Links to Exploration step
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Morphological alterations in the caudate, putamen, pallidum, and thalamus in Parkinson's disease</title><author><name sortKey="Garg, Amanmeet" sort="Garg, Amanmeet" uniqKey="Garg A" first="Amanmeet" last="Garg">Amanmeet Garg</name><affiliation><nlm:aff id="aff1"><institution>Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University</institution><country>Burnaby, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Appel Cresswell, Silke" sort="Appel Cresswell, Silke" uniqKey="Appel Cresswell S" first="Silke" last="Appel-Cresswell">Silke Appel-Cresswell</name><affiliation><nlm:aff id="aff2"><institution>Neurology, Pacific Parkinson's Research Center, University of British Columbia</institution><country>Vancouver, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Popuri, Karteek" sort="Popuri, Karteek" uniqKey="Popuri K" first="Karteek" last="Popuri">Karteek Popuri</name><affiliation><nlm:aff id="aff1"><institution>Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University</institution><country>Burnaby, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Mckeown, Martin J" sort="Mckeown, Martin J" uniqKey="Mckeown M" first="Martin J." last="Mckeown">Martin J. Mckeown</name><affiliation><nlm:aff id="aff2"><institution>Neurology, Pacific Parkinson's Research Center, University of British Columbia</institution><country>Vancouver, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Beg, Mirza Faisal" sort="Beg, Mirza Faisal" uniqKey="Beg M" first="Mirza Faisal" last="Beg">Mirza Faisal Beg</name><affiliation><nlm:aff id="aff1"><institution>Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University</institution><country>Burnaby, BC, Canada</country></nlm:aff></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">PMC</idno><idno type="pmid">25873854</idno><idno type="pmc">4379878</idno><idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4379878</idno><idno type="RBID">PMC:4379878</idno><idno type="doi">10.3389/fnins.2015.00101</idno><date when="2015">2015</date><idno type="wicri:Area/Pmc/Corpus">000986</idno><idno type="wicri:explorRef" wicri:stream="Pmc" wicri:step="Corpus" wicri:corpus="PMC">000986</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Morphological alterations in the caudate, putamen, pallidum, and thalamus in Parkinson's disease</title><author><name sortKey="Garg, Amanmeet" sort="Garg, Amanmeet" uniqKey="Garg A" first="Amanmeet" last="Garg">Amanmeet Garg</name><affiliation><nlm:aff id="aff1"><institution>Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University</institution><country>Burnaby, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Appel Cresswell, Silke" sort="Appel Cresswell, Silke" uniqKey="Appel Cresswell S" first="Silke" last="Appel-Cresswell">Silke Appel-Cresswell</name><affiliation><nlm:aff id="aff2"><institution>Neurology, Pacific Parkinson's Research Center, University of British Columbia</institution><country>Vancouver, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Popuri, Karteek" sort="Popuri, Karteek" uniqKey="Popuri K" first="Karteek" last="Popuri">Karteek Popuri</name><affiliation><nlm:aff id="aff1"><institution>Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University</institution><country>Burnaby, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Mckeown, Martin J" sort="Mckeown, Martin J" uniqKey="Mckeown M" first="Martin J." last="Mckeown">Martin J. Mckeown</name><affiliation><nlm:aff id="aff2"><institution>Neurology, Pacific Parkinson's Research Center, University of British Columbia</institution><country>Vancouver, BC, Canada</country></nlm:aff></affiliation></author><author><name sortKey="Beg, Mirza Faisal" sort="Beg, Mirza Faisal" uniqKey="Beg M" first="Mirza Faisal" last="Beg">Mirza Faisal Beg</name><affiliation><nlm:aff id="aff1"><institution>Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University</institution><country>Burnaby, BC, Canada</country></nlm:aff></affiliation></author></analytic><series><title level="j">Frontiers in Neuroscience</title><idno type="ISSN">1662-4548</idno><idno type="eISSN">1662-453X</idno><imprint><date when="2015">2015</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en"><p>Like many neurodegenerative diseases, the clinical symptoms of Parkinsons disease (PD) do not manifest until significant progression of the disease has already taken place, motivating the need for sensitive biomarkers of the disease. While structural imaging is a potentially attractive method due to its widespread availability and non-invasive nature, global morphometric measures (e.g., volume) have proven insensitive to subtle disease change. Here we use individual surface displacements from deformations of an average surface model to capture disease related changes in shape of the subcortical structures in PD. Data were obtained from both the University of British Columbia (<italic>UBC</italic>) [<italic>n</italic> = 54 healthy controls (HC) and <italic>n</italic> = 55 Parkinsons disease (PD) patients] and the publicly available Parkinsons Progression Markers Initiative (<italic>PPMI</italic>) [<italic>n</italic> = 137 (HC) and <italic>n</italic> = 189 (PD)] database. A high dimensional non-rigid registration algorithm was used to register target segmentation labels (caudate, putamen, pallidum, and thalamus) to a set of segmentation labels defined on the average-template. The vertex-wise surface displacements were significantly different between PD and HC in thalamic and caudate structures. However, overall displacements did not correlate with disease severity, as assessed by the Unified Parkinson's Disease Rating Scale (UPDRS). The results from this study suggest disease-relevant shape abnormalities can be robustly detected in subcortical structures in PD. Future studies will be required to determine if shape changes in subcortical structures are seen in the prodromal phases of the disease.</p></div></front><back><div1 type="bibliography"><listBibl><biblStruct><analytic><author><name sortKey="Apostolova, L G" uniqKey="Apostolova L">L. G. Apostolova</name></author><author><name sortKey="Beyer, M" uniqKey="Beyer M">M. Beyer</name></author><author><name sortKey="Green, A E" uniqKey="Green A">A. E. Green</name></author><author><name sortKey="Hwang, K S" uniqKey="Hwang K">K. S. Hwang</name></author><author><name sortKey="Morra, J H" uniqKey="Morra J">J. H. Morra</name></author><author><name sortKey="Chou, Y Y" uniqKey="Chou Y">Y.-Y. Chou</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Beg, M" uniqKey="Beg M">M. Beg</name></author><author><name sortKey="Miller, M" uniqKey="Miller M">M. Miller</name></author><author><name sortKey="Trouve, A" uniqKey="Trouve A">A. Trouvé</name></author><author><name sortKey="Younes, L" uniqKey="Younes L">L. Younes</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Bergouignan, L" uniqKey="Bergouignan L">L. Bergouignan</name></author><author><name sortKey="Chupin, M" uniqKey="Chupin M">M. Chupin</name></author><author><name sortKey="Czechowska, Y" uniqKey="Czechowska Y">Y. Czechowska</name></author><author><name sortKey="Kinkingnehun, S" uniqKey="Kinkingnehun S">S. Kinkingnéhun</name></author><author><name sortKey="Lemogne, C" uniqKey="Lemogne C">C. Lemogne</name></author><author><name sortKey="Le Bastard, G" uniqKey="Le Bastard G">G. Le Bastard</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Brenneis, C" uniqKey="Brenneis C">C. Brenneis</name></author><author><name sortKey="Egger, K" uniqKey="Egger K">K. Egger</name></author><author><name sortKey="Scherfler, C" uniqKey="Scherfler C">C. Scherfler</name></author><author><name sortKey="Seppi, K" uniqKey="Seppi K">K. Seppi</name></author><author><name sortKey="Schocke, M" uniqKey="Schocke M">M. Schocke</name></author><author><name sortKey="Poewe, W" uniqKey="Poewe W">W. Poewe</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Burton, E J" uniqKey="Burton E">E. J. Burton</name></author><author><name sortKey="Mckeith, I G" uniqKey="Mckeith I">I. G. McKeith</name></author><author><name sortKey="Burn, D J" uniqKey="Burn D">D. J. Burn</name></author><author><name sortKey="O Brien, J T" uniqKey="O Brien J">J. T. O'Brien</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Cardoso, E F" uniqKey="Cardoso E">E. F. Cardoso</name></author><author><name sortKey="Maia, F M" uniqKey="Maia F">F. M. Maia</name></author><author><name sortKey="Fregni, F" uniqKey="Fregni F">F. Fregni</name></author><author><name sortKey="Myczkowski, M L" uniqKey="Myczkowski M">M. L. Myczkowski</name></author><author><name sortKey="Melo, L M" uniqKey="Melo L">L. M. Melo</name></author><author><name sortKey="Sato, J A R" uniqKey="Sato J">J. A. R. Sato</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Davatzikos, C" uniqKey="Davatzikos C">C. Davatzikos</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="De Lau, L" uniqKey="De Lau L">L. de Lau</name></author><author><name sortKey="Breteler, M" uniqKey="Breteler M">M. Breteler</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Duchesne, S" uniqKey="Duchesne S">S. Duchesne</name></author><author><name sortKey="Rolland, Y" uniqKey="Rolland Y">Y. Rolland</name></author><author><name sortKey="Verin, M" uniqKey="Verin M">M. Vérin</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Fischl, B" uniqKey="Fischl B">B. Fischl</name></author><author><name sortKey="Kouwe, A V D" uniqKey="Kouwe A">A. V. D. Kouwe</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Fischl, B" uniqKey="Fischl B">B. Fischl</name></author><author><name sortKey="Salat, D H" uniqKey="Salat D">D. H. Salat</name></author><author><name sortKey="Busa, E" uniqKey="Busa E">E. Busa</name></author><author><name sortKey="Albert, M" uniqKey="Albert M">M. Albert</name></author><author><name sortKey="Dieterich, M" uniqKey="Dieterich M">M. Dieterich</name></author><author><name sortKey="Haselgrove, C" uniqKey="Haselgrove C">C. Haselgrove</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Garg, A" uniqKey="Garg A">A. Garg</name></author><author><name sortKey="Wong, D" uniqKey="Wong D">D. Wong</name></author><author><name sortKey="Popuri, K" uniqKey="Popuri K">K. Popuri</name></author><author><name sortKey="Poskitt, K J" uniqKey="Poskitt K">K. J. Poskitt</name></author><author><name sortKey="Fitzpatrick, K" uniqKey="Fitzpatrick K">K. Fitzpatrick</name></author><author><name sortKey="Bjornson, B" uniqKey="Bjornson B">B. Bjornson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Ghaemi, M" uniqKey="Ghaemi M">M. Ghaemi</name></author><author><name sortKey="Hilker, R" uniqKey="Hilker R">R. Hilker</name></author><author><name sortKey="Rudolf, J" uniqKey="Rudolf J">J. Rudolf</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Hammers, A" uniqKey="Hammers A">A. Hammers</name></author><author><name sortKey="Allom, R" uniqKey="Allom R">R. Allom</name></author><author><name sortKey="Koepp, M J" uniqKey="Koepp M">M. J. Koepp</name></author><author><name sortKey="Free, S L" uniqKey="Free S">S. L. Free</name></author><author><name sortKey="Myers, R" uniqKey="Myers R">R. Myers</name></author><author><name sortKey="Lemieux, L" uniqKey="Lemieux L">L. Lemieux</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Khan, A R" uniqKey="Khan A">A. R. Khan</name></author><author><name sortKey="Beg, M F" uniqKey="Beg M">M. F. Beg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Khan, A" uniqKey="Khan A">A. Khan</name></author><author><name sortKey="Wang, L" uniqKey="Wang L">L. Wang</name></author><author><name sortKey="Beg, M" uniqKey="Beg M">M. Beg</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Krabbe, K" uniqKey="Krabbe K">K. Krabbe</name></author><author><name sortKey="Karlsborg, M" uniqKey="Karlsborg M">M. Karlsborg</name></author><author><name sortKey="Hansen, A" uniqKey="Hansen A">A. Hansen</name></author><author><name sortKey="Werdelin, L" uniqKey="Werdelin L">L. Werdelin</name></author><author><name sortKey="Mehlsen, J" uniqKey="Mehlsen J">J. Mehlsen</name></author><author><name sortKey="Larsson, H B W" uniqKey="Larsson H">H. B. W. Larsson</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Li, W" uniqKey="Li W">W. Li</name></author><author><name sortKey="Liu, J" uniqKey="Liu J">J. Liu</name></author><author><name sortKey="Skidmore, F" uniqKey="Skidmore F">F. Skidmore</name></author><author><name sortKey="Liu, Y" uniqKey="Liu Y">Y. Liu</name></author><author><name sortKey="Tian, J" uniqKey="Tian J">J. Tian</name></author><author><name sortKey="Li, K" uniqKey="Li K">K. Li</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Lilliefors, H" uniqKey="Lilliefors H">H. Lilliefors</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Mckeown, M J" uniqKey="Mckeown M">M. J. McKeown</name></author><author><name sortKey="Uthama, A" uniqKey="Uthama A">A. Uthama</name></author><author><name sortKey="Abugharbieh, R" uniqKey="Abugharbieh R">R. Abugharbieh</name></author><author><name sortKey="Palmer, S" uniqKey="Palmer S">S. Palmer</name></author><author><name sortKey="Lewis, M" uniqKey="Lewis M">M. Lewis</name></author><author><name sortKey="Huang, X" uniqKey="Huang X">X. Huang</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Nagano Saito, A" uniqKey="Nagano Saito A">A. Nagano-Saito</name></author><author><name sortKey="Washimi, Y" uniqKey="Washimi Y">Y. Washimi</name></author><author><name sortKey="Arahata, Y" uniqKey="Arahata Y">Y. Arahata</name></author><author><name sortKey="Kachi, T" uniqKey="Kachi T">T. Kachi</name></author><author><name sortKey="Lerch, J P" uniqKey="Lerch J">J. P. Lerch</name></author><author><name sortKey="Evans, A C" uniqKey="Evans A">A. C. Evans</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Obeso, J A" uniqKey="Obeso J">J. A. Obeso</name></author><author><name sortKey="Rodriguez Oroz, M C" uniqKey="Rodriguez Oroz M">M. C. Rodriguez-Oroz</name></author><author><name sortKey="Stamelou, M" uniqKey="Stamelou M">M. Stamelou</name></author><author><name sortKey="Bhatia, K P" uniqKey="Bhatia K">K. P. Bhatia</name></author><author><name sortKey="Burn, D J" uniqKey="Burn D">D. J. Burn</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Paviour, D C" uniqKey="Paviour D">D. C. Paviour</name></author><author><name sortKey="Price, S L" uniqKey="Price S">S. L. Price</name></author><author><name sortKey="Jahanshahi, M" uniqKey="Jahanshahi M">M. Jahanshahi</name></author><author><name sortKey="Lees, A J" uniqKey="Lees A">A. J. Lees</name></author><author><name sortKey="Fox, N C" uniqKey="Fox N">N. C. Fox</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Pitcher, T L" uniqKey="Pitcher T">T. L. Pitcher</name></author><author><name sortKey="Melzer, T R" uniqKey="Melzer T">T. R. Melzer</name></author><author><name sortKey="Macaskill, M R" uniqKey="Macaskill M">M. R. MacAskill</name></author><author><name sortKey="Graham, C F" uniqKey="Graham C">C. F. Graham</name></author><author><name sortKey="Livingston, L" uniqKey="Livingston L">L. Livingston</name></author><author><name sortKey="Keenan, R J" uniqKey="Keenan R">R. J. Keenan</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Pol, H" uniqKey="Pol H">H. Pol</name></author><author><name sortKey="Van Der Flier, W" uniqKey="Van Der Flier W">W. van der Flier</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Redgrave, P" uniqKey="Redgrave P">P. Redgrave</name></author><author><name