Show simple item record

hal.structure.identifierInstitut Mines-Telecom
dc.contributor.authorGramfort, A.
HAL ID: 687
*
hal.structure.identifierCEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
dc.contributor.authorPeyré, Gabriel
HAL ID: 1211
*
hal.structure.identifierGraduate School of Informatics [Kyoto]
dc.contributor.authorCuturi, Marco
HAL ID: 3354
*
dc.date.accessioned2015-04-08T09:33:25Z
dc.date.available2015-04-08T09:33:25Z
dc.date.issued2015
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/14905
dc.descriptionLNCS n°9123
dc.language.isoenen
dc.subjectTransport of Neuroimaging Data
dc.subject.ddc006.3en
dc.titleFast Optimal Transport Averaging of Neuroimaging Data
dc.typeCommunication / Conférence
dc.contributor.editoruniversityotherGraduate School of Informatics;Japon
dc.contributor.editoruniversityotherNeuroSpin, CEA Saclay;France
dc.contributor.editoruniversityotherInstitut Mines-Télécom, Telecom ParisTech, CNRS LTCI;France
dc.description.abstractenKnowing how the Human brain is anatomically and function-ally organized at the level of a group of healthy individuals or patientsis the primary goal of neuroimaging research. Yet computing an averageof brain imaging data de ned over a voxel grid or a triangulation re-mains a challenge. Data are large, the geometry of the brain is complexand the between subjects variability leads to spatially or temporally non-overlapping e ects of interest. To address the problem of variability, dataare commonly smoothed before performing a linear group averaging. Inthis work we build on ideas originally introduced by Kantorovich [18] topropose a new algorithm that can average e ciently non-normalized datade ned over arbitrary discrete domains using transportation metrics. Weshow how Kantorovich means can be linked to Wasserstein barycenters inorder to take advantage of the entropic smoothing approach used by [7].It leads to a smooth convex optimization problem and an algorithm withstrong convergence guarantees. We illustrate the versatility of this tooland its empirical behavior on functional neuroimaging data, functionalMRI and magnetoencephalography (MEG) source estimates, de ned onvoxel grids and triangulations of the folded cortical surface.
dc.publisher.cityParisen
dc.identifier.citationpages261-272
dc.relation.ispartoftitleInformation Processing in Medical Imaging 24th International Conference, IPMI 2015, Sabhal Mor Ostaig, Isle of Skye, UK, June 28 - July 3, 2015, Proceedings
dc.relation.ispartofeditorSebastien Ourselin, Daniel C. Alexander, Carl-Fredrik Westin, M. Jorge Cardoso
dc.relation.ispartofpublnameSpringer
dc.relation.ispartofpublcityBerlin Heidelberg
dc.relation.ispartofdate2015
dc.relation.ispartofurl10.1007/978-3-319-19992-4
dc.identifier.urlsitehttps://arxiv.org/abs/1503.08596v2
dc.subject.ddclabelIntelligence artificielleen
dc.relation.ispartofisbn978-3-319-19991-7
dc.description.submittednonen
dc.identifier.doi10.1007/978-3-319-19992-4_20
dc.description.ssrncandidatenon
dc.description.halcandidateoui
dc.description.readershiprecherche
dc.description.audienceInternational
dc.date.updated2016-10-10T07:46:57Z
hal.author.functionaut
hal.author.functionaut
hal.author.functionaut


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record