Skip to main navigation Skip to search Skip to main content

Deep-FExt: Deep feature extraction for vessel segmentation and centerline prediction

  • Technical University of Munich

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

14 Scopus citations

Abstract

Feature extraction is a very crucial task in image and pixel (voxel) classification and regression in biomedical image modelling. In this work we present a feature extraction scheme based on inception models for pixel classification tasks. We extract features under multi-scale and multi-layer schemes through convolutional operators. Layers of Fully Convolutional Network are later stacked on these feature extraction layers and trained end-to-end for the purpose of classification. We test our model on the DRIVE and STARE public data sets for the purpose of segmentation and centerline detection and it outperforms most existing hand crafted or deterministic feature schemes found in literature. We achieve an average maximum Dice of 0.85 on the DRIVE data set which outperforms the scores from the second human annotator of this data set. We also achieve an average maximum Dice of 0.85 and kappa of 0.84 on the STARE data set. Even though these datasets are only 2-D we also propose ways of extending this feature extraction scheme to handle 3-D datasets.

Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging - 8th International Workshop, MLMI 2017, Held in Conjunction with MICCAI 2017, Proceedings
EditorsYinghuan Shi, Heung-Il Suk, Kenji Suzuki, Qian Wang
PublisherSpringer Verlag
Pages344-352
Number of pages9
ISBN (Print)9783319673882
DOIs
StatePublished - 2017
Event8th International Workshop on Machine Learning in Medical Imaging, MLMI 2017 held in conjunction with the 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 - Quebec City, Canada
Duration: 10 Sep 201710 Sep 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10541 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference8th International Workshop on Machine Learning in Medical Imaging, MLMI 2017 held in conjunction with the 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017
Country/TerritoryCanada
CityQuebec City
Period10/09/1710/09/17

Keywords

  • Biomedical image modelling
  • Centerline prediction
  • Convolutional Networks
  • Feature extraction
  • Image and pixel classification and regression
  • Inception models
  • Vessel segmentation

Fingerprint

Dive into the research topics of 'Deep-FExt: Deep feature extraction for vessel segmentation and centerline prediction'. Together they form a unique fingerprint.

Cite this