Publication series

NameProceedings - Applied Imagery Pattern Recognition Workshop
Volume2020-October
ISSN (Print)2164-2516

Conference

Conference2020 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2020
Country/TerritoryUnited States
CityWashington
Period10/13/2010/15/20

Keywords

  • Image segmentation
  • Medical image

Fingerprint

Dive into the research topics of 'Deep learning semantic segmentation for high-resolution medical volumes'. Together they form a unique fingerprint.

Cite this