sortKey="Rodriguez, M" uniqKey="Rodriguez M">M. Rodriguez</name></author><author><name sortKey="Smith, Y" uniqKey="Smith Y">Y. Smith</name></author><author><name sortKey="Rodriguez Oroz, M C" uniqKey="Rodriguez Oroz M">M. C. Rodriguez-Oroz</name></author><author><name sortKey="Lehericy, S" uniqKey="Lehericy S">S. Lehericy</name></author><author><name sortKey="Bergman, H" uniqKey="Bergman H">H. Bergman</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Schulz, J B" uniqKey="Schulz J">J. B. Schulz</name></author><author><name sortKey="Skalej, M" uniqKey="Skalej M">M. Skalej</name></author><author><name sortKey="Wedekind, D" uniqKey="Wedekind D">D. Wedekind</name></author><author><name sortKey="Luft, A R" uniqKey="Luft A">A. R. Luft</name></author><author><name sortKey="Abele, M" uniqKey="Abele M">M. Abele</name></author><author><name sortKey="Voigt, K" uniqKey="Voigt K">K. Voigt</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Sterling, N W" uniqKey="Sterling N">N. W. Sterling</name></author><author><name sortKey="Du, G" uniqKey="Du G">G. Du</name></author><author><name sortKey="Lewis, M M" uniqKey="Lewis M">M. M. Lewis</name></author><author><name sortKey="Dimaio, C" uniqKey="Dimaio C">C. Dimaio</name></author><author><name sortKey="Kong, L" uniqKey="Kong L">L. Kong</name></author><author><name sortKey="Eslinger, P J" uniqKey="Eslinger P">P. J. Eslinger</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Tzarouchi, L C" uniqKey="Tzarouchi L">L. C. Tzarouchi</name></author><author><name sortKey="Astrakas, L G" uniqKey="Astrakas L">L. G. Astrakas</name></author><author><name sortKey="Konitsiotis, S" uniqKey="Konitsiotis S">S. Konitsiotis</name></author><author><name sortKey="Tsouli, S" uniqKey="Tsouli S">S. Tsouli</name></author><author><name sortKey="Margariti, P" uniqKey="Margariti P">P. Margariti</name></author><author><name sortKey="Zikou, A" uniqKey="Zikou A">A. Zikou</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Worsley, K J" uniqKey="Worsley K">K. J. Worsley</name></author><author><name sortKey="Marrett, S" uniqKey="Marrett S">S. Marrett</name></author><author><name sortKey="Neelin, P" uniqKey="Neelin P">P. Neelin</name></author><author><name sortKey="Vandal, A C" uniqKey="Vandal A">A. C. Vandal</name></author><author><name sortKey="Friston, K J" uniqKey="Friston K">K. J. Friston</name></author><author><name sortKey="Evans, A C" uniqKey="Evans A">A. C. Evans</name></author></analytic></biblStruct><biblStruct><analytic><author><name sortKey="Worsley, K J" uniqKey="Worsley K">K. J. Worsley</name></author><author><name sortKey="Taylor, J E" uniqKey="Taylor J">J. E. Taylor</name></author><author><name sortKey="Tomaiuolo, F" uniqKey="Tomaiuolo F">F. Tomaiuolo</name></author><author><name sortKey="Lerch, J" uniqKey="Lerch J">J. Lerch</name></author></analytic></biblStruct></listBibl></div1></back></TEI><pmc article-type="research-article"><pmc-dir>properties open_access</pmc-dir>
<front><journal-meta><journal-id journal-id-type="nlm-ta">Front Neurosci</journal-id><journal-id journal-id-type="iso-abbrev">Front Neurosci</journal-id><journal-id journal-id-type="publisher-id">Front. Neurosci.</journal-id><journal-title-group><journal-title>Frontiers in Neuroscience</journal-title></journal-title-group><issn pub-type="ppub">1662-4548</issn><issn pub-type="epub">1662-453X</issn><publisher><publisher-name>Frontiers Media S.A.</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="pmid">25873854</article-id><article-id pub-id-type="pmc">4379878</article-id><article-id pub-id-type="doi">10.3389/fnins.2015.00101</article-id><article-categories><subj-group subj-group-type="heading"><subject>Neuroscience</subject><subj-group><subject>Original Research</subject></subj-group></subj-group></article-categories><title-group><article-title>Morphological alterations in the caudate, putamen, pallidum, and thalamus in Parkinson's disease</article-title></title-group><contrib-group><contrib contrib-type="author"><name><surname>Garg</surname><given-names>Amanmeet</given-names></name><xref ref-type="aff" rid="aff1"><sup>1</sup></xref><uri xlink:type="simple" xlink:href="http://community.frontiersin.org/people/u/203099"></uri></contrib><contrib contrib-type="author"><name><surname>Appel-Cresswell</surname><given-names>Silke</given-names></name><xref ref-type="aff" rid="aff2"><sup>2</sup></xref><uri xlink:type="simple" xlink:href="http://community.frontiersin.org/people/u/203185"></uri></contrib><contrib contrib-type="author"><name><surname>Popuri</surname><given-names>Karteek</given-names></name><xref ref-type="aff" rid="aff1"><sup>1</sup></xref><uri xlink:type="simple" xlink:href="http://community.frontiersin.org/people/u/223486"></uri></contrib><contrib contrib-type="author"><name><surname>McKeown</surname><given-names>Martin J.</given-names></name><xref ref-type="aff" rid="aff2"><sup>2</sup></xref><uri xlink:type="simple" xlink:href="http://community.frontiersin.org/people/u/9879"></uri></contrib><contrib contrib-type="author"><name><surname>Beg</surname><given-names>Mirza Faisal</given-names></name><xref ref-type="aff" rid="aff1"><sup>1</sup></xref><xref ref-type="author-notes" rid="fn001"><sup>*</sup></xref><uri xlink:type="simple" xlink:href="http://community.frontiersin.org/people/u/156181"></uri></contrib></contrib-group><aff id="aff1"><sup>1</sup><institution>Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University</institution><country>Burnaby, BC, Canada</country></aff><aff id="aff2"><sup>2</sup><institution>Neurology, Pacific Parkinson's Research Center, University of British Columbia</institution><country>Vancouver, BC, Canada</country></aff><author-notes><fn fn-type="edited-by"><p>Edited by: Christine Lucas Tardif, Max Planck Institute for Human Cognitive and Brain Sciences, Germany</p></fn><fn fn-type="edited-by"><p>Reviewed by: Yiming Xiao, McGill University, Canada; Natasha Lepore, University of Southern California, Children's Hospital Los Angeles, USA</p></fn><corresp id="fn001">*Correspondence: Mirza Faisal Beg, School of Engineering Science, Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6, Canada <email xlink:type="simple">mfbeg@sfu.ca</email></corresp><fn fn-type="other" id="fn002"><p>This article was submitted to Brain Imaging Methods, a section of the journal Frontiers in Neuroscience</p></fn></author-notes><pub-date pub-type="epub"><day>31</day><month>3</month><year>2015</year></pub-date><pub-date pub-type="collection"><year>2015</year></pub-date><volume>9</volume><elocation-id>101</elocation-id><history><date date-type="received"><day>24</day><month>11</month><year>2014</year></date><date date-type="accepted"><day>10</day><month>3</month><year>2015</year></date></history><permissions><copyright-statement>Copyright © 2015 Garg, Appel-Cresswell, Popuri, McKeown and Beg.</copyright-statement><copyright-year>2015</copyright-year><license license-type="open-access" xlink:href="http://creativecommons.org/licenses/by/4.0/"><license-p>This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.</license-p></license></permissions><abstract><p>Like many neurodegenerative diseases, the clinical symptoms of Parkinsons disease (PD) do not manifest until significant progression of the disease has already taken place, motivating the need for sensitive biomarkers of the disease. While structural imaging is a potentially attractive method due to its widespread availability and non-invasive nature, global morphometric measures (e.g., volume) have proven insensitive to subtle disease change. Here we use individual surface displacements from deformations of an average surface model to capture disease related changes in shape of the subcortical structures in PD. Data were obtained from both the University of British Columbia (<italic>UBC</italic>) [<italic>n</italic> = 54 healthy controls (HC) and <italic>n</italic> = 55 Parkinsons disease (PD) patients] and the publicly available Parkinsons Progression Markers Initiative (<italic>PPMI</italic>) [<italic>n</italic> = 137 (HC) and <italic>n</italic> = 189 (PD)] database. A high dimensional non-rigid registration algorithm was used to register target segmentation labels (caudate, putamen, pallidum, and thalamus) to a set of segmentation labels defined on the average-template. The vertex-wise surface displacements were significantly different between PD and HC in thalamic and caudate structures. However, overall displacements did not correlate with disease severity, as assessed by the Unified Parkinson's Disease Rating Scale (UPDRS). The results from this study suggest disease-relevant shape abnormalities can be robustly detected in subcortical structures in PD. Future studies will be required to determine if shape changes in subcortical structures are seen in the prodromal phases of the disease.</p></abstract><kwd-group><kwd>LDDMM</kwd><kwd>shape analysis</kwd><kwd>surface displacement</kwd><kwd>Parkinson's disease</kwd><kwd>prediction</kwd><kwd>brain MRI</kwd></kwd-group><counts><fig-count count="6"></fig-count><table-count count="4"></table-count><equation-count count="2"></equation-count><ref-count count="31"></ref-count><page-count count="14"></page-count><word-count count="8154"></word-count></counts></article-meta></front><body><sec id="s1"><title>1. Introduction</title><p>Parkinson's disease (PD) is the second most common age related neurodegenerative disorder after Alzheimers disease (de Lau and Breteler, <xref rid="B8" ref-type="bibr">2006</xref>). Routine clinical MRI is rarely used in diagnosis, and is often used only to rule out other conditions that may mimic PD (e.g., vascular Parkinsonism). Although there are no gross structural abnormalities seen in PD, the use of structural MRI is still potentially attractive as a biomarker because of the ubiquitous nature of the technology. Several studies have suggested subtle morphological alterations such as atrophy in the putamen and/or caudate (Schulz et al., <xref rid="B27" ref-type="bibr">1999</xref>; Ghaemi et al., <xref rid="B13" ref-type="bibr">2002</xref>; Krabbe et al., <xref rid="B17" ref-type="bibr">2005</xref>; Pitcher et al., <xref rid="B24" ref-type="bibr">2012</xref>). Most of these studies have looked at overall volume as a measure of atrophy as it easy to measure, invariant to position of the subject in the scanner, and, if appropriately normalized, directly comparable across subjects. However, in some structures such as the thalamus, volume may actually increase as a compensatory mechanism when cortical regions are damaged (Pol and van der Flier, <xref rid="B25" ref-type="bibr">2000</xref>), complicating overall volume as a marker of disease progression. This may be why the thalami undergo significant shape change with PD, even when no significant difference in volume can be detected (McKeown et al., <xref rid="B20" ref-type="bibr">2008</xref>), presumably on the basis of specific nuclei being affected and/or compensatory hypertrophy of other regions.</p><p>Commonly-used brain morphometric analysis methods to assess progression of neurodegenerative disease can be categorized into three general types: voxel-based, surface-based, and deformation-based methods. By far, the majority of morphometry studies to date have been based on voxel-based morphometry (VBM), because of its conceptual simplicity and widespread availability of suitable software. In VBM, subjects brain images are registered to a common template image, and statistical analyses are performed on a voxel-by-voxel basis on the registered subject brain images. This technique has demonstrated cortical loss in some brain areas in PD (Burton et al., <xref rid="B5" ref-type="bibr">2005</xref>; Nagano-Saito et al., <xref rid="B21" ref-type="bibr">2005</xref>) and in other PD-related diseases (e.g., Multi system atrophy) (Paviour et al., <xref rid="B23" ref-type="bibr">2006</xref>; Brenneis et al., <xref rid="B4" ref-type="bibr">2007</xref>; Tzarouchi et al., <xref rid="B29" ref-type="bibr">2010</xref>). Nevertheless, there is widespread recognition of the limitations of VBM: it tends to find focal changes as opposed to more spatially distributed changes (Davatzikos, <xref rid="B7" ref-type="bibr">2004</xref>), and may be insensitive to subtle morphological alterations (Bergouignan et al., <xref rid="B3" ref-type="bibr">2009</xref>).</p><p>An alternative approach is to use surface-based morphometry (SBM) for measuring shape changes. In SBM, a surface representation of a structural boundary is investigated rather than at the level of the individual voxels. Using such a method, PD related shape and volume changes in the hippocampus, caudate and ventricles can be detected (Apostolova et al., <xref rid="B1" ref-type="bibr">2010</xref>). In a similar approach, deformation-based morphometry (DBM), the deformation fields obtained from one-to-one non-rigid registrations are analyzed in place of the final registered images (Duchesne et al., <xref rid="B9" ref-type="bibr">2009</xref>). In our work, we use a SBM method where we model the change in shape as the deformation of a template surface to the individual target surfaces for the disease and control groups. The surface displacement metric obtained as a signed normal component of the displacement vector (from template to target surface) at each vertex on the template surface captures the deformation information. This metric directly models deformation on the surface of the template.</p><p>Here we present a SBM method that incorporates anatomically-defined regions of interest (ROIs). Accurate segmentation labels for subcortical ROIs were obtained via an automated registration based segmentation process (Khan et al., <xref rid="B16" ref-type="bibr">2008</xref>). These labels were then registered to a prototype via a non-linear registration algorithm (Beg et al., <xref rid="B2" ref-type="bibr">2005</xref>). The template for this study was generated via an alternating registration and averaging process (Khan and Beg, <xref rid="B15" ref-type="bibr">2009</xref>), that encapsulates information from the entire cohort and thus with minimal bias to data from a specific subject. Given a mean template, the surface displacement obtained from taking the difference in coordinates between the reference (template) and the deformed surfaces on a vertex-by vertex basis represent a feature for subsequent classification. Based on prior work (McKeown et al., <xref rid="B20" ref-type="bibr">2008</xref>; Apostolova et al., <xref rid="B1" ref-type="bibr">2010</xref>) we study the change in shape of caudate, thalamus, putamen, and pallidum structures due to PD. We processed the imaging data from two cohorts with PD patients and healthy control groups, and analyzed the surface displacement feature for the group level difference in the features.</p><p>The rest of the paper is organized as follows: in Section 2 the data and processing methods are introduced, followed by the results of experiments in Section 3. The results are further discussed and concluded in Section 4.</p></sec><sec id="s2"><title>2. Materials and methods</title><p>In this section we discuss the methods for extraction of shape features from the MRI data for the individual subcortical ROIs. The feature extraction process consists of a number of steps applied sequentially to perform the tasks of data pre-processing, construction of a prototype, diffeomorphic registration followed by final feature extraction. These features are then tested for differences between the PD and HC groups. The following sub-sections provide a description of these steps in detail.