Sparse multivariate measures of similarity between intra-modal neuroimaging datasets
An increasing number of neuroimaging studies are now based on either combining more than one data modality (inter-modal) or combining more than one measurement from the same modality (intra-modal). To date, most intra-modal studies using multivariate statistics have focused on differences between da...
Main Authors: | , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2015-10-01
|
Series: | Frontiers in Neuroscience |
Subjects: | |
Online Access: | http://journal.frontiersin.org/Journal/10.3389/fnins.2015.00366/full |
_version_ | 1811212293433720832 |
---|---|
author | Maria J. Rosa Maria J. Rosa Mitul A. Mehta Emilio eMerlo Pich Celine eRisterucci Fernando eZelaya A.A.T Simone eReinders Steve eWilliams Paola eDazzan Paola eDazzan Orla M Doyle Andre eMarquand Andre eMarquand |
author_facet | Maria J. Rosa Maria J. Rosa Mitul A. Mehta Emilio eMerlo Pich Celine eRisterucci Fernando eZelaya A.A.T Simone eReinders Steve eWilliams Paola eDazzan Paola eDazzan Orla M Doyle Andre eMarquand Andre eMarquand |
author_sort | Maria J. Rosa |
collection | DOAJ |
description | An increasing number of neuroimaging studies are now based on either combining more than one data modality (inter-modal) or combining more than one measurement from the same modality (intra-modal). To date, most intra-modal studies using multivariate statistics have focused on differences between datasets, for instance relying on classifiers to differentiate between effects in the data. However, to fully characterize these effects, multivariate methods able to measure similarities between datasets are needed. One classical technique for estimating the relationship between two datasets is canonical correlation analysis (CCA). However, in the context of high-dimensional data the application of CCA is extremely challenging. A recent extension of CCA, sparse CCA (SCCA), overcomes this limitation, by regularizing the model parameters while yielding a sparse solution. In this work, we modify SCCA with the aim of facilitating its application to high-dimensional neuroimaging data and finding meaningful multivariate image-to-image correspondences in intra-modal studies. In particular, we show how the optimal subset of variables can be estimated independently and we look at the information encoded in more than one set of SCCA transformations. We illustrate our framework using Arterial Spin Labelling data to investigate multivariate similarities between the effects of two antipsychotic drugs on cerebral blood flow. |
first_indexed | 2024-04-12T05:26:42Z |
format | Article |
id | doaj.art-6d679a038b484c60b64523767994e78d |
institution | Directory Open Access Journal |
issn | 1662-453X |
language | English |
last_indexed | 2024-04-12T05:26:42Z |
publishDate | 2015-10-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Neuroscience |
spelling | doaj.art-6d679a038b484c60b64523767994e78d2022-12-22T03:46:15ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2015-10-01910.3389/fnins.2015.00366162470Sparse multivariate measures of similarity between intra-modal neuroimaging datasetsMaria J. Rosa0Maria J. Rosa1Mitul A. Mehta2Emilio eMerlo Pich3Celine eRisterucci4Fernando eZelaya5A.A.T Simone eReinders6Steve eWilliams7Paola eDazzan8Paola eDazzan9Orla M Doyle10Andre eMarquand11Andre eMarquand12IXICO PLC.The Institute of Psychiatry, Psychology & NeuroscienceThe Institute of Psychiatry, Psychology & NeuroscienceF. Hoffman – La RocheF. Hoffman – La RocheThe Institute of Psychiatry, Psychology & NeuroscienceInstitute of Psychiatry, Psychology & NeuroscienceThe Institute of Psychiatry, Psychology & NeuroscienceInstitute of Psychiatry, Psychology & NeuroscienceNational Institute for Health Research Mental Health Biomedical Research CentreThe Institute of Psychiatry, Psychology & NeuroscienceThe Institute of Psychiatry, Psychology & NeuroscienceDonders Institute for Brain, Cognition and Behaviour, Radboud UniversityAn increasing number of neuroimaging studies are now based on either combining more than one data modality (inter-modal) or combining more than one measurement from the same modality (intra-modal). To date, most intra-modal studies using multivariate statistics have focused on differences between datasets, for instance relying on classifiers to differentiate between effects in the data. However, to fully characterize these effects, multivariate methods able to measure similarities between datasets are needed. One classical technique for estimating the relationship between two datasets is canonical correlation analysis (CCA). However, in the context of high-dimensional data the application of CCA is extremely challenging. A recent extension of CCA, sparse CCA (SCCA), overcomes this limitation, by regularizing the model parameters while yielding a sparse solution. In this work, we modify SCCA with the aim of facilitating its application to high-dimensional neuroimaging data and finding meaningful multivariate image-to-image correspondences in intra-modal studies. In particular, we show how the optimal subset of variables can be estimated independently and we look at the information encoded in more than one set of SCCA transformations. We illustrate our framework using Arterial Spin Labelling data to investigate multivariate similarities between the effects of two antipsychotic drugs on cerebral blood flow.http://journal.frontiersin.org/Journal/10.3389/fnins.2015.00366/fullAntipsychoticsArterial Spin Labelingmultivariate analysissparse canonical correlation analysisrepeated measurespharmacological MRI |
spellingShingle | Maria J. Rosa Maria J. Rosa Mitul A. Mehta Emilio eMerlo Pich Celine eRisterucci Fernando eZelaya A.A.T Simone eReinders Steve eWilliams Paola eDazzan Paola eDazzan Orla M Doyle Andre eMarquand Andre eMarquand Sparse multivariate measures of similarity between intra-modal neuroimaging datasets Frontiers in Neuroscience Antipsychotics Arterial Spin Labeling multivariate analysis sparse canonical correlation analysis repeated measures pharmacological MRI |
title | Sparse multivariate measures of similarity between intra-modal neuroimaging datasets |
title_full | Sparse multivariate measures of similarity between intra-modal neuroimaging datasets |
title_fullStr | Sparse multivariate measures of similarity between intra-modal neuroimaging datasets |
title_full_unstemmed | Sparse multivariate measures of similarity between intra-modal neuroimaging datasets |
title_short | Sparse multivariate measures of similarity between intra-modal neuroimaging datasets |
title_sort | sparse multivariate measures of similarity between intra modal neuroimaging datasets |
topic | Antipsychotics Arterial Spin Labeling multivariate analysis sparse canonical correlation analysis repeated measures pharmacological MRI |
url | http://journal.frontiersin.org/Journal/10.3389/fnins.2015.00366/full |
work_keys_str_mv | AT mariajrosa sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT mariajrosa sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT mitulamehta sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT emilioemerlopich sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT celineeristerucci sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT fernandoezelaya sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT aatsimoneereinders sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT steveewilliams sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT paolaedazzan sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT paolaedazzan sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT orlamdoyle sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT andreemarquand sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets AT andreemarquand sparsemultivariatemeasuresofsimilaritybetweenintramodalneuroimagingdatasets |