Weakly Supervised Segmentation by a Deep Geodesic Prior

Aliasghar Mortazi*, Naji Khosravan, Drew A. Torigian, Sila Kurugol, Ulas Bagci

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

The performance of the state-of-the-art image segmentation methods heavily relies on the high-quality annotations, which are not easily affordable, particularly for medical data. To alleviate this limitation, in this study, we propose a weakly supervised image segmentation method based on a deep geodesic prior. We hypothesize that integration of this prior information can reduce the adverse effects of weak labels in segmentation accuracy. Our proposed algorithm is based on a prior information, extracted from an auto-encoder, trained to map objects’ geodesic maps to their corresponding binary maps. The obtained information is then used as an extra term in the loss function of the segmentor. In order to show efficacy of the proposed strategy, we have experimented segmentation of cardiac substructures with clean and two levels of noisy labels (L1, L2). Our experiments showed that the proposed algorithm boosted the performance of baseline deep learning-based segmentation for both clean and noisy labels by $$4.4\%$$, $$4.6\%$$ (L1), and $$6.3\%$$ (L2) in dice score, respectively. We also showed that the proposed method was more robust in the presence of high-level noise due to the existence of shape priors.

Original languageEnglish (US)
Title of host publicationMachine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings
EditorsHeung-Il Suk, Mingxia Liu, Chunfeng Lian, Pingkun Yan
PublisherSpringer
Pages238-246
Number of pages9
ISBN (Print)9783030326913
DOIs
StatePublished - 2019
Externally publishedYes
Event10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China
Duration: Oct 13 2019Oct 13 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11861 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Country/TerritoryChina
CityShenzhen
Period10/13/1910/13/19

Keywords

  • Deep learning
  • Geodesic prior
  • Medical image segmentation
  • Shape prior
  • Weakly supervised

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Weakly Supervised Segmentation by a Deep Geodesic Prior'. Together they form a unique fingerprint.

Cite this