</p><sec><title>2.1. Subjects and scans</title><sec><title>2.1.1. University of British Columbia (<italic>UBC</italic>) dataset</title><p>Data were taken from 55 non-demented PD subjects and 54 healthy subjects seen at the <italic>UBC Pacific Parkinson's Research Center</italic>. Table <xref ref-type="table" rid="T1">1</xref> gives the clinical and demographic details of the PD and control groups. Written informed consent was obtained from all subjects prior to participation in the study. This study was approved by the Clinical Research Ethics Board of the University of British Columbia and conforms to the Declaration of Helsinki.</p><table-wrap id="T1" position="float"><label>Table 1</label><caption><p><bold>Demographics for the data in the <italic>UBC</italic> dataset</bold>.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="left" valign="top" rowspan="1" colspan="1"><bold>Variable</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>Healthy control</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>PD subjects</bold></th></tr></thead><tbody><tr><td align="left" valign="top" rowspan="1" colspan="1">Number</td><td align="center" valign="top" rowspan="1" colspan="1">54</td><td align="center" valign="top" rowspan="1" colspan="1">55</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">Age</td><td align="center" valign="top" rowspan="1" colspan="1">46.16 ± 16.80</td><td align="center" valign="top" rowspan="1" colspan="1">64 ± 8.11</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">Gender</td><td align="center" valign="top" rowspan="1" colspan="1">23M/25F</td><td align="center" valign="top" rowspan="1" colspan="1">36M/19F</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">UPDRS</td><td align="center" valign="top" rowspan="1" colspan="1">N.A.</td><td align="center" valign="top" rowspan="1" colspan="1">25.89 ± 11.02</td></tr></tbody></table></table-wrap><p>Patients were examined after overnight medication withdrawal with >12 h for L-dopa and >18 h for dopamine agonists. Exclusion criteria included atypical parkinson's (cerebellar ataxia, prominent dementia, early postural instability), symmetrical onset of symptoms, evidence of severe memory impairment or signs of dementia [Montreal Cognitive Assessment (MoCA) scores < 24] or a history of cerebrovascular disease or other neurological disorders. The Unified Parkinson's Disease Rating Scale (UPDRS) motor scale data were recorded and used in further analysis.</p><p>Subjects were scanned using a Philips Achieva 3.0 T scanner (Philips, Best, The Netherlands), with a 8-channel head coil. A memory foam pillow was used to minimize head motion. The acquisition parameters for the 3D T1-TFE sequence were as follows: 170 axial slices, repetition time = 7.7 ms, echo time = 3.6 ms, flip angle 8°, field of view 256 × 200 mm, acquired matrix size 256 × 200, and voxel size 1 × 1 × 1 mm<sup>3</sup>.</p></sec><sec><title>2.1.2. Parkinson's progression marker initiative (<italic>PPMI</italic>) cohort</title><p>The data available under the <italic>PPMI</italic> project was obtained from the LONI Image data archive (<ext-link ext-link-type="uri" xlink:href="https://ida.loni.usc.edu">https://ida.loni.usc.edu</ext-link>). The data acquisition details are available on the website for the PPMI project and can be obtained from the url (<ext-link ext-link-type="uri" xlink:href="http://www.ppmi-info.org/">http://www.ppmi-info.org/</ext-link>). The data from the baseline visit was selected for the analysis. We randomly chose a subset of subjects so that they would be age-matched with other subject groups. The demographic details for this subset are presented in the Table <xref ref-type="table" rid="T2">2</xref>.</p><table-wrap id="T2" position="float"><label>Table 2</label><caption><p><bold>Demographics for the data in the <italic>PPMI</italic> dataset</bold>.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="left" valign="top" rowspan="1" colspan="1"><bold>Variable</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>Healthy control</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>PD patients</bold></th></tr></thead><tbody><tr><td align="left" valign="top" rowspan="1" colspan="1">Number</td><td align="center" valign="top" rowspan="1" colspan="1">137</td><td align="center" valign="top" rowspan="1" colspan="1">189</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">Age</td><td align="center" valign="top" rowspan="1" colspan="1">63.85 ± 7.46</td><td align="center" valign="top" rowspan="1" colspan="1">68.02 ± 4.77</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">Gender</td><td align="center" valign="top" rowspan="1" colspan="1">75M/62F</td><td align="center" valign="top" rowspan="1" colspan="1">115M/74F</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">UPDRS</td><td align="center" valign="top" rowspan="1" colspan="1">N.A.</td><td align="center" valign="top" rowspan="1" colspan="1">33.62 ± 13.36</td></tr></tbody></table></table-wrap><p>The <italic>PPMI</italic> cohort is a multicenter study where multiple scanners and imaging protocols are used. An example protocol for a MPRAGE sequence as used in the study is as follows: Siemens Magnetom TrioTim sungo MR B17 3T scanner (Siemens, Germany). T1 weighted images were acquired with the MPRAGE sequence with the following parameters: 176 axial slices, repetition time = 2300 ms, echo time = 2.98 ms, flip angle = 9°, field of view = 256 mm, acquired matrix size = 256 × 200, and voxel size 1 × 1 × 1 mm<sup>3</sup>. Some other scanners used in the study include GE Signa 3.0T, GE Discovery 3.0T (G.E., USA), Philips Acheiva 1.5T (Philips, Netherlands), Siemens TrioTim 3.0T, Siemens Symphony 1.5T, Siemens Verio 3.0T, Siemens Espree 1.5T (Siemens, Germany). Extensive pre-processing and data normalization was performed by the members of the PPMI-core group prior to release of the data for public use. For detailed image acquisition protocols and scanner specific preprocessing the reader is referred to the website of the PPMI cohort at <ext-link ext-link-type="uri" xlink:href="http://www.ppmi-info.org">www.ppmi-info.org</ext-link></p></sec></sec><sec><title>2.2. Data analysis process</title><p>In this section, we present the details of the image analysis steps that was used to obtain the surface displacement data for individual subcortical structures from the raw MRI data. We briefly describe the registration algorithm, followed by the methods for segmentation of anatomical structures, prototype creation and surface displacement computation. The process flow diagram (Figure <xref ref-type="fig" rid="F1">1</xref>) illustrates the stages of the process.</p><fig id="F1" position="float"><label>Figure 1</label><caption><p><bold>A process flow diagram for the computation of the surface displacement feature from the raw MRI data</bold>.</p></caption><graphic xlink:href="fnins-09-00101-g0001"></graphic></fig><sec><title>2.2.1. Large-deformation diffeomorphic metric mapping (LDDMM)</title><p>Diffeomorphic registration methods are desirable in processing of medical imaging data because of their inherent smoothness and ability to model large and small displacements. Here we briefly describe the LDDMM (Beg et al., <xref rid="B2" ref-type="bibr">2005</xref>) process, which generates a diffeomorphic transformation by minimizing the difference between the source and transformed target images.</p><p>Let us define Ω ⊂ ℝ<sup>3</sup> as the coordinate space of the source image, and <italic>G</italic>:Ω ↔ Ω as the set of diffeomorphic transformations on Ω. The LDDMM algorithm seeks a geodesic ϕ : [0, 1] → <italic>G</italic> where each point ϕ<sub><italic>t</italic></sub> ∈ <italic>G, t</italic> ∈ [0, 1] is a diffeomorphism on the domain Ω. Then the source image <italic>I</italic><sub>0</sub> evolves along the path to the target image <italic>I</italic><sub>1</sub> according to ϕ<sub><italic>t</italic></sub><italic>I</italic><sub>0</sub> = <italic>I</italic><sub>0</sub> ◦ ϕ<sup>−1</sup><sub><italic>t</italic></sub>. At the endpoint <italic>t</italic> = 1, the source <italic>I</italic><sub>0</sub> is connected to the target image via <italic>I</italic><sub>1</sub> = ϕ<sub>1</sub><italic>I</italic><sub>0</sub> = <italic>I</italic><sub>0</sub> ◦ ϕ<sup>−1</sup><sub>1</sub>. The associated velocity field <italic>v</italic>, taken from the space of smooth velocity fields <italic>V</italic> on the domain Ω ⊂ ℝ<sup>3</sup>, is a solution to the differential equation <inline-formula><mml:math id="M3"><mml:mrow><mml:msub><mml:mover accent="true"><mml:mi>ϕ</mml:mi><mml:mo>˙</mml:mo></mml:mover><mml:mi>t</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> = <italic>v</italic><sub><italic>t</italic></sub> (ϕ<sub><italic>t</italic></sub>), <italic>t</italic> ∈ [0, 1] satisfying</p><disp-formula id="E1"><label>(1)</label><mml:math id="M1"><mml:mtable columnalign="left"><mml:mtr><mml:mtd><mml:mover accent="true"><mml:mi>v</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>=</mml:mo><mml:mi>v</mml:mi><mml:mo>:</mml:mo><mml:msub><mml:mover accent="true"><mml:mi>ϕ</mml:mi><mml:mo>˙</mml:mo></mml:mover><mml:mi>t</mml:mi></mml:msub></mml:mtd></mml:mtr><mml:mtr><mml:mtd><mml:mtext> </mml:mtext><mml:mo>=</mml:mo><mml:mi>arg</mml:mi><mml:mtext> </mml:mtext><mml:msub><mml:mtext>min</mml:mtext><mml:mrow><mml:msub><mml:mi>V</mml:mi><mml:mi>t</mml:mi></mml:msub><mml:mo stretchy="false">(</mml:mo><mml:msub><mml:mi>ϕ</mml:mi><mml:mi>t</mml:mi></mml:msub><mml:mo stretchy="false">)</mml:mo></mml:mrow></mml:msub><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mstyle displaystyle="true"><mml:mrow><mml:msubsup><mml:mo>∫</mml:mo><mml:mn>0</mml:mn><mml:mn>1</mml:mn></mml:msubsup><mml:mrow><mml:mo>‖</mml:mo><mml:msub><mml:mi>v</mml:mi><mml:mi>t</mml:mi></mml:msub><mml:msup><mml:mo>‖</mml:mo><mml:mn>2</mml:mn></mml:msup></mml:mrow></mml:mrow></mml:mstyle><mml:mtext>d</mml:mtext><mml:mi>t</mml:mi><mml:mo>+</mml:mo><mml:mfrac><mml:mn>1</mml:mn><mml:mrow><mml:msup><mml:mi>σ</mml:mi><mml:mn>2</mml:mn></mml:msup></mml:mrow></mml:mfrac><mml:mo>‖</mml:mo><mml:msub><mml:mi>I</mml:mi><mml:mn>0</mml:mn></mml:msub><mml:mo>°</mml:mo><mml:msubsup><mml:mi>ϕ</mml:mi><mml:mn>1</mml:mn><mml:mrow><mml:mo>−</mml:mo><mml:mn>1</mml:mn></mml:mrow></mml:msubsup><mml:mo>−</mml:mo><mml:msub><mml:mi>I</mml:mi><mml:mn>1</mml:mn></mml:msub><mml:msubsup><mml:mo>‖</mml:mo><mml:mrow><mml:msup><mml:mi>L</mml:mi><mml:mn>2</mml:mn></mml:msup></mml:mrow><mml:mn>2</mml:mn></mml:msubsup></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mtd></mml:mtr></mml:mtable></mml:math></disp-formula><p>By integrating the optimizer <inline-formula><mml:math id="M2"><mml:mover accent="true"><mml:mi>v</mml:mi><mml:mo>^</mml:mo></mml:mover></mml:math></inline-formula> of this cost function we get the optimal change of coordinates <inline-formula><mml:math id="M14"><mml:mrow><mml:mi>φ</mml:mi><mml:mo>=</mml:mo><mml:msubsup><mml:mi>ϕ</mml:mi><mml:mn>1</mml:mn><mml:mover accent="true"><mml:mi>v</mml:mi><mml:mo>^</mml:mo></mml:mover></mml:msubsup></mml:mrow></mml:math></inline-formula>. The superscript <italic>v</italic> in ϕ<sup><italic>v</italic></sup> is used to explicitly denote the dependence of ϕ on the associated velocity field <italic>v</italic>. The mapping <inline-formula><mml:math id="M4"><mml:mrow><mml:mi>φ</mml:mi><mml:mo>=</mml:mo><mml:msubsup><mml:mi>ϕ</mml:mi><mml:mn>1</mml:mn><mml:mover accent="true"><mml:mi>v</mml:mi><mml:mo>^</mml:mo></mml:mover></mml:msubsup></mml:mrow></mml:math></inline-formula> is guaranteed to be a diffeomorphism by enforcing sufficient smoothness on the elements of <italic>V</italic>. We do this by defining a norm on <italic>V</italic> through a 3 × 3 differential operator <italic>L</italic> of the type <italic>L</italic> = (αΔ + γ)<sup>α</sup><italic>I</italic><sub>3×3</sub> where α > 1.5 in 3D space such that ‖<italic>f</italic>‖<sub><italic>V</italic></sub> = ‖<italic>Lf</italic>‖<sub><italic>L</italic><sub>2</sub></sub>, and ‖ · ‖<sub><italic>L</italic><sub>2</sub></sub> is the standard <italic>L</italic><sub>2</sub> norm for the square integrable functions defined on Ω. The gradient of the cost function (Equation 1) is given by the following Freche derivative in <italic>V</italic>:</p><disp-formula id="E2"><label>(2)</label><mml:math id="M5"><mml:mrow><mml:msub><mml:mo>∇</mml:mo><mml:mi>v</mml:mi></mml:msub><mml:msub><mml:mi>E</mml:mi><mml:mi>t</mml:mi></mml:msub><mml:mo>=</mml:mo><mml:mn>2</mml:mn><mml:msub><mml:mover accent="true"><mml:mi>v</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mi>t</mml:mi></mml:msub><mml:mo>−</mml:mo><mml:mi>K</mml:mi><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mfrac><mml:mn>2</mml:mn><mml:mrow><mml:msup><mml:mi>σ</mml:mi><mml:mn>2</mml:mn></mml:msup></mml:mrow></mml:mfrac><mml:mo>|</mml:mo><mml:mi>D</mml:mi><mml:msubsup><mml:mi>ϕ</mml:mi><mml:mrow><mml:mi>t</mml:mi><mml:mo>,</mml:mo><mml:mn>1</mml:mn></mml:mrow><mml:mover accent="true"><mml:mi>v</mml:mi><mml:mo>^</mml:mo></mml:mover></mml:msubsup><mml:mo>|</mml:mo><mml:mo>∇</mml:mo><mml:msubsup><mml:mi>J</mml:mi><mml:mi>t</mml:mi><mml:mn>0</mml:mn></mml:msubsup><mml:mo stretchy="false">(</mml:mo><mml:msubsup><mml:mi>J</mml:mi><mml:mi>t</mml:mi><mml:mn>0</mml:mn></mml:msubsup><mml:mo>−</mml:mo><mml:msubsup><mml:mi>J</mml:mi><mml:mi>t</mml:mi><mml:mn>1</mml:mn></mml:msubsup><mml:mo stretchy="false">)</mml:mo></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula><p>Where <italic>J</italic><sup>0</sup><sub><italic>t</italic></sub> = <italic>I</italic><sub>0</sub> ◦ ϕ<sub><italic>t</italic>, 0</sub> and <italic>J</italic><sup>1</sup><sub><italic>t</italic></sub> = <italic>I</italic><sub>1</sub> ◦ ϕ<sub><italic>t</italic>, 1</sub>, |<italic>Dg</italic>| is the determinant of the Jacobian matrix and <italic>K</italic> is a compact self-adjoint operator <italic>K</italic> : <italic>L</italic><sub>2</sub> (Ω, ℝ<sup>3</sup>) → <italic>V</italic> uniquely defined by 〈<italic>a, b</italic>〉<sub><italic>L</italic><sub>2</sub></sub> = 〈<italic>K a, b</italic>〉<sub><italic>V</italic></sub> such that for any smooth vector field <italic>f</italic> ∈ <italic>V, k</italic>(<italic>L</italic><sup>†</sup><italic>L</italic>)<italic>f</italic> = <italic>f</italic> holds. Also <italic>L</italic><sup>†</sup> is the adjoint of <italic>L</italic> and the notation ϕ<sub><italic>s, t</italic></sub> = ϕ<sub><italic>s</italic></sub> · ϕ<sup>−1</sup><sub><italic>j</italic></sub> is employed. Finally the parameter <inline-formula><mml:math id="M6"><mml:mrow><mml:mfrac><mml:mn>1</mml:mn><mml:mrow><mml:msup><mml:mi>σ</mml:mi><mml:mn>2</mml:mn></mml:msup></mml:mrow></mml:mfrac></mml:mrow></mml:math></inline-formula> provides weighted optimization between the regularization and data matching components, and is chosen to be the same for all matchings.</p><p>In order to compute <italic>v</italic>, this variational gradient is used in the standard gradient descent procedure, yielding the update <italic>v</italic><sup><italic>n</italic> + 1</sup> = <italic>v<sup>n</sup></italic> − ε ∇<italic><sub>v<sup>n</sup></sub>E</italic> where <italic>n</italic> denotes the simulation number.