A joint network optimization framework to predict clinical severity from resting state functional MRI data
We propose a novel optimization framework to predict clinical severity from resting state fMRI (rs-fMRI) data. Our model consists of two coupled terms. The first term decomposes the correlation matrices into a sparse set of representative subnetworks that define a network manifold. These subnetworks...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2020-02-01
|
Series: | NeuroImage |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S105381191930905X |
id |
doaj-cdeef0259ada415b947541e8f0afeaf4 |
---|---|
record_format |
Article |
spelling |
doaj-cdeef0259ada415b947541e8f0afeaf42020-11-25T03:12:10ZengElsevierNeuroImage1095-95722020-02-01206116314A joint network optimization framework to predict clinical severity from resting state functional MRI dataN.S. D’Souza0M.B. Nebel1N. Wymbs2S.H. Mostofsky3A. Venkataraman4Department of Electrical and Computer Engineering, Johns Hopkins University, USA; Corresponding author.Center for Neurodevelopmental & Imaging Research, Kennedy Krieger Institute, USACenter for Neurodevelopmental & Imaging Research, Kennedy Krieger Institute, USACenter for Neurodevelopmental & Imaging Research, Kennedy Krieger Institute, USA; Department of Neurology, Johns Hopkins School of Medicine, USA; Department of Psychiatry and Behavioral Science, Johns Hopkins School of Medicine, USADepartment of Electrical and Computer Engineering, Johns Hopkins University, USAWe propose a novel optimization framework to predict clinical severity from resting state fMRI (rs-fMRI) data. Our model consists of two coupled terms. The first term decomposes the correlation matrices into a sparse set of representative subnetworks that define a network manifold. These subnetworks are modeled as rank-one outer-products which correspond to the elemental patterns of co-activation across the brain; the subnetworks are combined via patient-specific non-negative coefficients. The second term is a linear regression model that uses the patient-specific coefficients to predict a measure of clinical severity. We validate our framework on two separate datasets in a ten fold cross validation setting. The first is a cohort of fifty-eight patients diagnosed with Autism Spectrum Disorder (ASD). The second dataset consists of sixty three patients from a publicly available ASD database. Our method outperforms standard semi-supervised frameworks, which employ conventional graph theoretic and statistical representation learning techniques to relate the rs-fMRI correlations to behavior. In contrast, our joint network optimization framework exploits the structure of the rs-fMRI correlation matrices to simultaneously capture group level effects and patient heterogeneity. Finally, we demonstrate that our proposed framework robustly identifies clinically relevant networks characteristic of ASD.http://www.sciencedirect.com/science/article/pii/S105381191930905XMatrix factorizationDictionary learningFunctional magnetic resonance imagingClinical severity |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
N.S. D’Souza M.B. Nebel N. Wymbs S.H. Mostofsky A. Venkataraman |
spellingShingle |
N.S. D’Souza M.B. Nebel N. Wymbs S.H. Mostofsky A. Venkataraman A joint network optimization framework to predict clinical severity from resting state functional MRI data NeuroImage Matrix factorization Dictionary learning Functional magnetic resonance imaging Clinical severity |
author_facet |
N.S. D’Souza M.B. Nebel N. Wymbs S.H. Mostofsky A. Venkataraman |
author_sort |
N.S. D’Souza |
title |
A joint network optimization framework to predict clinical severity from resting state functional MRI data |
title_short |
A joint network optimization framework to predict clinical severity from resting state functional MRI data |
title_full |
A joint network optimization framework to predict clinical severity from resting state functional MRI data |
title_fullStr |
A joint network optimization framework to predict clinical severity from resting state functional MRI data |
title_full_unstemmed |
A joint network optimization framework to predict clinical severity from resting state functional MRI data |
title_sort |
joint network optimization framework to predict clinical severity from resting state functional mri data |
publisher |
Elsevier |
series |
NeuroImage |
issn |
1095-9572 |
publishDate |
2020-02-01 |
description |
We propose a novel optimization framework to predict clinical severity from resting state fMRI (rs-fMRI) data. Our model consists of two coupled terms. The first term decomposes the correlation matrices into a sparse set of representative subnetworks that define a network manifold. These subnetworks are modeled as rank-one outer-products which correspond to the elemental patterns of co-activation across the brain; the subnetworks are combined via patient-specific non-negative coefficients. The second term is a linear regression model that uses the patient-specific coefficients to predict a measure of clinical severity. We validate our framework on two separate datasets in a ten fold cross validation setting. The first is a cohort of fifty-eight patients diagnosed with Autism Spectrum Disorder (ASD). The second dataset consists of sixty three patients from a publicly available ASD database. Our method outperforms standard semi-supervised frameworks, which employ conventional graph theoretic and statistical representation learning techniques to relate the rs-fMRI correlations to behavior. In contrast, our joint network optimization framework exploits the structure of the rs-fMRI correlation matrices to simultaneously capture group level effects and patient heterogeneity. Finally, we demonstrate that our proposed framework robustly identifies clinically relevant networks characteristic of ASD. |
topic |
Matrix factorization Dictionary learning Functional magnetic resonance imaging Clinical severity |
url |
http://www.sciencedirect.com/science/article/pii/S105381191930905X |
work_keys_str_mv |
AT nsdsouza ajointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT mbnebel ajointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT nwymbs ajointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT shmostofsky ajointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT avenkataraman ajointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT nsdsouza jointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT mbnebel jointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT nwymbs jointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT shmostofsky jointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata AT avenkataraman jointnetworkoptimizationframeworktopredictclinicalseverityfromrestingstatefunctionalmridata |
_version_ |
1724651165613817856 |