Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis.
A relatively large number of studies have investigated the power of structural magnetic resonance imaging (sMRI) data to discriminate patients with schizophrenia from healthy controls. However, very few of them have also included patients with bipolar disorder, allowing the clinically relevant discr...
Main Authors: | , , , , , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Public Library of Science (PLoS)
2017-01-01
|
Series: | PLoS ONE |
Online Access: | http://europepmc.org/articles/PMC5398548?pdf=render |
_version_ | 1818360502370172928 |
---|---|
author | Raymond Salvador Joaquim Radua Erick J Canales-Rodríguez Aleix Solanes Salvador Sarró José M Goikolea Alicia Valiente Gemma C Monté María Del Carmen Natividad Amalia Guerrero-Pedraza Noemí Moro Paloma Fernández-Corcuera Benedikt L Amann Teresa Maristany Eduard Vieta Peter J McKenna Edith Pomarol-Clotet |
author_facet | Raymond Salvador Joaquim Radua Erick J Canales-Rodríguez Aleix Solanes Salvador Sarró José M Goikolea Alicia Valiente Gemma C Monté María Del Carmen Natividad Amalia Guerrero-Pedraza Noemí Moro Paloma Fernández-Corcuera Benedikt L Amann Teresa Maristany Eduard Vieta Peter J McKenna Edith Pomarol-Clotet |
author_sort | Raymond Salvador |
collection | DOAJ |
description | A relatively large number of studies have investigated the power of structural magnetic resonance imaging (sMRI) data to discriminate patients with schizophrenia from healthy controls. However, very few of them have also included patients with bipolar disorder, allowing the clinically relevant discrimination between both psychotic diagnostics. To assess the efficacy of sMRI data for diagnostic prediction in psychosis we objectively evaluated the discriminative power of a wide range of commonly used machine learning algorithms (ridge, lasso, elastic net and L0 norm regularized logistic regressions, a support vector classifier, regularized discriminant analysis, random forests and a Gaussian process classifier) on main sMRI features including grey and white matter voxel-based morphometry (VBM), vertex-based cortical thickness and volume, region of interest volumetric measures and wavelet-based morphometry (WBM) maps. All possible combinations of algorithms and data features were considered in pairwise classifications of matched samples of healthy controls (N = 127), patients with schizophrenia (N = 128) and patients with bipolar disorder (N = 128). Results show that the selection of feature type is important, with grey matter VBM (without data reduction) delivering the best diagnostic prediction rates (averaging over classifiers: schizophrenia vs. healthy 75%, bipolar disorder vs. healthy 63% and schizophrenia vs. bipolar disorder 62%) whereas algorithms usually yielded very similar results. Indeed, those grey matter VBM accuracy rates were not even improved by combining all feature types in a single prediction model. Further multi-class classifications considering the three groups simultaneously made evident a lack of predictive power for the bipolar group, probably due to its intermediate anatomical features, located between those observed in healthy controls and those found in patients with schizophrenia. Finally, we provide MRIPredict (https://www.nitrc.org/projects/mripredict/), a free tool for SPM, FSL and R, to easily carry out voxelwise predictions based on VBM images. |
first_indexed | 2024-12-13T21:01:49Z |
format | Article |
id | doaj.art-ba90e241a26a400091c09c7242e86cef |
institution | Directory Open Access Journal |
issn | 1932-6203 |
language | English |
last_indexed | 2024-12-13T21:01:49Z |
publishDate | 2017-01-01 |
publisher | Public Library of Science (PLoS) |
record_format | Article |
series | PLoS ONE |
spelling | doaj.art-ba90e241a26a400091c09c7242e86cef2022-12-21T23:31:34ZengPublic Library of Science (PLoS)PLoS ONE1932-62032017-01-01124e017568310.1371/journal.pone.0175683Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis.Raymond SalvadorJoaquim RaduaErick J Canales-RodríguezAleix SolanesSalvador SarróJosé M GoikoleaAlicia ValienteGemma C MontéMaría Del Carmen NatividadAmalia Guerrero-PedrazaNoemí MoroPaloma Fernández-CorcueraBenedikt L AmannTeresa MaristanyEduard VietaPeter J McKennaEdith Pomarol-ClotetA relatively large number of studies have investigated the power of structural magnetic resonance imaging (sMRI) data to discriminate patients with schizophrenia from healthy controls. However, very few of them have also included patients with bipolar disorder, allowing the clinically relevant discrimination between both psychotic diagnostics. To assess the efficacy of sMRI data for diagnostic prediction in psychosis we objectively evaluated the discriminative power of a wide range of commonly used machine learning algorithms (ridge, lasso, elastic net and L0 norm regularized logistic regressions, a support vector classifier, regularized discriminant analysis, random forests and a Gaussian process classifier) on main sMRI features including grey and white matter voxel-based morphometry (VBM), vertex-based cortical thickness and volume, region of interest volumetric measures and wavelet-based morphometry (WBM) maps. All possible combinations of algorithms and data features were considered in pairwise classifications of matched samples of healthy controls (N = 127), patients with schizophrenia (N = 128) and patients with bipolar disorder (N = 128). Results show that the selection of feature type is important, with grey matter VBM (without data reduction) delivering the best diagnostic prediction rates (averaging over classifiers: schizophrenia vs. healthy 75%, bipolar disorder vs. healthy 63% and schizophrenia vs. bipolar disorder 62%) whereas algorithms usually yielded very similar results. Indeed, those grey matter VBM accuracy rates were not even improved by combining all feature types in a single prediction model. Further multi-class classifications considering the three groups simultaneously made evident a lack of predictive power for the bipolar group, probably due to its intermediate anatomical features, located between those observed in healthy controls and those found in patients with schizophrenia. Finally, we provide MRIPredict (https://www.nitrc.org/projects/mripredict/), a free tool for SPM, FSL and R, to easily carry out voxelwise predictions based on VBM images.http://europepmc.org/articles/PMC5398548?pdf=render |
spellingShingle | Raymond Salvador Joaquim Radua Erick J Canales-Rodríguez Aleix Solanes Salvador Sarró José M Goikolea Alicia Valiente Gemma C Monté María Del Carmen Natividad Amalia Guerrero-Pedraza Noemí Moro Paloma Fernández-Corcuera Benedikt L Amann Teresa Maristany Eduard Vieta Peter J McKenna Edith Pomarol-Clotet Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis. PLoS ONE |
title | Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis. |
title_full | Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis. |
title_fullStr | Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis. |
title_full_unstemmed | Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis. |
title_short | Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis. |
title_sort | evaluation of machine learning algorithms and structural features for optimal mri based diagnostic prediction in psychosis |
url | http://europepmc.org/articles/PMC5398548?pdf=render |
work_keys_str_mv | AT raymondsalvador evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT joaquimradua evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT erickjcanalesrodriguez evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT aleixsolanes evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT salvadorsarro evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT josemgoikolea evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT aliciavaliente evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT gemmacmonte evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT mariadelcarmennatividad evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT amaliaguerreropedraza evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT noemimoro evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT palomafernandezcorcuera evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT benediktlamann evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT teresamaristany evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT eduardvieta evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT peterjmckenna evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis AT edithpomarolclotet evaluationofmachinelearningalgorithmsandstructuralfeaturesforoptimalmribaseddiagnosticpredictioninpsychosis |