</p><p>The optimal mapping ϕ<sub>0, 1</sub> is used in further steps to transform the images from the source space to the target image space.</p></sec><sec><title>2.2.2. FS + LDDMM segmentation</title><p>The FS + LDDMM segmentation steps (Khan et al., <xref rid="B16" ref-type="bibr">2008</xref>) combine the probabilistic labels obtained from the FreeSurfer (FS) program along with the LDDMM registration with ground truth template data, to create accurate labels for the subcortical ROIs in the target MRI data.</p><p>Freesurfer (v4.5.0) (Fischl et al., <xref rid="B11" ref-type="bibr">2002</xref>) was utilized to obtain initial segmentation of subcortical structures for each MRI image volume. With this process, non-brain tissue was removed from the images, followed by automated Talairach transformation and segmentation of caudate, pallidum, putamen, thalamus (Fischl et al., <xref rid="B11" ref-type="bibr">2002</xref>; Fischl and Kouwe, <xref rid="B10" ref-type="bibr">2004</xref>). Subsequently, an ROI was defined for each structure on the target and template MR images using FS labels and manual labels, respectively. These ROIs from target images were then aligned via an intensity-based affine transformation to those in the template images. A bounding box, predefined in the template space using the extent of the template FS labels plus a 12-voxel padding, was used to generate sub-volumes. These pre-processed MRI sub-volumes were then registered via the LDDMM method as described above (Section 2.2.1) to obtain the final segmentation labels.</p><p>Within the FS + LDDMM process, the LDDMM registration was performed in a multi-stage fashion, each stage using different image pairs and with each subsequent stage initialized with the velocity vector fields, of the previous stage. In the first stage, the FS labels of the hippocampus, amygdala, and lateral ventricles were used, in the second stage, Gaussian smoothed (σ = 5) MRI images were used, and in the final stage non-smoothed MRI images were used. The velocity vector fields were discretized into 5 timesteps. Finally, the atlas segmentations for each hemisphere were propagated to the target by applying the LDDMM and affine transformation, using linear interpolation to maintain precision when resampling the segmentations. The propagated segmentations from each template were fused with equal weights for each template. This resulted in a binary segmentation volume with the same dimensions and orientation as the original MRI in the native image space with each subcortical region of interest represented by a voxel intensity of 1 (intensity threshold = 127.5) and background intensity as 0.</p><p>The templates for segmentation of the MRI images in this study were obtained by manual segmentation of 6 healthy control subjects from the UBC scanner but not included in this study. The protocols for definition of structural boundaries for all the structures of interest were obtained from previously published work (Hammers et al., <xref rid="B14" ref-type="bibr">2003</xref>).</p></sec><sec><title>2.2.3. Segmentation quality check</title><p>The final labeling for each structure obtained from the segmentation process was checked for segmentation accuracy through an extensive quality control process. A surface model was fit for each binary segmentation volume via the marching cubes algorithm to provide a set of vertices and triangles representing the segmentation boundaries. These surfaces were overlayed on the original MRI images in the three orthogonal views to check for accuracy of segmentation labels (e.g., Figure <xref ref-type="fig" rid="F2">2</xref>). A visual verification of each such visualization was performed by an expert in neuro-anatomy. In conventional quality control approaches, the subjects with inaccurate segmentation for any structure are removed from the subsequent analysis. In our work, all subjects were found to have accurate segmentation labels and were included in the subsequent analyses. The demographics reported in Tables <xref ref-type="table" rid="T1">1</xref>, <xref ref-type="table" rid="T2">2</xref> present the set of subjects with accurate and acceptable segmentation labels. The detailed visualizations for all the subjects in the analysis are available at the website (<ext-link ext-link-type="uri" xlink:href="http://www.autobrainmapping.com">www.autobrainmapping.com</ext-link>).</p><fig id="F2" position="float"><label>Figure 2</label><caption><p><bold>Saggital <italic>upper left</italic> Coronal <italic>upper right</italic> Axial <italic>lower left</italic> views of the segmentation outlines overlayed on the corresponding MRI slices</bold>. The 3D surface renderings show the smooth surfaces <italic>lower right</italic>.</p></caption><graphic xlink:href="fnins-09-00101-g0002"></graphic></fig></sec><sec><title>2.2.4. Unbiased average template (prototype) generation</title><p>The choice of template for registration influences the accuracy of the surface displacement data. To this end we created a “prototype” for the cohort (an average template) from the binary segmentation labels obtained for individual structures (Khan and Beg, <xref rid="B15" ref-type="bibr">2009</xref>). Pre-processed data were obtained by affine alignment of each binary subcortical image to an initial template, followed by extraction of a sub-volume ROI in the prototype space. The process alternated between (1) registration (LDDMM) of individual ROIs to a template ROI and (2) computation of an average from the registered ROIs in the template space. The average computed in the previous step formed the template for registration in the next step. A healthy control subject was chosen as an initial template and the pipeline was run for three iterations to obtain the final, average unbiased template. A subset of normal subjects (<italic>n</italic> = 10) from the <italic>UBC</italic> dataset were selected for creation of the prototype.</p></sec><sec><title>2.2.5. Surface displacement</title><p>The prototype generated above (Section 2.2.4) was used for computation of the surface displacement data. The target binary labels were pre-registered to the prototype prior to non-rigid registration. Two different approaches were considered for the pre-registration step: (1) using a 6 degrees-of-freedom (DOF) rigid transformation (2) using a 9 DOF affine transformation. The rigid pre-registration approach corrects only for the translational and rotational discrepancies between the target structures, whereas the affine approach corrects the scale discrepancy as well. Both pre-registered sets (rigid and affine) were further used in the pipeline to extract two sets of the surface displacement features.</p><p>High dimensional non-rigid registration (LDDMM) from the prototype segmentation image, <italic>M<sub>avg</sub></italic>, to each pre-registered segmentation image <italic>M</italic><sub><italic>j</italic></sub> was performed to obtain the mapping ϕ<sub><italic>M<sub>avg</sub></italic>, <italic>M<sub>j</sub></italic></sub> = <italic>LDDMM</italic>(<italic>M<sub>avg</sub></italic>, <italic>M</italic><sub><italic>j</italic></sub>). The injected-surface was then computed as <inline-formula><mml:math id="M7"><mml:mrow><mml:msub><mml:mover accent="true"><mml:mi>S</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mi>i</mml:mi></mml:msub><mml:mo>=</mml:mo><mml:msubsup><mml:mi>ϕ</mml:mi><mml:mrow><mml:msub><mml:mi>M</mml:mi><mml:mrow><mml:mi>a</mml:mi><mml:mi>v</mml:mi><mml:mi>g</mml:mi></mml:mrow></mml:msub><mml:mo>,</mml:mo><mml:msub><mml:mi>M</mml:mi><mml:mi>j</mml:mi></mml:msub></mml:mrow><mml:mrow><mml:mo>−</mml:mo><mml:mn>1</mml:mn></mml:mrow></mml:msubsup><mml:mo stretchy="false">(</mml:mo><mml:mover accent="true"><mml:mi>S</mml:mi><mml:mo>¯</mml:mo></mml:mover><mml:mo stretchy="false">)</mml:mo></mml:mrow></mml:math></inline-formula>, where each <inline-formula><mml:math id="M8"><mml:mrow><mml:msub><mml:mover accent="true"><mml:mi>S</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mi>i</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> has the same set of corresponding nodes, obtained from the template mesh, <overline><italic>S</italic></overline>, and thus all segmentations from different subjects can be compared at a vertex-wise level. One benefit of this surface injection technique is that it can deal with many types of topological defects that can be present in the automated segmentation by enforcing a smoothness in the deformation that ignores holes and handles, as has been previously suggested (Khan et al., <xref rid="B16" ref-type="bibr">2008</xref>).</p><p>The vertex-wise correspondence in the meshes for each subcortical ROI across subjects enables the quantification of the disease related atrophy or hypertrophy on the surface of the ROI. The deformed surfaces may lie inside (atrophy) or outside (hypertrophy) of the prototype surface. To achieve this we used a signed closest point distance metric computed at each vertex on the prototype surface to that on the target surface. At each node, <italic>a</italic>∈ <inline-formula><mml:math id="M9"><mml:mrow><mml:msub><mml:mover accent="true"><mml:mi>S</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mi>i</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula>, <italic>m</italic>∈ <overline><italic>S</italic></overline>, we find the dot-product of the displacement from prototype to the deformed injected surface, <inline-formula><mml:math id="M10"><mml:mover accent="true"><mml:mi>d</mml:mi><mml:mo>→</mml:mo></mml:mover></mml:math></inline-formula><sub><italic>m, a</italic></sub> and the surface normal on the average, <inline-formula><mml:math id="M11"><mml:mover accent="true"><mml:mi>n</mml:mi><mml:mo>→</mml:mo></mml:mover></mml:math></inline-formula><sub><italic>m</italic></sub>. The normal distance, <italic>d<sup>norm</sup></italic>(<italic>m, a</italic>) = <inline-formula><mml:math id="M12"><mml:mover accent="true"><mml:mi>d</mml:mi><mml:mo>→</mml:mo></mml:mover></mml:math></inline-formula><sub><italic>m, a</italic></sub> · <inline-formula><mml:math id="M13"><mml:mover accent="true"><mml:mi>n</mml:mi><mml:mo>→</mml:mo></mml:mover></mml:math></inline-formula><sub><italic>m</italic></sub> is negative or positive when the deformed surface is respectively inward or outward relative to the average, effectively an indication of the deformation.</p></sec></sec><sec><title>2.3. Statistical analysis</title><p>In this section we describe the methods for the statistical analysis of the volume data and surface displacement features with the aim to find the group level differences in the data. Volume difference would suggest a global morphological alteration whereas surface displacement difference would suggest a localized shape change as a disease effect. The statistical tests for the surface displacement data were performed using both the rigidly and affinely pre-registered surfaces.</p><sec><title>2.3.1. Volume: group difference</title><p>The volume measurements from the surfaces generated from segmentation labels were tested for the statistical differences between the PD and HC groups. A two-tailed <italic>t</italic>-test was performed between the data from the two groups. The significance level was maintained at <italic>p</italic> < 0.05 for all tests. The test was repeated for both <italic>UBC</italic> and <italic>PPMI</italic> cohorts.</p></sec><sec><title>2.3.2. Surface displacement data: group difference</title><p>The surface displacement data as described in the Section 2.2.5 provided a signed distance value at each vertex on the prototype surface for individual structures. The surfaces for each target subject were in vertex-to-vertex correspondence enabling a direct comparison of the displacement data. A vertex-wise comparison of group difference across the cohort between the patient and healthy control groups provided insight into the spatially-localized, disease-related alteration in shape of the subcortical structures.</p><p>A linear model was fit to regress out the effect of variation in age and gender effects from the SD data. The SD data were kept as the outcome variable and the age and gender were used as predictors in the model. The residuals from the linear model fit were used in the statistical analysis to test for the effect of disease on the data. Vertex wise group difference analysis was conducted using SurfStat software (Worsley et al., <xref rid="B30" ref-type="bibr">1996</xref>), which employs Random Field Theory to correct for multiple comparisons (Worsley et al., <xref rid="B31" ref-type="bibr">2004</xref>). The vertex wise comparison and cluster-forming thresholds were set at <italic>p</italic> < 0.05. The contrast between the two groups was evaluated as <italic>HC</italic>—<italic>PD</italic>, where a positive <italic>t</italic>-value suggested atrophy in patients in comparison to the healthy controls and <italic>vice versa</italic>.</p></sec><sec><title>2.3.3. Relationship with clinical score</title><p>The surface displacement data were then tested for their potential to predict the UPDRS scores in both the studies.</p><p>The surface displacement data were rearranged into a 1-dimensional vector for each structure. These displacement vectors for all structures were then concatenated into a long column vector for each subject. In order to avoid the curse of dimensionality (very large dimensional data in a small sample size), dimensionality reduction was performed using Principal Component Analysis (PCA) decomposition. Sufficient number of PCs were retained to account for 95% of the variability in the data. The PC loadings were tested for the potential to predict the clinical scores in a linear model via linear regression. A leave-one-out procedure was conducted where each patients' score was predicted based on the model fit to the data from the remainder of the patient group. The PC loadings were the predictors and the clinical score were the response variables. A 2-D scatter plot (e.g., <bold>Figure 6</bold>) between the predicted vs. the actual score was obtained and a least squares line was fit to check for statistically significant relationship between the two. The coefficients of the linear model and their statistics are reported.</p><p>The UPDRS score in the <italic>UBC</italic> and <italic>PPMI</italic> cohorts was tested for association to the surface displacement data. We tested the UPDRS scores to be normally distributed via a lilleofors test (Lilliefors, <xref rid="B19" ref-type="bibr">1967</xref>). The data rejected the null hypothesis of “not normally distributed.”</p></sec></sec></sec><sec id="s3"><title>3. Results</title><p>The processed MRI data provided segmentation of subcortical structures for the caudate, putamen, thalamus, and pallidum in the left and right hemispheres (Figure <xref ref-type="fig" rid="F2">2</xref>). The segmentation labels for the anatomical structures were thoroughly checked for accuracy via visualization of surface outlines overlayed on the MRI slices (Figure <xref ref-type="fig" rid="F2">2</xref>). In our data, all subjects in the two cohorts were found to have acceptable (surface outline following structure boundary) and accurate segmentation labels, and were retained for subsequent analysis. Similarly, the surface displacement computed for each structure was visualized to check for presence of inaccuracies (extreme displacement values) due to registration errors. As an example, visualizations for three subjects in the patient and control groups are presented for the caudate and pallidum (right), and putamen and thalamus (left) structures (Figure <xref ref-type="fig" rid="F3">3</xref>).