Automatic 2D Segmentation of an Intracardiac Catheter Based on MSER Blob Detector and Eccentricity

DOI

10.2991/iwci-18.2018.5How to use a DOI?

Keywords

cardiac ultrasound, endovascular surgery, MSER, shape analysis, catheter detection

Abstract

The present study describes an algorithm for automatic catheter detection and segmentation based on echocardiography data. The catheter area was recognized and then delineated by combination of detection and segmentation techniques such as the maximally stable extremal regions (MSER) algorithm, feature analysis and Kittler-Illingworth thresholding algorithm. Regions processed by MSER detector were restricted by eccentricity. Eccentricity, assessing the shape of the particular region, was chosen as the main feature. Morphological closing was applied at the pre-processing step. After applying MSER blob detector, detection accuracy of the catheter made up 86.7±11.5%. After performing an additional restriction based on the shape analysis, the accuracy increased to 92.8±6.6%. The proposed algorithm allows performing automatic detection and segmentation of the catheter inside the heart based on 2D echocardiography data.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Viacheslav Danilov
AU  - Igor Skirnevskiy
AU  - Olga Gerget
AU  - Roman Manakov
PY  - 2018/08
DA  - 2018/08
TI  - Automatic 2D Segmentation of an Intracardiac Catheter Based on MSER Blob Detector and Eccentricity
BT  - Proceedings of the Vth International workshop "Critical infrastructures: Contingency management, Intelligent, Agent-based, Cloud computing and Cyber security" (IWCI 2018)
PB  - Atlantis Press
SP  - 26
EP  - 30
SN  - 1951-6851
UR  - https://doi.org/10.2991/iwci-18.2018.5
DO  - 10.2991/iwci-18.2018.5
ID  - Danilov2018/08
ER  -