Uncertainty-aware Ensemble of Classifiers for Segmenting Brain MRI Data

Al-Taie, Ahmed; Hahn, Horst K.; Linsen, Lars

Uncertainty-aware Ensemble of Classifiers for Segmenting Brain MRI Data

Files

041-050.pdf (1.79 MB)

Date

2014

Authors

Al-Taie, Ahmed
Hahn, Horst K.
Linsen, Lars

Publisher

The Eurographics Association

Abstract

Estimating and visualizing uncertainty in medical image segmentation has become an active research area due to the necessity of making medical experts aware of possibly wrong segmentation decisions. Still, to our knowledge all these methods are based on a single choice of the underlying segmentation approach. Segmentation using an ensemble of classifiers (or committee machine) use multiple classifiers to increase the performance when compared to applying a single classifier. In this paper, we propose methods to estimate uncertainties in segmentations produced by ensembles of classifiers. We investigate and compare the different combining strategies of the segmentation results of the ensemble members from an uncertainty point of view. We discuss why some combining strategies tend to perform better than others. Also, we visualize the estimated uncertainties using a color mapping in image space and propose a post-segmentations correction step to reclassify the noisy pixels in the final result based on the statistical uncertainty.

        @inproceedings{:10.2312/vcbm.20141182
https::/diglib.eg.org/handle/10.2312/vcbm.20141182.041-050
,
booktitle = {Eurographics Workshop on Visual Computing for Biology and Medicine
},
editor = {Ivan Viola and Katja Buehler and Timo Ropinski
},
title = {{Uncertainty-aware Ensemble of Classifiers for Segmenting Brain MRI Data
}},
author = {Al-Taie, Ahmed and 
Hahn, Horst K. and 
Linsen, Lars
},
year = {2014
},
publisher = {The Eurographics Association
},
ISSN = {2070-5778
},
ISBN = {978-3-905674-62-0
},
DOI = {/10.2312/vcbm.20141182
https://diglib.eg.org/handle/10.2312/vcbm.20141182.041-050
}
}