</p><fig id="F3" position="float"><label>Figure 3</label><caption><p><bold>Visualization for the quality control of the surface displacement data</bold>. Data for three representative subjects from the healthy control and patients groups, respectively have been presented from the <italic>UBC</italic> dataset for left thalamus and caudate and right pallidum and putamen structures. Color (legend) represents surface displacement from the prototype surface.</p></caption><graphic xlink:href="fnins-09-00101-g0003"></graphic></fig><p>The volumes of structures were significantly different (<italic>p</italic> < 0.05) for the right thalamus (<italic>p</italic> = 0.034, <italic>t</italic> = 2.15), right putamen (<italic>p</italic> = 0.041, <italic>t</italic> = 2.07), and left thalamus (<italic>p</italic> = 0.035, <italic>t</italic> = 2.14) in the <italic>UBC</italic> dataset. In contrast, volumes in the <italic>PPMI</italic> dataset did not present a statistically significant difference for all structures (Table <xref ref-type="table" rid="T3">3</xref>). In the rigid pre-registration case, the vertex-wise group analysis did not show any statistically significant difference. However, in the affine pre-registration case, the analysis of the surface displacement data found vertex clusters with significant difference in the displacements between patient and control groups (Figures <xref ref-type="fig" rid="F4">4</xref>, <xref ref-type="fig" rid="F5">5</xref>). As the rigid pre-registration failed to highlight the significant differences between the CN and PD groups, we present the results from only the affine case in the rest of this paper. Disease-related shape changes in subcortical structures were widespread. Table <xref ref-type="table" rid="T4">4</xref> presents the top 2 clusters ordered by the residuals of the model fit with their number of vertices, average <italic>t</italic>-value, and <italic>p</italic>-value for each structure.</p><table-wrap id="T3" position="float"><label>Table 3</label><caption><p><bold>Results for the analysis of group level difference in the volume of the subcortical structures between the participants in the Parkinson's disease and healthy control groups</bold>.</p></caption><table frame="hsides" rules="groups"><thead><tr><th rowspan="1" colspan="1"></th><th align="center" valign="top" colspan="2" rowspan="1"><bold>UBC</bold></th><th align="center" valign="top" colspan="2" rowspan="1"><bold>PPMI</bold></th></tr><tr><th align="left" valign="top" rowspan="1" colspan="1"><bold>Structure</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>p</italic>-value</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>t</italic>-stat</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>p</italic>-value</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>t</italic>-stat</bold></th></tr></thead><tbody><tr><td align="left" valign="top" rowspan="1" colspan="1">L-thal</td><td align="center" valign="top" rowspan="1" colspan="1">0.035<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.14</td><td align="center" valign="top" rowspan="1" colspan="1">0.406</td><td align="center" valign="top" rowspan="1" colspan="1">0.83</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">R-thal</td><td align="center" valign="top" rowspan="1" colspan="1">0.034<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.15</td><td align="center" valign="top" rowspan="1" colspan="1">0.180</td><td align="center" valign="top" rowspan="1" colspan="1">1.35</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">L-caud</td><td align="center" valign="top" rowspan="1" colspan="1">0.347</td><td align="center" valign="top" rowspan="1" colspan="1">−0.95</td><td align="center" valign="top" rowspan="1" colspan="1">0.262</td><td align="center" valign="top" rowspan="1" colspan="1">−1.12</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">R-caud</td><td align="center" valign="top" rowspan="1" colspan="1">0.092</td><td align="center" valign="top" rowspan="1" colspan="1">−1.70</td><td align="center" valign="top" rowspan="1" colspan="1">0.053</td><td align="center" valign="top" rowspan="1" colspan="1">−1.94</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">L-pall</td><td align="center" valign="top" rowspan="1" colspan="1">0.581</td><td align="center" valign="top" rowspan="1" colspan="1">−0.55</td><td align="center" valign="top" rowspan="1" colspan="1">0.227</td><td align="center" valign="top" rowspan="1" colspan="1">−1.21</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">R-pall</td><td align="center" valign="top" rowspan="1" colspan="1">0.606</td><td align="center" valign="top" rowspan="1" colspan="1">0.52</td><td align="center" valign="top" rowspan="1" colspan="1">0.945</td><td align="center" valign="top" rowspan="1" colspan="1">0.07</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">L-put</td><td align="center" valign="top" rowspan="1" colspan="1">0.126</td><td align="center" valign="top" rowspan="1" colspan="1">1.54</td><td align="center" valign="top" rowspan="1" colspan="1">0.959</td><td align="center" valign="top" rowspan="1" colspan="1">−0.05</td></tr><tr><td align="left" valign="top" rowspan="1" colspan="1">R-put</td><td align="center" valign="top" rowspan="1" colspan="1">0.041<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.07</td><td align="center" valign="top" rowspan="1" colspan="1">0.400</td><td align="center" valign="top" rowspan="1" colspan="1">0.84</td></tr></tbody></table><table-wrap-foot><p><italic>Statistical significance at p < 0.05 marked by asterisk (<sup>*</sup>)</italic>.</p></table-wrap-foot></table-wrap><fig id="F4" position="float"><label>Figure 4</label><caption><p><bold>Results for the vertex-wise group difference analysis of the surface displacement data for the left and right, caudate and thalamus structures in the <italic>UBC</italic> and <italic>PPMI</italic> datasets</bold>. The colored patches represent the <italic>t</italic>-values in the areas with statistically significant (<italic>p</italic> < 0.05) difference between the patient and healthy control group. Gray colored area had no statistically significant difference between the groups. Positive <italic>t</italic>-value represents lower value in the patient group.</p></caption><graphic xlink:href="fnins-09-00101-g0004"></graphic></fig><fig id="F5" position="float"><label>Figure 5</label><caption><p><bold>Results for the vertex-wise group difference analysis of the surface displacement data for left and right, pallidum and putamen structures in the <italic>UBC</italic> and <italic>PPMI</italic> datasets</bold>. The colored patches represent the <italic>t</italic>-values in the areas with statistically significant (<italic>p</italic> < 0.05) difference between the patient and healthy control group. Gray colored area had no statistically significant difference between the groups. Positive <italic>t</italic>-value represents lower value in the patient group.</p></caption><graphic xlink:href="fnins-09-00101-g0005"></graphic></fig><table-wrap id="T4" position="float"><label>Table 4</label><caption><p><bold>Results for the vertex-wise group difference in the surface displacement (affine pre-registered) between Parkinson's disease patients [<italic>n</italic> = 55 (<italic>UBC</italic>), <italic>n</italic> = 189 (<italic>PPMI</italic>)] and healthy controls [<italic>n</italic> = 54 (<italic>UBC</italic>), <italic>n</italic> = 137 (<italic>PPMI</italic>)] for 8 subcortical structures</bold>.</p></caption><table frame="hsides" rules="groups"><thead><tr><th align="left" valign="top" rowspan="1" colspan="1"><bold>Structure</bold><break></break><bold>(Total no. of vertices)</bold></th><th align="center" valign="top" colspan="5" rowspan="1"><bold>UBC PD</bold></th><th align="center" valign="top" colspan="5" rowspan="1"><bold>PPMI</bold></th></tr><tr><th rowspan="1" colspan="1"></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>No. of vertices</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>p</italic>-value</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>t</italic>-value</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>HC group</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>PD group</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>No. of vertices</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>p</italic>-value</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold><italic>t</italic>-value</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>HC group</bold></th><th align="center" valign="top" rowspan="1" colspan="1"><bold>PD group</bold></th></tr></thead><tbody><tr><td align="left" colspan="11" valign="top" rowspan="1">L-thal (2212)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">145</td><td align="center" valign="top" rowspan="1" colspan="1">0.001<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.73</td><td align="center" valign="top" rowspan="1" colspan="1">0.15±0.05</td><td align="center" valign="top" rowspan="1" colspan="1">0.04±0.06</td><td align="center" valign="top" rowspan="1" colspan="1">159</td><td align="center" valign="top" rowspan="1" colspan="1">0.000<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.21</td><td align="center" valign="top" rowspan="1" colspan="1">0.18±0.04</td><td align="center" valign="top" rowspan="1" colspan="1">0.11±0.04</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">111</td><td align="center" valign="top" rowspan="1" colspan="1">0.008<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.06</td><td align="center" valign="top" rowspan="1" colspan="1">0.09±0.05</td><td align="center" valign="top" rowspan="1" colspan="1">−0.08±0.05</td><td align="center" valign="top" rowspan="1" colspan="1">152</td><td align="center" valign="top" rowspan="1" colspan="1">0.001<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.21</td><td align="center" valign="top" rowspan="1" colspan="1">0.06±0.05</td><td align="center" valign="top" rowspan="1" colspan="1">−0.02±0.04</td></tr><tr><td align="left" colspan="11" valign="top" rowspan="1">R-thal (2146)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">244</td><td align="center" valign="top" rowspan="1" colspan="1">0.003<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.32</td><td align="center" valign="top" rowspan="1" colspan="1">0.57±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">0.46±0.04</td><td align="center" valign="top" rowspan="1" colspan="1">185</td><td align="center" valign="top" rowspan="1" colspan="1">0.000<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.27</td><td align="center" valign="top" rowspan="1" colspan="1">0.48±0.05</td><td align="center" valign="top" rowspan="1" colspan="1">0.43±0.04</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">68</td><td align="center" valign="top" rowspan="1" colspan="1">0.132</td><td align="center" valign="top" rowspan="1" colspan="1">−2.53</td><td align="center" valign="top" rowspan="1" colspan="1">0.11±0.06</td><td align="center" valign="top" rowspan="1" colspan="1">0.28±0.06</td><td align="center" valign="top" rowspan="1" colspan="1">166</td><td align="center" valign="top" rowspan="1" colspan="1">0.003<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.47</td><td align="center" valign="top" rowspan="1" colspan="1">0.31±0.04</td><td align="center" valign="top" rowspan="1" colspan="1">0.23±0.04</td></tr><tr><td align="left" colspan="11" valign="top" rowspan="1">L-caud (2096)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">62</td><td align="center" valign="top" rowspan="1" colspan="1">0.023<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.49</td><td align="center" valign="top" rowspan="1" colspan="1">0.13±0.04</td><td align="center" valign="top" rowspan="1" colspan="1">0.02±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">88</td><td align="center" valign="top" rowspan="1" colspan="1">0.008<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.08</td><td align="center" valign="top" rowspan="1" colspan="1">0.02±0.06</td><td align="center" valign="top" rowspan="1" colspan="1">−0.07±0.05</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">37</td><td align="center" valign="top" rowspan="1" colspan="1">0.102<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">−2.69</td><td align="center" valign="top" rowspan="1" colspan="1">−0.04±0.01</td><td align="center" valign="top" rowspan="1" colspan="1">0.03±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">110</td><td align="center" valign="top" rowspan="1" colspan="1">0.019<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">−2.80</td><td align="center" valign="top" rowspan="1" colspan="1">0.15±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">0.23±0.03</td></tr><tr><td align="left" colspan="11" valign="top" rowspan="1">R-caud (2116)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">170</td><td align="center" valign="top" rowspan="1" colspan="1">0.003<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.84</td><td align="center" valign="top" rowspan="1" colspan="1">0.09±0.08</td><td align="center" valign="top" rowspan="1" colspan="1">−0.11±0.06</td><td align="center" valign="top" rowspan="1" colspan="1">306</td><td align="center" valign="top" rowspan="1" colspan="1">0.000<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.75</td><td align="center" valign="top" rowspan="1" colspan="1">0.11±0.05</td><td align="center" valign="top" rowspan="1" colspan="1">0.04±0.06</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">176</td><td align="center" valign="top" rowspan="1" colspan="1">0.004<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">−2.69</td><td align="center" valign="top" rowspan="1" colspan="1">0.17±0.04</td><td align="center" valign="top" rowspan="1" colspan="1">0.32±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">152</td><td align="center" valign="top" rowspan="1" colspan="1">0.004<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.88</td><td align="center" valign="top" rowspan="1" colspan="1">−0.12±0.07</td><td align="center" valign="top" rowspan="1" colspan="1">−0.22±0.07</td></tr><tr><td align="left" colspan="11" valign="top" rowspan="1">L-pall (522)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">115</td><td align="center" valign="top" rowspan="1" colspan="1">0.001<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.70</td><td align="center" valign="top" rowspan="1" colspan="1">0.60±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">0.55±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">270</td><td align="center" valign="top" rowspan="1" colspan="1">0.000<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.01</td><td align="center" valign="top" rowspan="1" colspan="1">0.44±0.04</td><td align="center" valign="top" rowspan="1" colspan="1">0.42±0.04</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">14</td><td align="center" valign="top" rowspan="1" colspan="1">0.086</td><td align="center" valign="top" rowspan="1" colspan="1">2.58</td><td