Training Multi-organ Segmentation Networks with Sample Selection by Relaxed Upper Confident Bound

  • Yan Wang*
  • , Yuyin Zhou
  • , Peng Tang
  • , Wei Shen
  • , Elliot K. Fishman
  • , Alan L. Yuille
  • *Corresponding author for this work

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

17 Scopus citations

Abstract

Convolutional neural networks (CNNs), especially fully convolutional networks, have been widely applied to automatic medical image segmentation problems, e.g., multi-organ segmentation. Existing CNN-based segmentation methods mainly focus on looking for increasingly powerful network architectures, but pay less attention to data sampling strategies for training networks more effectively. In this paper, we present a simple but effective sample selection method for training multi-organ segmentation networks. Sample selection exhibits an exploitation-exploration strategy, i.e., exploiting hard samples and exploring less frequently visited samples. Based on the fact that very hard samples might have annotation errors, we propose a new sample selection policy, named Relaxed Upper Confident Bound (RUCB). Compared with other sample selection policies, e.g., Upper Confident Bound (UCB), it exploits a range of hard samples rather than being stuck with a small set of very hard ones, which mitigates the influence of annotation errors during training. We apply this new sample selection policy to training a multi-organ segmentation network on a dataset containing 120 abdominal CT scans and show that it boosts segmentation performance significantly.

Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings
EditorsAlejandro F. Frangi, Gabor Fichtinger, Julia A. Schnabel, Carlos Alberola-López, Christos Davatzikos
PublisherSpringer Verlag
Pages434-442
Number of pages9
ISBN (Print)9783030009366
DOIs
StatePublished - 2018
Externally publishedYes
Event21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 16 Sep 201820 Sep 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11073 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
Country/TerritorySpain
CityGranada
Period16/09/1820/09/18

Fingerprint

Dive into the research topics of 'Training Multi-organ Segmentation Networks with Sample Selection by Relaxed Upper Confident Bound'. Together they form a unique fingerprint.

Cite this