On the relevance of automatically selected single-voxel MRS and multimodal MRI and MRSI features for brain tumour differentiation

Authors:
Geert J. Postma;Jan Luts;Albert J. Idema;Margarida Julií-Sapé;Ángel Moreno-Torres;Witek Gajewicz;Johan A. K. Suykens;Arend Heerschap;Sabine Van Huffel;Lutgarde M. C. Buydens
Affiliations:
Institute for Molecules and Materials, Radboud University Nijmegen, Heijendaalseweg 135, 6525 AJ Nijmegen, The Netherlands;Department of Electrical Engineering (ESAT), Research Division SCD, Katholieke Universiteit Leuven, Kasteelpark Arenberg 10, B-3001 Leuven, Belgium;Department of Neurosurgery, Radboud University Nijmegen, University Medical Center, Geert Grooteplein Z18, PO Box 9101, 6500 HB Nijmegen, The Netherlands;Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain and Un ...;Research Department, Centre Diagnòstic Pedralbes, Esplugues de Llobregat, Barcelona, Spain and CIBER-BBN, Esplugues de Llobregat, Spain;Department of Radiology, Medical University Lodz, 90-156Lodz, Kopcinskiego 22, Poland;Department of Electrical Engineering (ESAT), Research Division SCD, Katholieke Universiteit Leuven, Kasteelpark Arenberg 10, B-3001 Leuven, Belgium;Department of Radiology, Radboud University Nijmegen, University Medical Center, Geert Grooteplein Z18, PO Box 9101, 6500 HB Nijmegen, The Netherlands;Department of Electrical Engineering (ESAT), Research Division SCD, Katholieke Universiteit Leuven, Kasteelpark Arenberg 10, B-3001 Leuven, Belgium;Institute for Molecules and Materials, Radboud University Nijmegen, Heijendaalseweg 135, 6525 AJ Nijmegen, The Netherlands
Venue:
Computers in Biology and Medicine
Year:
2011

Citing 5
Cited 0

Selection of relevant features and examples in machine learning

Artificial Intelligence - Special issue on relevance
Theoretical and Empirical Analysis of ReliefF and RReliefF

Machine Learning
An introduction to variable and feature selection

The Journal of Machine Learning Research
A combined MRI and MRSI based multiclass system for brain tumour recognition using LS-SVMs with class probabilities and feature selection

Artificial Intelligence in Medicine
Bagging Linear Sparse Bayesian Learning Models for Variable Selection in Cancer Diagnosis

IEEE Transactions on Information Technology in Biomedicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to evaluate the relevance of magnetic resonance (MR) features selected by automatic feature selection techniques to build classifiers for differential diagnosis and tissue segmentation two data sets containing MR spectroscopy data from patients with brain tumours were investigated. The automatically selected features were evaluated using literature and clinical experience. It was observed that a significant part of the automatically selected features correspond to what is known from the literature and clinical experience. We conclude that automatic feature selection is a useful tool to obtain relevant and possibly interesting features, but evaluation of the obtained features remains necessary.