<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-04199794v1</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-22T19:17:34+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Comprehensive analysis of synthetic learning applied to neonatal brain MRI segmentation</title>
            <author role="aut">
              <persName>
                <forename type="first">R</forename>
                <surname>Valabregue</surname>
              </persName>
              <idno type="halauthorid">510107-0</idno>
              <affiliation ref="#struct-542029"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">F</forename>
                <surname>Girka</surname>
              </persName>
              <idno type="halauthorid">2892083-0</idno>
              <affiliation ref="#struct-542029"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">A</forename>
                <surname>Pron</surname>
              </persName>
              <idno type="halauthorid">2080792-0</idno>
              <affiliation ref="#struct-202046"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">F</forename>
                <surname>Rousseau</surname>
              </persName>
              <idno type="halauthorid">887978-0</idno>
              <affiliation ref="#struct-491118"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">G</forename>
                <surname>Auzias</surname>
              </persName>
              <idno type="halauthorid">2583204-0</idno>
              <affiliation ref="#struct-202046"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>romain</forename>
                <surname>valabregue</surname>
              </persName>
              <email type="md5">9f304a7188d287624061b4d70e4d6ea7</email>
              <email type="domain">upmc.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2023-09-08 09:44:39</date>
              <date type="whenModified">2026-01-19 16:46:21</date>
              <date type="whenReleased">2023-09-11 10:51:57</date>
              <date type="whenProduced">2023-09-08</date>
              <date type="whenEndEmbargoed">2023-09-08</date>
              <ref type="file" target="https://hal.science/hal-04199794v1/document">
                <date notBefore="2023-09-08"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-04199794v1/file/main.pdf" id="file-4199794-3662712">
                <date notBefore="2023-09-08"/>
              </ref>
              <ref type="externalLink" target="http://arxiv.org/pdf/2309.05306"/>
            </edition>
            <edition n="v2">
              <date type="whenSubmitted">2024-06-11 17:18:15</date>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="2100">
                <persName>
                  <forename>romain</forename>
                  <surname>valabregue</surname>
                </persName>
                <email type="md5">9f304a7188d287624061b4d70e4d6ea7</email>
                <email type="domain">upmc.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-04199794</idno>
            <idno type="halUri">https://hal.science/hal-04199794</idno>
            <idno type="halBibtex">valabregue:hal-04199794</idno>
            <idno type="halRefHtml">2023</idno>
            <idno type="halRef">2023</idno>
            <availability status="restricted">
              <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal<ref corresp="#file-4199794-3662712"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt/>
          <notesStmt>
            <note type="audience" n="2">International</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Comprehensive analysis of synthetic learning applied to neonatal brain MRI segmentation</title>
                <author role="aut">
                  <persName>
                    <forename type="first">R</forename>
                    <surname>Valabregue</surname>
                  </persName>
                  <idno type="halauthorid">510107-0</idno>
                  <affiliation ref="#struct-542029"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">F</forename>
                    <surname>Girka</surname>
                  </persName>
                  <idno type="halauthorid">2892083-0</idno>
                  <affiliation ref="#struct-542029"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">A</forename>
                    <surname>Pron</surname>
                  </persName>
                  <idno type="halauthorid">2080792-0</idno>
                  <affiliation ref="#struct-202046"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">F</forename>
                    <surname>Rousseau</surname>
                  </persName>
                  <idno type="halauthorid">887978-0</idno>
                  <affiliation ref="#struct-491118"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">G</forename>
                    <surname>Auzias</surname>
                  </persName>
                  <idno type="halauthorid">2583204-0</idno>
                  <affiliation ref="#struct-202046"/>
                </author>
              </analytic>
              <monogr>
                <imprint/>
              </monogr>
              <idno type="arxiv">2309.05306</idno>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">Synthetic MRI</term>
                <term xml:lang="en">synthetic</term>
                <term xml:lang="en">MRI</term>
                <term xml:lang="en">brain</term>
                <term xml:lang="en">newborn</term>
                <term xml:lang="en">segmentation</term>
                <term xml:lang="en">deep learning</term>
                <term xml:lang="en">deep learning segmentation newborn brain MRI synthetic dHCP</term>
                <term xml:lang="en">Synthetic training images</term>
                <term xml:lang="en">dHCP</term>
                <term xml:lang="en">Brain</term>
                <term xml:lang="en">Infant</term>
                <term xml:lang="en">Newborn</term>
                <term xml:lang="en">Segmentation</term>
                <term xml:lang="en">Deep learning DL</term>
              </keywords>
              <classCode scheme="halDomain" n="stat.ml">Statistics [stat]/Machine Learning [stat.ML]</classCode>
              <classCode scheme="halDomain" n="info.info-im">Computer Science [cs]/Medical Imaging</classCode>
              <classCode scheme="halTypology" n="UNDEFINED">Preprints, Working Papers, ...</classCode>
              <classCode scheme="halOldTypology" n="UNDEFINED">Preprints, Working Papers, ...</classCode>
