#167:
S. Candemir,
K. Palaniappan, and
Y. S. Akgul
IEEE Int. Symposium on Biomedical Imaging (ISBI),
2013
Abstract,
Bibtex,
PlainText,
PDF,
DOI,
Google Scholar
One of the first steps of computer-aided systems is robustly detect the anatomical boundaries. Literature has several successful energy minimization based algorithms which are applied to medical images. However, these algorithms depend on parameters which need to be tuned for a meaningful solution. One of the important parameters is the regularization parameter (lambda) which is generally estimated in an ad-hoc manner and is used for the whole data set. In this paper we claim that lambda can be learned by local features which hold the regional characteristics of the image. We propose a lambda estimation system which is modeled as a multi-class classification scheme. We demonstrate the performance of the approach within graph cut segmentation framework via qualitative results on chest X-rays. Experimental results indicate that predicted parameters produce better segmentation results.
@inproceedings{Sema:ISBI-2013-graph-cut,
author = "S. Candemir and K. Palaniappan and Y. S. Akgul",
title = "Multi-class regularization parameter learning for graph cut image segmentation",
year = 2013,
booktitle = "IEEE Int. Symposium on Biomedical Imaging (ISBI)",
keywords = "segmentation, graph methods, chest x-ray, machine learning, biomedical",
doi = "10.1109/ISBI.2013.6556813"
}
S. Candemir, K. Palaniappan, and Y. S. Akgul. Multi-class regularization parameter learning for graph cut image segmentation. IEEE Int. Symposium on Biomedical Imaging (ISBI), 2013.