Dataset Information

Non-negative matrix factorization of multimodal MRI, fMRI and phenotypic data reveals differential changes in default mode subnetworks in ADHD.

ABSTRACT: In the multimodal neuroimaging framework, data on a single subject are collected from inherently different sources such as functional MRI, structural MRI, behavioral and/or phenotypic information. The information each source provides is not independent; a subset of features from each modality maps to one or more common latent dimensions, which can be interpreted using generative models. These latent dimensions, or "topics," provide a sparse summary of the generative process behind the features for each individual. Topic modeling, an unsupervised generative model, has been used to map seemingly disparate features to a common domain. We use Non-Negative Matrix Factorization (NMF) to infer the latent structure of multimodal ADHD data containing fMRI, MRI, phenotypic and behavioral measurements. We compare four different NMF algorithms and find that the sparsest decomposition is also the most differentiating between ADHD and healthy patients. We identify dimensions that map to interpretable, recognizable dimensions such as motion, default mode network activity, and other such features of the input data. For example, structural and functional graph theory features related to default mode subnetworks clustered with the ADHD-Inattentive diagnosis. Structural measurements of the default mode network (DMN) regions such as the posterior cingulate, precuneus, and parahippocampal regions were all related to the ADHD-Inattentive diagnosis. Ventral DMN subnetworks may have more functional connections in ADHD-I, while dorsal DMN may have less. ADHD topics are dependent upon diagnostic site, suggesting diagnostic differences across geographic locations. We assess our findings in light of the ADHD-200 classification competition, and contrast our unsupervised, nominated topics with previously published supervised learning methods. Finally, we demonstrate the validity of these latent variables as biomarkers by using them for classification of ADHD in 730 patients. Cumulatively, this manuscript addresses how multimodal data in ADHD can be interpreted by latent dimensions.

SUBMITTER: Anderson A

PROVIDER: S-EPMC4063903 | biostudies-literature | 2014 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Non-negative matrix factorization of multimodal MRI, fMRI and phenotypic data reveals differential changes in default mode subnetworks in ADHD.

Anderson Ariana A Douglas Pamela K PK Douglas Pamela K PK Kerr Wesley T WT Haynes Virginia S VS Yuille Alan L AL Xie Jianwen J Wu Ying Nian YN Brown Jesse A JA Cohen Mark S MS

NeuroImage 20131219

In the multimodal neuroimaging framework, data on a single subject are collected from inherently different sources such as functional MRI, structural MRI, behavioral and/or phenotypic information. The information each source provides is not independent; a subset of features from each modality maps to one or more common latent dimensions, which can be interpreted using generative models. These latent dimensions, or "topics," provide a sparse summary of the generative process behind the features f ...[more]

PMID: 24361664

Similar Datasets

Project description:The amounts and types of available multimodal tumor data are rapidly increasing, and their integration is critical for fully understanding the underlying cancer biology and personalizing treatment. However, the development of methods for effectively integrating multimodal data in a principled manner is lagging behind our ability to generate the data. In this article, we introduce an extension to a multiview nonnegative matrix factorization algorithm (NNMF) for dimensionality reduction and integration of heterogeneous data types and compare the predictive modeling performance of the method on unimodal and multimodal data. We also present a comparative evaluation of our novel multiview approach and current data integration methods. Our work provides an efficient method to extend an existing dimensionality reduction method. We report rigorous evaluation of the method on large-scale quantitative protein and phosphoprotein tumor data from the Clinical Proteomic Tumor Analysis Consortium (CPTAC) acquired using state-of-the-art liquid chromatography mass spectrometry. Exome sequencing and RNA-Seq data were also available from The Cancer Genome Atlas for the same tumors. For unimodal data, in case of breast cancer, transcript levels were most predictive of estrogen and progesterone receptor status and copy number variation of human epidermal growth factor receptor 2 status. For ovarian and colon cancers, phosphoprotein and protein levels were most predictive of tumor grade and stage and residual tumor, respectively. When multiview NNMF was applied to multimodal data to predict outcomes, the improvement in performance is not overall statistically significant beyond unimodal data, suggesting that proteomics data may contain more predictive information regarding tumor phenotypes than transcript levels, probably due to the fact that proteins are the functional gene products and therefore a more direct measurement of the functional state of the tumor. Here, we have applied our proposed approach to multimodal molecular data for tumors, but it is generally applicable to dimensionality reduction and joint analysis of any type of multimodal data.

Project description:Deficits characteristic of attention deficit/hyperactivity disorder (ADHD), including poor attention and inhibitory control, are at least partially alleviated by factors that increase engagement of attention, suggesting a hypodopaminergic reward deficit. Lapses of attention are associated with attenuated deactivation of the default mode network (DMN), a distributed brain system normally deactivated during tasks requiring attention to the external world. Task-related DMN deactivation has been shown to be attenuated in ADHD relative to controls. We hypothesised that motivational incentives to balance speed against restraint would increase task engagement during an inhibitory control task, enhancing DMN deactivation in ADHD. We also hypothesised that methylphenidate, an indirect dopamine agonist, would tend to normalise abnormal patterns of DMN deactivation.We obtained functional magnetic resonance images from 18 methylphenidate-responsive children with ADHD (DSM-IV combined subtype) and 18 pairwise-matched typically developing children aged 9-15 years while they performed a paced Go/No-go task. We manipulated motivational incentive to balance response speed against inhibitory control, and tested children with ADHD both on and off methylphenidate.When children with ADHD were off-methylphenidate and task incentive was low, event-related DMN deactivation was significantly attenuated compared to controls, but the two groups did not differ under high motivational incentives. The modulation of DMN deactivation by incentive in the children with ADHD, off-methylphenidate, was statistically significant, and significantly greater than in typically developing children. When children with ADHD were on-methylphenidate, motivational modulation of event-related DMN deactivation was abolished, and no attenuation relative to their typically developing peers was apparent in either motivational condition.During an inhibitory control task, children with ADHD exhibit a raised motivational threshold at which task-relevant stimuli become sufficiently salient to deactivate the DMN. Treatment with methylphenidate normalises this threshold, rendering their pattern of task-related DMN deactivation indistinguishable from that of typically developing children.

Dataset Information

Non-negative matrix factorization of multimodal MRI, fMRI and phenotypic data reveals differential changes in default mode subnetworks in ADHD.

Publications

Non-negative matrix factorization of multimodal MRI, fMRI and phenotypic data reveals differential changes in default mode subnetworks in ADHD.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets