Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation
Neurodegenerative diseases are frequently associated with structural changes in the brain. Magnetic resonance imaging (MRI) scans can show these variations and therefore can be used as a supportive feature for a number of neurodegenerative diseases. The hippocampus has been known to be a biomarker f...
Main Authors: | , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Hindawi Limited
2015-01-01
|
Series: | Computational and Mathematical Methods in Medicine |
Online Access: | http://dx.doi.org/10.1155/2015/814104 |
id |
doaj-e7a8e7f1c17348fe98952f9a3c143d85 |
---|---|
record_format |
Article |
spelling |
doaj-e7a8e7f1c17348fe98952f9a3c143d852020-11-24T22:54:58ZengHindawi LimitedComputational and Mathematical Methods in Medicine1748-670X1748-67182015-01-01201510.1155/2015/814104814104Feature Selection Based on Machine Learning in MRIs for Hippocampal SegmentationSabina Tangaro0Nicola Amoroso1Massimo Brescia2Stefano Cavuoti3Andrea Chincarini4Rosangela Errico5Paolo Inglese6Giuseppe Longo7Rosalia Maglietta8Andrea Tateo9Giuseppe Riccio10Roberto Bellotti11Istituto Nazionale di Fisica Nucleare, Sezione di Bari, Via Orabona 4, 70125 Bari, ItalyIstituto Nazionale di Fisica Nucleare, Sezione di Bari, Via Orabona 4, 70125 Bari, ItalyAstronomical Observatory of Capodimonte, INAF, Salita Moiariello 16, 80131 Napoli, ItalyAstronomical Observatory of Capodimonte, INAF, Salita Moiariello 16, 80131 Napoli, ItalyIstituto Nazionale di Fisica Nucleare, Sezione di Genova, Via Dodecaneso 33, 16146 Genova, ItalyDipartimento Interateneo di Fisica, Università degli Studi di Bari, Via Amendola 173, 70126 Bari, ItalyIstituto Nazionale di Fisica Nucleare, Sezione di Bari, Via Orabona 4, 70125 Bari, ItalyDipartimento di Fisica, Università degli Studi di Napoli Federico II, Complesso Universitario di Monte Sant’Angelo, Via Cintia 1, 80126 Napoli, ItalyIstituto di Studi sui Sistemi Intelligenti per l’Automazione, CNR, Via Giovanni Amendola 122/D-I, 70126 Bari, ItalyDipartimento Interateneo di Fisica, Università degli Studi di Bari, Via Amendola 173, 70126 Bari, ItalyAstronomical Observatory of Capodimonte, INAF, Salita Moiariello 16, 80131 Napoli, ItalyIstituto Nazionale di Fisica Nucleare, Sezione di Bari, Via Orabona 4, 70125 Bari, ItalyNeurodegenerative diseases are frequently associated with structural changes in the brain. Magnetic resonance imaging (MRI) scans can show these variations and therefore can be used as a supportive feature for a number of neurodegenerative diseases. The hippocampus has been known to be a biomarker for Alzheimer disease and other neurological and psychiatric diseases. However, it requires accurate, robust, and reproducible delineation of hippocampal structures. Fully automatic methods are usually the voxel based approach; for each voxel a number of local features were calculated. In this paper, we compared four different techniques for feature selection from a set of 315 features extracted for each voxel: (i) filter method based on the Kolmogorov-Smirnov test; two wrapper methods, respectively, (ii) sequential forward selection and (iii) sequential backward elimination; and (iv) embedded method based on the Random Forest Classifier on a set of 10 T1-weighted brain MRIs and tested on an independent set of 25 subjects. The resulting segmentations were compared with manual reference labelling. By using only 23 feature for each voxel (sequential backward elimination) we obtained comparable state-of-the-art performances with respect to the standard tool FreeSurfer.http://dx.doi.org/10.1155/2015/814104 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Sabina Tangaro Nicola Amoroso Massimo Brescia Stefano Cavuoti Andrea Chincarini Rosangela Errico Paolo Inglese Giuseppe Longo Rosalia Maglietta Andrea Tateo Giuseppe Riccio Roberto Bellotti |
spellingShingle |
Sabina Tangaro Nicola Amoroso Massimo Brescia Stefano Cavuoti Andrea Chincarini Rosangela Errico Paolo Inglese Giuseppe Longo Rosalia Maglietta Andrea Tateo Giuseppe Riccio Roberto Bellotti Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation Computational and Mathematical Methods in Medicine |
author_facet |
Sabina Tangaro Nicola Amoroso Massimo Brescia Stefano Cavuoti Andrea Chincarini Rosangela Errico Paolo Inglese Giuseppe Longo Rosalia Maglietta Andrea Tateo Giuseppe Riccio Roberto Bellotti |
author_sort |
Sabina Tangaro |
title |
Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation |
title_short |
Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation |
title_full |
Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation |
title_fullStr |
Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation |
title_full_unstemmed |
Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation |
title_sort |
feature selection based on machine learning in mris for hippocampal segmentation |
publisher |
Hindawi Limited |
series |
Computational and Mathematical Methods in Medicine |
issn |
1748-670X 1748-6718 |
publishDate |
2015-01-01 |
description |
Neurodegenerative diseases are frequently associated with
structural changes in the brain. Magnetic resonance imaging (MRI)
scans can show these variations and therefore can be used as a supportive
feature for a number of neurodegenerative diseases. The hippocampus
has been known to be a biomarker for Alzheimer disease and other neurological
and psychiatric diseases. However, it requires accurate, robust,
and reproducible delineation of hippocampal structures. Fully automatic
methods are usually the voxel based approach; for each voxel a number
of local features were calculated. In this paper, we compared four different
techniques for feature selection from a set of 315 features extracted
for each voxel: (i) filter method based on the Kolmogorov-Smirnov test;
two wrapper methods, respectively, (ii) sequential forward selection and
(iii) sequential backward elimination; and (iv) embedded method based
on the Random Forest Classifier on a set of 10 T1-weighted brain MRIs
and tested on an independent set of 25 subjects. The resulting segmentations
were compared with manual reference labelling. By using only
23 feature for each voxel (sequential backward elimination) we obtained
comparable state-of-the-art performances with respect to the standard
tool FreeSurfer. |
url |
http://dx.doi.org/10.1155/2015/814104 |
work_keys_str_mv |
AT sabinatangaro featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT nicolaamoroso featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT massimobrescia featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT stefanocavuoti featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT andreachincarini featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT rosangelaerrico featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT paoloinglese featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT giuseppelongo featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT rosaliamaglietta featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT andreatateo featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT giuseppericcio featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation AT robertobellotti featureselectionbasedonmachinelearninginmrisforhippocampalsegmentation |
_version_ |
1725658595354214400 |