align="center" valign="top" rowspan="1" colspan="1">0.33±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">0.28±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">4</td><td align="center" valign="top" rowspan="1" colspan="1">0.555</td><td align="center" valign="top" rowspan="1" colspan="1">−2.13</td><td align="center" valign="top" rowspan="1" colspan="1">0.19±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">0.23±0.00</td></tr><tr><td align="left" colspan="11" valign="top" rowspan="1">R-pall (468)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">63</td><td align="center" valign="top" rowspan="1" colspan="1">0.001<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.59</td><td align="center" valign="top" rowspan="1" colspan="1">0.33±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">0.31±0.01</td><td align="center" valign="top" rowspan="1" colspan="1">147</td><td align="center" valign="top" rowspan="1" colspan="1">0.000<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.02</td><td align="center" valign="top" rowspan="1" colspan="1">0.50±0.04</td><td align="center" valign="top" rowspan="1" colspan="1">0.48±0.05</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">95</td><td align="center" valign="top" rowspan="1" colspan="1">0.001<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.92</td><td align="center" valign="top" rowspan="1" colspan="1">0.44±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">0.41±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">29</td><td align="center" valign="top" rowspan="1" colspan="1">0.011<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.08</td><td align="center" valign="top" rowspan="1" colspan="1">0.30±0.01</td><td align="center" valign="top" rowspan="1" colspan="1">0.28±0.01</td></tr><tr><td align="left" colspan="11" valign="top" rowspan="1">L-put (1996)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">92</td><td align="center" valign="top" rowspan="1" colspan="1">0.002<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">−2.94</td><td align="center" valign="top" rowspan="1" colspan="1">−0.09±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">−0.03±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">157</td><td align="center" valign="top" rowspan="1" colspan="1">0.000<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">3.19</td><td align="center" valign="top" rowspan="1" colspan="1">0.28±0.07</td><td align="center" valign="top" rowspan="1" colspan="1">0.21±0.06</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">51</td><td align="center" valign="top" rowspan="1" colspan="1">0.097</td><td align="center" valign="top" rowspan="1" colspan="1">−3.00</td><td align="center" valign="top" rowspan="1" colspan="1">−0.07±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">0.06±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">67</td><td align="center" valign="top" rowspan="1" colspan="1">0.032</td><td align="center" valign="top" rowspan="1" colspan="1">2.51</td><td align="center" valign="top" rowspan="1" colspan="1">0.10±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">0.05±0.03</td></tr><tr><td align="left" colspan="11" valign="top" rowspan="1">R-put (1974)</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">75</td><td align="center" valign="top" rowspan="1" colspan="1">0.008<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.48</td><td align="center" valign="top" rowspan="1" colspan="1">0.15±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">0.08±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">73</td><td align="center" valign="top" rowspan="1" colspan="1">0.002<sup>*</sup></td><td align="center" valign="top" rowspan="1" colspan="1">2.58</td><td align="center" valign="top" rowspan="1" colspan="1">0.30±0.07</td><td align="center" valign="top" rowspan="1" colspan="1">0.26±0.07</td></tr><tr><td rowspan="1" colspan="1"></td><td align="left" valign="top" rowspan="1" colspan="1">55</td><td align="center" valign="top" rowspan="1" colspan="1">0.103</td><td align="center" valign="top" rowspan="1" colspan="1">−2.65</td><td align="center" valign="top" rowspan="1" colspan="1">0.07±0.02</td><td align="center" valign="top" rowspan="1" colspan="1">0.13±0.03</td><td align="center" valign="top" rowspan="1" colspan="1">90</td><td align="center" valign="top" rowspan="1" colspan="1">0.012</td><td align="center" valign="top" rowspan="1" colspan="1">2.45</td><td align="center" valign="top" rowspan="1" colspan="1">0.11±0.05</td><td align="center" valign="top" rowspan="1" colspan="1">0.07±0.04</td></tr></tbody></table><table-wrap-foot><p><italic>Data presented for top 2 vertex clusters and their corresponding mean value for the t-statistic, p-value and the mean ± standard-deviation of the surface displacement in the HC and PD group. Structures and clusters with statistically significant (p < 0.05) difference marked with an asterisk (<sup>*</sup>)</italic>.</p></table-wrap-foot></table-wrap><p>Spatial clusters with statistically significant difference in the surface displacement feature in the two groups were found on all the structures. Two key observations appear from the statistical comparisons: (1) Structures with net-inward deformation for the PD group (+ve <italic>t</italic>-value, right putamen, right, and left pallidum) and (2) structures with both net-inward and outward deformation for the PD group (−ve and +ve <italic>t</italic>-value, left thalamus, right, and left caudate). Additionally, similar spatial locations with difference of same (+ve or −ve <italic>t</italic>-value) nature were observed in the two datasets (e.g., right and left pallidum). The clinical score prediction experiment with the data from all structures used simultaneously as a single column vector was not able to predict the UPDRS scores with the statistical significance (<italic>p</italic> > 0.5).</p></sec><sec id="s4"><title>4. Discussion</title><p>Analysis of local shape change in comparison to global measures of morphology highlights the importance of examining spatially-localized alterations as a disease-related effect. In an extrinsic approach, the features extracted from the deformation fields rely on the choice of template to which individual images/surfaces are registered. These features can then be compared between groups for effect of a disease or clinical intervention. We computed the surface displacement metric as the signed closest point distance at each vertex of the surface model to quantify the net surface deformation. Surface displacement data (affine pre-registered) showed very strong differences between the two groups for all 4 structures (Table <xref ref-type="table" rid="T4">4</xref>, Figures <xref ref-type="fig" rid="F4">4</xref>, <xref ref-type="fig" rid="F5">5</xref>). Such differences were observed in <italic>UBC</italic> and the <italic>PPMI</italic> datasets, where patches of vertex-clusters were present throughout the surface (Figures <xref ref-type="fig" rid="F4">4</xref>, <xref ref-type="fig" rid="F5">5</xref>). Both atrophy (positive <italic>t</italic>-value) and hypertrophy (negative <italic>t</italic>-value) were present in the patient group, suggesting compensatory alteration in shape within the same structure.</p><p>It is interesting to note that the volume of the subcortical structures did not show statistically significant difference for any of the structures in the <italic>PPMI</italic> dataset (Table <xref ref-type="table" rid="T3">3</xref>). In contrast, many clusters with statistically significant (<italic>p</italic> < 0.05) difference in the surface displacement metric were observed in the vertex-wise group difference analysis (Table <xref ref-type="table" rid="T4">4</xref>, Figures <xref ref-type="fig" rid="F4">4</xref>, <xref ref-type="fig" rid="F5">5</xref>). Surface displacement being a local, sensitive measure of shape change is able to present group level differences. In contrast, the gross measurement of the structural volume is shown to not yield sensitivity to disease changes. This finding is consistent with previous observations (McKeown et al., <xref rid="B20" ref-type="bibr">2008</xref>) where volume did not show any difference between the disease and control group, but a shape feature was able to show statistically significant results.</p><sec><title>4.1. Technical strengths of the proposed analysis approach</title><p>We applied an automated registration based segmentation approach (FS + LDDMM) which has been validated to provide accurate segmentation labels (Section 2.2.2) (Khan et al., <xref rid="B16" ref-type="bibr">2008</xref>). The application of the automated method with multiple templates is expected to yield accurate segmentation of the subcortical ROI, unaffected from potential variability due to a manual labeling procedure. This segmentation method has already been shown to provide high quality segmentation for segmentation of caudate, thalamus, putamen, and hippocampus (average dice coefficient = 0.85). Another validation study in the context of segmentation of pediatric MRI images found the dice coefficients of 0.89 for thalamus, 0.89 for caudate and 0.87 for putamen, thereby emphasizing accurate segmentation results (Garg et al., <xref rid="B12" ref-type="bibr">2014</xref>).</p><p>The choice of templates for segmentation is known to impact segmentation quality (Garg et al., <xref rid="B12" ref-type="bibr">2014</xref>). In our study, we used templates (<italic>n</italic> = 6) from the <italic>UBC</italic> cohort which were manually segmented by a neuro-anatomy expert. The templates do differ from the data in the <italic>PPMI</italic> cohort, which, in itself, being a collection of data acquired at different sites, has inherent inhomogeneities due to scanner characteristics and acquisition protocols. In order to ensure good quality segmentation of the data in the two cohorts we performed thorough segmentation quality check as explained in Section 2.2.3. All the subjects in the two cohorts were found to have accurate segmentation labels, and hence were retained for subsequent analysis. This prevented the errors in segmentation from propagating into subsequent processing and analysis steps. Additionally, the use of an unbiased prototype (Sections 2.2.4, 2.2.5) to obtain the surface displacement metric ensures that the deformation data is free from bias toward a subset of the group. Therefore, the statistical outcomes obtained in our study represent the characteristic differences in the anatomy captured by the MRI data.</p></sec><sec><title>4.2. Strength of the shape feature</title><p>Structural MRI is a potential marker for alterations observed in PD as it can assess brain systems associated with motor and non-motor deficits. It has been shown that less sensitive measures of change in morphology such as volume presents contradictory results for putamen (e.g., Krabbe et al., <xref rid="B17" ref-type="bibr">2005</xref>; Pitcher et al., <xref rid="B24" ref-type="bibr">2012</xref>) vs. (Schulz et al., <xref rid="B27" ref-type="bibr">1999</xref>), Ghaemi et al. (<xref rid="B13" ref-type="bibr">2002</xref>) and caudate atrophy (e.g., Pitcher et al., <xref rid="B24" ref-type="bibr">2012</xref> vs. Schulz et al., <xref rid="B27" ref-type="bibr">1999</xref>; Ghaemi et al., <xref rid="B13" ref-type="bibr">2002</xref>). Additionally, thalami in PD showed significant shape change despite no significant difference in volume (McKeown et al., <xref rid="B20" ref-type="bibr">2008</xref>). Results from our study are in alignment with the previous observations where volumes of anatomical structures did not show a statistically significant change, whereas widespread shape change was observed as an effect of Parkinson's disease. Our shape analysis approach and its derived features contain both global and local shape information and show ability to capture sensitive disease related shape change in two large and independent datasets.</p><p>Previous work (McKeown et al., <xref rid="B20" ref-type="bibr">2008</xref>) exploring differences in thalamic shape utilized a SPHARM representation. A recent morphometric analysis (Apostolova et al., <xref rid="B1" ref-type="bibr">2010</xref>) also detected changes in basal ganglia structures between PD subjects with and without dementia. Apostolova's method focused on volume, where the radial distance—an intuitive measure of ROI thickness—was used as morphometric feature. Similarly, other work (Sterling et al., <xref rid="B28" ref-type="bibr">2013</xref>) also modeled shape with a SPHARM-PDM representation and found areas with significant differences in the caudate and putamen structures. In contrast, our work quantified the shape change with a surface displacement measure at every vertex in the average surface model for each sucortical region. The high dimensional non-rigid registration algorithm (LDDMM, Beg et al., <xref rid="B2" ref-type="bibr">2005</xref>) used in this study has been shown to account for the non-linearity of the anatomical shape space. The metric thus derived is able to accurately capture the shape alteration due to disease. Using this metric, we were able to spatially localize the areas with significant disease related alteration.</p><p>We found both similarities as well as differences in the anatomical location of shape changes between the two cohorts. The observed differences can be attributed to the difference in scanners, image acquisition protocols, inhomogeneities due to multi-center data collection in the PPMI cohort and differences in clinical features. Nonetheless, there are very encouraging overlapping changes in the PD groups in the two cohorts, mainly in sensorimotor and limbic areas and we will preliminarily discuss the functional implications of these most robust changes: The globus pallidus shows the most widespread changes with several areas of atrophy bilaterally in both cohorts; in the two datasets, areas of atrophy overlap along the medial aspect of the globus pallidus internus, most likely in areas implied in motor function (Obeso et al., <xref rid="B22" ref-type="bibr">2015</xref>). In the caudate, atrophy is present along the dorso(medial) tail of the caudate on the right, affecting areas belonging to sensorimotor, frontostriatal circuits (Redgrave et al., <xref rid="B26" ref-type="bibr">2010</xref>). On the left, fewer changes are observed, there is some overlap for atrophy along the ventral tail of the caudate, again likely implying sensorimotor circuits. In the left putamen, we found mostly hypertrophy in the UBC cohort but atrophy in the PPMI cohort without clear areas of overlap between the groups. On the right, areas of atrophy overlap in the dorsal area of the posterior putamen, which is associated with sensorimotor function (Redgrave et al., <xref rid="B26" ref-type="bibr">2010</xref>). Both areas of hypertrophy and atrophy were observed in the thalamus, atrophy with partial overlap between the two groups is found in mediodorsal aspects of the thalami bilaterally. The mediodorsal thalamus is part of the limbic circuit and has been implied in depression in PD (Cardoso et al., <xref rid="B6" ref-type="bibr">2009</xref>; Li et al., <xref rid="B18" ref-type="bibr">2010</xref>). The shape feature computed in our study presents the potential to detect the differences between PD and control groups. The discriminant function that can be developed from this work could then be applied in longitudinal studies to see if it is able to detect alterations in subjects in the prodromal phase of PD.