Rib Segmentation in Chest Radiographs by Support Vector Machine

DOI

10.2991/emcs-16.2016.391How to use a DOI?

Keywords

Rib segmentation; Support vector machine (SVM); Wavelet transformation; Gaussian derivative; Feature extraction

Abstract

The segmentation of ribs is one of the key problems for computer-aided diagnosis. This paper presented a novel scheme for rib segmentation in chest radiographs. The Gaussian filter was used repeatedly to remove uneven background of chest. Detail images could be obtained by use of the multi-scale wavelet decomposition, and then Gaussian derivative was employed to extract features from the detail images. The rib model for support vector machine (SVM) was established, which was employed to classify ribs. In our study, a method of sample selection with property reduction was used to reduce calculation time. The experimental result shows that proposed scheme can segment ribs in chest radiographs effectively.

Copyright

© 2016, 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  - Guodong Zhang
AU  - Haiping Wu
AU  - Wei Guo
PY  - 2016/01
DA  - 2016/01
TI  - Rib Segmentation in Chest Radiographs by Support Vector Machine
BT  - Proceedings of the 2016 International Conference on Education, Management, Computer and Society
PB  - Atlantis Press
SP  - 1564
EP  - 1567
SN  - 2352-538X
UR  - https://doi.org/10.2991/emcs-16.2016.391
DO  - 10.2991/emcs-16.2016.391
ID  - Zhang2016/01
ER  -