</p></sec><sec><title>4.3. Shape differences and clinical association</title><p>The surface displacement shape feature was not able to predict the UPDRS motor score (Figure <xref ref-type="fig" rid="F6">6</xref>). The UPDRS motor score is a fairly crude clinical measurement as it combines different features of PD such as tremor and postural instability which have differing anatomical bases and this might partially explain why we did not find a correlation of the UPDRS motor score with shape changes. In order to identify correlations of shape changes with clinical indices, future studies should examine motor subscores for rigidity, bradykinesia, tremor and axial stability separately, assess mood, cognitive and reward function, control for more affected side and handedness and match groups closely for age and gender.</p><fig id="F6" position="float"><label>Figure 6</label><caption><p><bold>Scatter plots for the predicted and actual clinical scores from the leave one out prediction experiment using the PC loadings from the PCA decomposition of surface displacements where data from all structures was combined</bold>. Experimental data from the <bold>(A)</bold><italic>UBC</italic> dataset and <bold>(B)</bold><italic>PPMI</italic> dataset. Blue line is the least squares fit between the predicted and actual clinical score values.</p></caption><graphic xlink:href="fnins-09-00101-g0006"></graphic></fig></sec><sec><title>4.4. Limitation and conclusions</title><p>In our study, we considered the pre-registration of the binary segmentations using rigid and affine transformations, respectively (see Section 2.2.5). The surface displacement feature computed from the rigid registered data did not show a statistically significant difference between the groups (HC vs. PD), whereas, the affine registered data showed spatial clusters with significant difference (Table <xref ref-type="table" rid="T4">4</xref>, Figures <xref ref-type="fig" rid="F4">4</xref>, <xref ref-type="fig" rid="F5">5</xref>). This suggests that the scale variability among the target binary segmentations has a confounding effect on the CN vs. PD group differences, and the diseases related changes are likely smaller than the variability in the scale of the structures seen across the population. The use of an affine pre-registration step between the binary segmentations removed the scale-related variability, however, in the process, it may also have removed some of the scale related changes seen due to PD. Hence, using the structures which are changing due to the disease, to further remove effects due to overall scale change across the population is likely sub-optimal. Further, the cross-talk between the normally occurring scale variability of structures in the population vs. the changes in scale due to disease (such as atrophy, or hypertrophy) likely affects all shape analysis methods that rely on surface deformation based morphometry. This observation highlights the need for shape assessment methods to deal with scale-related variability in the population better. As a suggestion, perhaps accounting for scale variability using a feature that does not change with disease such as the cranial vault instead of the binary segmentations can be investigated.</p><p>We applied volumetric registration method to obtain the mappings between the source and target binary images. As the binary images carry shape information at the boundary pixels and the pixels in the interior contain minimal or no information regarding shape of the structures, such methods are expected to provide registration accuracy and shape information, at par with surface registration methods. However, a direct comparison of the two sets of methods is beyond the focus of the current work and forms a scope for future investigation. Additionally, in order to be used as a biomarker in the clinical setting, the method needs to be tested further on much larger and independent data cohorts for its ability to predict clinical features and the ability to detect changes in the prodromal stages of the disease.</p><p>In conclusion, we present a first study to investigate the change in shape in Parkinson's disease tested on a large publicly available dataset (<italic>PPMI</italic>) and validated on an independently acquired dataset at <italic>UBC</italic>. Our results suggest that systemic changes in the shape of subcortical structures (caudate, pallidum, putamen and thalamus) can be non-invasively assessed in PD <italic>in vivo</italic>. The surface displacement feature encodes the spatially localized shape information. In this study, we have been able to highlight regions on the surface of subcortical structures that show changes in PD. The automated method presented in the study provides a new avenue to assess the progress of the neuro-degenerative processes. Further validation using data from larger cohorts is needed to assess the predictive capability of this method.</p></sec><sec><title>Conflict of interest statement</title><p>The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.</p></sec></sec></body><back><fn-group><fn fn-type="financial-disclosure"><p><bold>Funding</bold>. The authors acknowledge and thank the <funding-source id="GS1">Natural Sciences and Engineering Research Council (NSERC)</funding-source>, <funding-source id="GS2">Michael Smith Foundation for Health Research (MSFHR)</funding-source>, <funding-source id="GS3">Canadian Institute of Health Research (CIHR)</funding-source>, and MITACS Canada for their generous funding support. Part of this work was funded by a grant from the <funding-source id="GS5">National Parkinson's Foundation (NPF) to MJM and SC</funding-source>, as well as the <funding-source id="GS6">PPRI/UBC Chair</funding-source> in Parkinson's research (MJM).</p></fn></fn-group><ack><p>Data used in the preparation of this article were obtained from the Parkinsons Progression Markers Initiative (PPMI) database (<ext-link ext-link-type="uri" xlink:href="http://www.ppmi-info.org/data">www.ppmi-info.org/data</ext-link>). For up-to-date information on the study, visit <ext-link ext-link-type="uri" xlink:href="http://www.ppmi-info.org">www.ppmi-info.org</ext-link>. PPMI, a public-private partnership, is funded by the Michael J. Fox Foundation for Parkinsons Research and funding partners, including Abbott, Avid Radiopharmaceuticals, Biogen Idec, Bristol-Myers Squibb, Covance, lan, GE Healthcare, Genentech, GSK-GlaxoSmithKline, Lilly, Merck, MSD-Meso Scale Discovery, Pfizer, Roche, and UCB (<ext-link ext-link-type="uri" xlink:href="http://www.ppmi-info.org/fundingpartners">www.ppmi-info.org/fundingpartners</ext-link>).</p></ack><ref-list><title>References</title><ref id="B1"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Apostolova</surname><given-names>L. G.</given-names></name><name><surname>Beyer</surname><given-names>M.</given-names></name><name><surname>Green</surname><given-names>A. E.</given-names></name><name><surname>Hwang</surname><given-names>K. S.</given-names></name><name><surname>Morra</surname><given-names>J. H.</given-names></name><name><surname>Chou</surname><given-names>Y.-Y.</given-names></name><etal></etal></person-group>. (<year>2010</year>). <article-title>Hippocampal, caudate, and ventricular changes in Parkinson's disease with and without dementia</article-title>. <source>Mov. Disord</source>. <volume>25</volume>, <fpage>687</fpage>–<lpage>695</lpage>. <pub-id pub-id-type="doi">10.1002/mds.22799</pub-id><pub-id pub-id-type="pmid">20437538</pub-id></mixed-citation></ref><ref id="B2"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Beg</surname><given-names>M.</given-names></name><name><surname>Miller</surname><given-names>M.</given-names></name><name><surname>Trouvé</surname><given-names>A.</given-names></name><name><surname>Younes</surname><given-names>L.</given-names></name></person-group> (<year>2005</year>). <article-title>Computing large deformation metric mappings via geodesic flows of diffeomorphisms</article-title>. <source>Int. J. Comput. Vis</source>. <volume>61</volume>, <fpage>139</fpage>–<lpage>157</lpage>. <pub-id pub-id-type="doi">10.1023/B:VISI.0000043755.93987.aa</pub-id><pub-id pub-id-type="pmid">19698793</pub-id></mixed-citation></ref><ref id="B3"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bergouignan</surname><given-names>L.</given-names></name><name><surname>Chupin</surname><given-names>M.</given-names></name><name><surname>Czechowska</surname><given-names>Y.</given-names></name><name><surname>Kinkingnéhun</surname><given-names>S.</given-names></name><name><surname>Lemogne</surname><given-names>C.</given-names></name><name><surname>Le Bastard</surname><given-names>G.</given-names></name><etal></etal></person-group>. (<year>2009</year>). <article-title>Can voxel based morphometry, manual segmentation and automated segmentation equally detect hippocampal volume differences in acute depression?</article-title><source>Neuroimage</source><volume>45</volume>, <fpage>29</fpage>–<lpage>37</lpage>. <pub-id pub-id-type="doi">10.1016/j.neuroimage.2008.11.006</pub-id><pub-id pub-id-type="pmid">19071222</pub-id></mixed-citation></ref><ref id="B4"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Brenneis</surname><given-names>C.</given-names></name><name><surname>Egger</surname><given-names>K.</given-names></name><name><surname>Scherfler</surname><given-names>C.</given-names></name><name><surname>Seppi</surname><given-names>K.</given-names></name><name><surname>Schocke</surname><given-names>M.</given-names></name><name><surname>Poewe</surname><given-names>W.</given-names></name><etal></etal></person-group>. (<year>2007</year>). <article-title>Progression of brain atrophy in multiple system atrophy. A longitudinal VBM study</article-title>. <source>J. Neurol</source>. <volume>254</volume>, <fpage>191</fpage>–<lpage>196</lpage>. <pub-id pub-id-type="doi">10.1007/s00415-006-0325-6</pub-id><pub-id pub-id-type="pmid">17334661</pub-id></mixed-citation></ref><ref id="B5"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Burton</surname><given-names>E. J.</given-names></name><name><surname>McKeith</surname><given-names>I. G.</given-names></name><name><surname>Burn</surname><given-names>D. J.</given-names></name><name><surname>O'Brien</surname><given-names>J. T.</given-names></name></person-group> (<year>2005</year>). <article-title>Brain atrophy rates in Parkinson's disease with and without dementia using serial magnetic resonance imaging</article-title>. <source>Mov. Disord</source>. <volume>20</volume>, <fpage>1571</fpage>–<lpage>1576</lpage>. <pub-id pub-id-type="doi">10.1002/mds.20652</pub-id><pub-id pub-id-type="pmid">16116613</pub-id></mixed-citation></ref><ref id="B6"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cardoso</surname><given-names>E. F.</given-names></name><name><surname>Maia</surname><given-names>F. M.</given-names></name><name><surname>Fregni</surname><given-names>F.</given-names></name><name><surname>Myczkowski</surname><given-names>M. L.</given-names></name><name><surname>Melo</surname><given-names>L. M.</given-names></name><name><surname>Sato</surname><given-names>J. A. R.</given-names></name><etal></etal></person-group>. (<year>2009</year>). <article-title>Depression in Parkinson's disease: convergence from voxel-based morphometry and functional magnetic resonance imaging in the limbic thalamus</article-title>. <source>Neuroimage</source><volume>47</volume>, <fpage>467</fpage>–<lpage>472</lpage>. <pub-id pub-id-type="doi">10.1016/j.neuroimage.2009.04.059</pub-id><pub-id pub-id-type="pmid">19398020</pub-id></mixed-citation></ref><ref id="B7"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Davatzikos</surname><given-names>C.</given-names></name></person-group> (<year>2004</year>). <article-title>Why voxel-based morphometric analysis should be used with great caution when characterizing group differences</article-title>. <source>Neuroimage</source><volume>23</volume>, <fpage>17</fpage>–<lpage>20</lpage>. <pub-id pub-id-type="doi">10.1016/j.neuroimage.2004.05.010</pub-id><pub-id pub-id-type="pmid">15325347</pub-id></mixed-citation></ref><ref id="B8"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>de Lau</surname><given-names>L.</given-names></name><name><surname>Breteler</surname><given-names>M.</given-names></name></person-group> (<year>2006</year>). <article-title>Epidemiology of Parkinson's disease</article-title>. <source>Lancet Neurol</source>. <volume>13</volume><issue>Suppl. 1</issue>, <fpage>2</fpage>–<lpage>9</lpage><pub-id pub-id-type="doi">10.1016/S1474-4422(06)70471-9</pub-id></mixed-citation></ref><ref id="B9"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Duchesne</surname><given-names>S.</given-names></name><name><surname>Rolland</surname><given-names>Y.</given-names></name><name><surname>Vérin</surname><given-names>M.</given-names></name></person-group> (<year>2009</year>). <article-title>Automated computer differential classification in Parkinsonian Syndromes via pattern analysis on MRI</article-title>. <source>Acad. Radiol</source>. <volume>16</volume>, <fpage>61</fpage>–<lpage>70</lpage>. <pub-id pub-id-type="doi">10.1016/j.acra.2008.05.024</pub-id><pub-id pub-id-type="pmid">19064213</pub-id></mixed-citation></ref><ref id="B10"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Fischl</surname><given-names>B.</given-names></name><name><surname>Kouwe</surname><given-names>A. V. D.</given-names></name></person-group> (<year>2004</year>). <article-title>Automatically parcellating the human cerebral cortex</article-title>. <source>Cereb. Cortex</source><volume>14</volume>, <fpage>11</fpage>–<lpage>22</lpage>. <pub-id pub-id-type="doi">10.1093/cercor/bhg087</pub-id><pub-id pub-id-type="pmid">14654453</pub-id></mixed-citation></ref><ref id="B11"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Fischl</surname><given-names>B.</given-names></name><name><surname>Salat</surname><given-names>D. H.</given-names></name><name><surname>Busa</surname><given-names>E.</given-names></name><name><surname>Albert</surname><given-names>M.</given-names></name><name><surname>Dieterich</surname><given-names>M.</given-names></name><name><surname>Haselgrove</surname><given-names>C.</given-names></name><etal></etal></person-group>. (<year>2002</year>). <article-title>Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain</article-title>. <source>Neuron</source><volume>33</volume>, <fpage>341</fpage>–<lpage>355</lpage>. <pub-id pub-id-type="doi">10.1016/S0896-6273(02)00569-X</pub-id><pub-id pub-id-type="pmid">11832223</pub-id></mixed-citation></ref><ref id="B12"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Garg</surname><given-names>A.</given-names></name><name><surname>Wong</surname><given-names>D.</given-names></name><name><surname>Popuri</surname><given-names>K.</given-names></name><name><surname>Poskitt</surname><given-names>K. J.</given-names></name><name><surname>Fitzpatrick</surname><given-names>K.</given-names></name><name><surname>Bjornson</surname><given-names>B.</given-names></name><etal></etal></person-group> (<year>2014</year>). <article-title>Manually segmented template library for 8-year-old pediatric brain MRI data with 16 subcortical structures</article-title>. <source>J. Med. Imaging</source><volume>1</volume>:<fpage>034502</fpage><pub-id pub-id-type="doi">10.1117/1.JMI.1.3.034502</pub-id></mixed-citation></ref><ref id="B13"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ghaemi</surname><given-names>M.</given-names></name><name><surname>Hilker</surname><given-names>R.</given-names></name><name><surname>Rudolf</surname><given-names>J.</given-names></name></person-group> (<year>2002</year>). <article-title>Differentiating multiple system atrophy from Parkinson's disease: contribution of striatal and midbrain MRI volumetry and multi-tracer PET imaging</article-title>. <source>J. Neurol. Neurosurg. Psychiatry</source><volume>73</volume>, <fpage>517</fpage>–<lpage>523</lpage>. <pub-id pub-id-type="doi">10.1136/jnnp.73.5.517</pub-id><pub-id pub-id-type="pmid">12397143</pub-id></mixed-citation></ref><ref id="B14"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hammers</surname><given-names>A.</given-names></name><name><surname>Allom</surname><given-names>R.</given-names></name><name><surname>Koepp</surname><given-names>M. J.</given-names></name><name><surname>Free</surname><given-names>S. L.</given-names></name><name><surname>Myers</surname><given-names>R.</given-names></name><name><surname>Lemieux</surname><given-names>L.</given-names></name><etal></etal></person-group>. (<year>2003</year>). <article-title>Three-dimensional maximum probability atlas of the human brain, with particular reference to the temporal lobe</article-title>. <source>Hum. Brain Mapp</source>. <volume>19</volume>, <fpage>224</fpage>–<lpage>247</lpage>. <pub-id pub-id-type="doi">10.1002/hbm.10123</pub-id><pub-id pub-id-type="pmid">12874777</pub-id></mixed-citation></ref><ref id="B15"><mixed-citation publication-type="book"><person-group person-group-type="author"><name><surname>Khan</surname><given-names>A. R.</given-names></name><name><surname>Beg</surname><given-names>M. F.</given-names></name></person-group> (<year>2009</year>). <article-title>Multi-structure whole brain registration and population average</article-title>, in <source>Annual International Conference of the IEEE Engineering in Medicine and Biology Society</source> (<publisher-loc>Minneapolis, MN</publisher-loc>) 2009, <fpage>5797</fpage>–<lpage>5800</lpage>. <pub-id pub-id-type="doi">10.1109/IEMBS.2009.5335196</pub-id></mixed-citation></ref><ref id="B16"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Khan</surname><given-names>A.</given-names></name><name><surname>Wang</surname><given-names>L.</given-names></name><name><surname>Beg</surname><given-names>M.</given-names></name></person-group> (<year>2008</year>). <article-title>FreeSurfer-initiated fully-automated subcortical brain segmentation in MRI using large deformation diffeomorphic metric mapping</article-title>. <source>Neuroimage</source><volume>41</volume>, <fpage>735</fpage>–<lpage>746</lpage>. <pub-id pub-id-type="doi">10.1016/j.neuroimage.2008.03.024</pub-id><pub-id pub-id-type="pmid">18455931</pub-id></mixed-citation></ref><ref id="B17"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Krabbe</surname><given-names>K.</given-names></name><name><surname>Karlsborg</surname><given-names>M.</given-names></name><name><surname>Hansen</surname><given-names>A.</given-names></name><name><surname>Werdelin</surname><given-names>L.</given-names></name><name><surname>Mehlsen</surname><given-names>J.</given-names></name><name><surname>Larsson</surname><given-names>H. B. W.</given-names></name><etal></etal></person-group>. (<year>2005</year>). <article-title>Increased intracranial volume in Parkinson's disease</article-title>. <source>J. Neurol. Sci</source>. <volume>239</volume>, <fpage>45</fpage>–<lpage>52</lpage>. <pub-id pub-id-type="doi">10.1016/j.jns.2005.07.013</pub-id><pub-id pub-id-type="pmid">16225890</pub-id></mixed-citation></ref><ref id="B18"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Li</surname><given-names>W.</given-names></name><name><surname>Liu</surname><given-names>J.</given-names></name><name><surname>Skidmore</surname><given-names>F.</given-names></name><name><surname>Liu</surname><given-names>Y.</given-names></name><name><surname>Tian</surname><given-names>J.</given-names></name><name><surname>Li</surname><given-names>K.</given-names></name></person-group> (<year>2010</year>). <article-title>White matter microstructure changes in the thalamus in Parkinson disease with depression: a diffusion tensor MR imaging study</article-title>. <source>Am. J. Neuroradiol</source>. <volume>31</volume>, <fpage>1861</fpage>–<lpage>1866</lpage>. <pub-id pub-id-type="doi">10.3174/ajnr.A2195</pub-id><pub-id pub-id-type="pmid">20705702</pub-id></mixed-citation></ref><ref id="B19"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lilliefors</surname><given-names>H.</given-names></name></person-group> (<year>1967</year>). <article-title>On the Kolmogorov-Smirnov test for normality with mean and variance unknown</article-title>. <source>J. Am. Stat. Assoc</source>. <volume>62</volume>, <fpage>399</fpage>–<lpage>402</lpage><pub-id pub-id-type="doi">10.1080/01621459.1967.10482916</pub-id></mixed-citation></ref><ref id="B20"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>McKeown</surname><given-names>M. J.</given-names></name><name><surname>Uthama</surname><given-names>A.</given-names></name><name><surname>Abugharbieh</surname><given-names>R.</given-names></name><name><surname>Palmer</surname><given-names>S.</given-names></name><name><surname>Lewis</surname><given-names>M.</given-names></name><name><surname>Huang</surname><given-names>X.</given-names></name></person-group> (<year>2008</year>). <article-title>Shape (but not volume) changes in the thalami in Parkinson disease</article-title>. <source>BMC Neurol</source>. <volume>8</volume>:<fpage>8</fpage>. <pub-id pub-id-type="doi">10.1186/1471-2377-8-8</pub-id><pub-id pub-id-type="pmid">18412976</pub-id></mixed-citation></ref><ref id="B21"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nagano-Saito</surname><given-names>A.</given-names></name><name><surname>Washimi</surname><given-names>Y.</given-names></name><name><surname>Arahata</surname><given-names>Y.</given-names></name><name><surname>Kachi</surname><given-names>T.</given-names></name><name><surname>Lerch</surname><given-names>J. P.</given-names></name><name><surname>Evans</surname><given-names>A. C.</given-names></name><etal></etal></person-group>. (<year>2005</year>). <article-title>Cerebral atrophy and its relation to cognitive impairment in Parkinson disease</article-title>. <source>Neurology</source><volume>64</volume>, <fpage>224</fpage>–<lpage>229</lpage>. <pub-id pub-id-type="doi">10.1212/01.WNL.0000149510.41793.50</pub-id><pub-id pub-id-type="pmid">15668417</pub-id></mixed-citation></ref><ref id="B22"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Obeso</surname><given-names>J. A.</given-names></name><name><surname>Rodriguez-Oroz</surname><given-names>M. C.</given-names></name><name><surname>Stamelou</surname><given-names>M.</given-names></name><name><surname>Bhatia</surname><given-names>K. P.</given-names></name><name><surname>Burn</surname><given-names>D. J.</given-names></name></person-group> (<year>2015</year>). <article-title>The expanding universe of disorders of the basal ganglia</article-title>. <source>Lancet</source><volume>384</volume>, <fpage>523</fpage>–<lpage>531</lpage>. <pub-id pub-id-type="doi">10.1016/S0140-6736(13)62418-6</pub-id><pub-id pub-id-type="pmid">24954674</pub-id></mixed-citation></ref><ref id="B23"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Paviour</surname><given-names>D. C.</given-names></name><name><surname>Price</surname><given-names>S. L.</given-names></name><name><surname>Jahanshahi</surname><given-names>M.</given-names></name><name><surname>Lees</surname><given-names>A. J.</given-names></name><name><surname>Fox</surname><given-names>N. C.</given-names></name></person-group> (<year>2006</year>). <article-title>Longitudinal MRI in progressive supranuclear palsy and multiple system atrophy: rates and regions of atrophy</article-title>. <source>Brain J. Neurol</source>. <volume>129(Pt 4)</volume>, <fpage>1040</fpage>–<lpage>1049</lpage>. <pub-id pub-id-type="doi">10.1093/brain/awl021</pub-id><pub-id pub-id-type="pmid">16455792</pub-id></mixed-citation></ref><ref id="B24"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pitcher</surname><given-names>T. L.</given-names></name><name><surname>Melzer</surname><given-names>T. R.</given-names></name><name><surname>MacAskill</surname><given-names>M. R.</given-names></name><name><surname>Graham</surname><given-names>C. F.</given-names></name><name><surname>Livingston</surname><given-names>L.</given-names></name><name><surname>Keenan</surname><given-names>R. J.</given-names></name><etal></etal></person-group>. (<year>2012</year>). <article-title>Reduced striatal volumes in Parkinson's disease: a magnetic resonance imaging study</article-title>. <source>Transl. Neurodegener</source>. <volume>1</volume>:<fpage>17</fpage>. <pub-id pub-id-type="doi">10.1186/2047-9158-1-17</pub-id><pub-id pub-id-type="pmid">23210661</pub-id></mixed-citation></ref><ref id="B25"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pol</surname><given-names>H.</given-names></name><name><surname>van der Flier</surname><given-names>W.</given-names></name></person-group> (<year>2000</year>). <article-title>Frontal lobe damage and thalamic volume changes</article-title>. <source>Neuroreport</source><volume>11</volume>, <fpage>3039</fpage>–<lpage>3041</lpage>. <pub-id pub-id-type="doi">10.1097/00001756-200009110-00042</pub-id><pub-id pub-id-type="pmid">11006990</pub-id></mixed-citation></ref><ref id="B26"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Redgrave</surname><given-names>P.</given-names></name><name><surname>Rodriguez</surname><given-names>M.</given-names></name><name><surname>Smith</surname><given-names>Y.</given-names></name><name><surname>Rodriguez-Oroz</surname><given-names>M. C.</given-names></name><name><surname>Lehericy</surname><given-names>S.</given-names></name><name><surname>Bergman</surname><given-names>H.</given-names></name><etal></etal></person-group>. (<year>2010</year>). <article-title>Goal-directed and habitual control in the basal ganglia: implications for Parkinson's disease</article-title>. <source>Nat. Rev. Neurosci</source>. <volume>11</volume>, <fpage>760</fpage>–<lpage>772</lpage>. <pub-id pub-id-type="doi">10.1038/nrn2915</pub-id><pub-id pub-id-type="pmid">20944662</pub-id></mixed-citation></ref><ref id="B27"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Schulz</surname><given-names>J. B.</given-names></name><name><surname>Skalej</surname><given-names>M.</given-names></name><name><surname>Wedekind</surname><given-names>D.</given-names></name><name><surname>Luft</surname><given-names>A. R.</given-names></name><name><surname>Abele</surname><given-names>M.</given-names></name><name><surname>Voigt</surname><given-names>K.</given-names></name><etal></etal></person-group>. (<year>1999</year>). <article-title>Magnetic resonance imagingbased volumetry differentiates idiopathic Parkinson's syndrome from multiple system atrophy and progressive supranuclear palsy</article-title>. <source>Ann. Neurol</source>. <volume>45</volume>, <fpage>65</fpage>–<lpage>74</lpage>. <pub-id pub-id-type="doi">10.1002/1531-8249(199901)45:1<65::AID-ART12>3.0.CO;2-1</pub-id><pub-id pub-id-type="pmid">9894879</pub-id></mixed-citation></ref><ref id="B28"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sterling</surname><given-names>N. W.</given-names></name><name><surname>Du</surname><given-names>G.</given-names></name><name><surname>Lewis</surname><given-names>M. M.</given-names></name><name><surname>Dimaio</surname><given-names>C.</given-names></name><name><surname>Kong</surname><given-names>L.</given-names></name><name><surname>Eslinger</surname><given-names>P. J.</given-names></name><etal></etal></person-group>. (<year>2013</year>). <article-title>Striatal shape in Parkinson's disease</article-title>. <source>Neurobiol. Aging</source><volume>34</volume>, <fpage>2510</fpage>–<lpage>2516</lpage>. <pub-id pub-id-type="doi">10.1016/j.neurobiolaging.2013.05.017</pub-id><pub-id pub-id-type="pmid">23820588</pub-id></mixed-citation></ref><ref id="B29"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Tzarouchi</surname><given-names>L. C.</given-names></name><name><surname>Astrakas</surname><given-names>L. G.</given-names></name><name><surname>Konitsiotis</surname><given-names>S.</given-names></name><name><surname>Tsouli</surname><given-names>S.</given-names></name><name><surname>Margariti</surname><given-names>P.</given-names></name><name><surname>Zikou</surname><given-names>A.</given-names></name><etal></etal></person-group>. (<year>2010</year>). <article-title>Voxel-based morphometry and Voxel-based relaxometry in parkinsonian variant of multiple system atrophy</article-title>. <source>J. Neuroimaging</source><volume>20</volume>, <fpage>260</fpage>–<lpage>266</lpage>. <pub-id pub-id-type="doi">10.1111/j.1552-6569.2008.00343.x</pub-id><pub-id pub-id-type="pmid">19187475</pub-id></mixed-citation></ref><ref id="B30"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Worsley</surname><given-names>K. J.</given-names></name><name><surname>Marrett</surname><given-names>S.</given-names></name><name><surname>Neelin</surname><given-names>P.</given-names></name><name><surname>Vandal</surname><given-names>A. C.</given-names></name><name><surname>Friston</surname><given-names>K. J.</given-names></name><name><surname>Evans</surname><given-names>A. C.</given-names></name></person-group> (<year>1996</year>). <article-title>A unified statistical approach for determining significant signals in images of cerebral activation</article-title>. <source>Hum. Brain Mapp</source>. <volume>4</volume>, <fpage>58</fpage>–<lpage>73</lpage>. <pub-id pub-id-type="doi">10.1002/(SICI)1097-0193(1996)4:1<58::AID-HBM4>3.0.CO;2-O</pub-id><pub-id pub-id-type="pmid">20408186</pub-id></mixed-citation></ref><ref id="B31"><mixed-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Worsley</surname><given-names>K. J.</given-names></name><name><surname>Taylor</surname><given-names>J. E.</given-names></name><name><surname>Tomaiuolo</surname><given-names>F.</given-names></name><name><surname>Lerch</surname><given-names>J.</given-names></name></person-group> (<year>2004</year>). <article-title>Unified univariate and multivariate random field theory</article-title>. <source>Neuroimage</source><volume>23</volume><issue>Suppl. 1</issue>, <fpage>S189</fpage>–<lpage>S195</lpage>. <pub-id pub-id-type="doi">10.1016/j.neuroimage.2004.07.026</pub-id><pub-id pub-id-type="pmid">15501088</pub-id></mixed-citation></ref></ref-list></back></pmc></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Canada/explor/ParkinsonCanadaV1/Data/Pmc/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 0009860 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Pmc/Corpus/biblio.hfd -nk 0009860 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Canada
|area= ParkinsonCanadaV1
|flux= Pmc
|étape= Corpus
|type= RBID
|clé=
|texte=
}}
| This area was generated with Dilib version V0.6.29. |  |