              <classCode scheme="halTreeTypology" n="UNDEFINED.PREPRINT">Preprints, Working Papers, ... - Preprint</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>Brain segmentation from neonatal MRI images is a very challenging task due to large changes in the shape of cerebral structures and variations in signal intensities reflecting the gestational process. In this context, there is a clear need for segmentation techniques that are robust to variations in image contrast and to the spatial configuration of anatomical structures. In this work, we evaluate the potential of synthetic learning, a contrast-independent model trained using synthetic images generated from the ground truth labels of very few subjects.We base our experiments on the dataset released by the developmental Human Connectome Project, for which high-quality T1- and T2-weighted images are available for more than 700 babies aged between 26 and 45 weeks post-conception. First, we confirm the impressive performance of a standard Unet trained on a few T2-weighted volumes, but also confirm that such models learn intensity-related features specific to the training domain. We then evaluate the synthetic learning approach and confirm its robustness to variations in image contrast by reporting the capacity of such a model to segment both T1- and T2-weighted images from the same individuals. However, we observe a clear influence of the age of the baby on the predictions. We improve the performance of this model by enriching the synthetic training set with realistic motion artifacts and over-segmentation of the white matter. Based on extensive visual assessment, we argue that the better performance of the model trained on real T2w data may be due to systematic errors in the ground truth. We propose an original experiment combining two definitions of the ground truth allowing us to show that learning from real data will reproduce any systematic bias from the training set, while synthetic models can avoid this limitation. Overall, our experiments confirm that synthetic learning is an effective solution for segmenting neonatal brain MRI. Our adapted synthetic learning approach combines key features that will be instrumental for large multi-site studies and clinical applications.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="regrouplaboratory" xml:id="struct-542029" status="VALID">
          <idno type="IdRef">162466021</idno>
          <idno type="ISNI">0000000406205939</idno>
          <idno type="RNSR">200918525B</idno>
          <idno type="ROR">https://ror.org/050gn5214</idno>
          <orgName>Institut du Cerveau = Paris Brain Institute</orgName>
          <orgName type="acronym">ICM</orgName>
          <desc>
            <address>
              <addrLine>47-83 Boulevard de l'Hôpital 75651 Paris Cedex 13</addrLine>
              <country key="FR"/>
            </address>
          </desc>
          <listRelation>
            <relation active="#struct-300068" type="direct"/>
            <relation name="U1127" active="#struct-303623" type="direct"/>
            <relation active="#struct-353778" type="direct"/>
            <relation active="#struct-413221" type="direct"/>
            <relation name="UM75" active="#struct-413221" type="direct"/>
            <relation name="UMR7225" active="#struct-441569" type="direct"/>
          </listRelation>
        </org>
        <org type="laboratory" xml:id="struct-202046" status="VALID">
          <idno type="IdRef">171900928</idno>
          <idno type="RNSR">201220346T</idno>
          <idno type="ROR">043hw6336</idno>
          <orgName>Institut de Neurosciences de la Timone</orgName>
          <orgName type="acronym">INT</orgName>
          <date type="start">2012-01-01</date>
          <desc>
            <address>
              <addrLine>Faculté de Médecine - Bâtiment Neurosciences 27, Bd Jean Moulin - 13385 Marseille Cedex 05</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.int.univ-amu.fr/</ref>
          </desc>
          <listRelation>
            <relation name="UMR7289" active="#struct-198056" type="direct"/>
            <relation name="UMR7289" active="#struct-441569" type="direct"/>
          </listRelation>
        </org>
        <org type="laboratory" xml:id="struct-491118" status="VALID">
          <idno type="RNSR">201220084H</idno>
          <idno type="ROR">https://ror.org/05mqemx33</idno>
          <orgName>Laboratoire de Traitement de l'Information Medicale</orgName>
          <orgName type="acronym">LaTIM</orgName>
          <desc>
            <address>
              <addrLine>CHRU Morvan, Bâtiment 1, 1er étage, bureau 1056, 2 avenue Foch, 29609 Brest, France</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://latim.univ-brest.fr</ref>
          </desc>
          <listRelation>
            <relation active="#struct-300314" type="direct"/>
            <relation name="UMR1101" active="#struct-303623" type="direct"/>
            <relation active="#struct-356603" type="direct"/>
            <relation active="#struct-481355" type="direct"/>
            <relation active="#struct-302102" type="indirect"/>
            <relation active="#struct-519898" type="direct"/>
          </listRelation>
        </org>
        <org type="regroupinstitution" xml:id="struct-300068" status="VALID">
          <idno type="ROR">https://ror.org/00pg5jh14</idno>
          <orgName>Assistance publique - Hôpitaux de Paris (AP-HP)</orgName>
          <orgName type="acronym">AP-HP</orgName>
          <desc>
            <address>
              <country key="FR"/>
            </address>
          </desc>
        </org>
        <org type="institution" xml:id="struct-303623" status="VALID">
          <idno type="IdRef">026388278</idno>
          <idno type="ROR">https://ror.org/02vjkv261</idno>
          <orgName>Institut National de la Santé et de la Recherche Médicale</orgName>
          <orgName type="acronym">INSERM</orgName>
          <desc>
            <address>
              <addrLine>101, rue de Tolbiac, 75013 Paris</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.inserm.fr</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-353778" status="VALID">
          <idno type="ROR">https://ror.org/02mh9a093</idno>
          <orgName>CHU Pitié-Salpêtrière [AP-HP]</orgName>
          <desc>
            <address>
              <addrLine>47-83 Boulevard de l'Hôpital, 75013 Paris</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.aphp.fr/contenu/hopital-universitaire-pitie-salpetriere-0</ref>
          </desc>
          <listRelation>
            <relation active="#struct-300068" type="direct"/>
            <relation active="#struct-413221" type="direct"/>
          </listRelation>
        </org>
        <org type="regroupinstitution" xml:id="struct-413221" status="VALID">
          <idno type="IdRef">221333754</idno>
          <idno type="ROR">https://ror.org/02en5vm52</idno>
          <orgName>Sorbonne Université</orgName>
          <orgName type="acronym">SU</orgName>
          <date type="start">2018-01-01</date>
          <desc>
            <address>
              <addrLine>21 rue de l’École de médecine - 75006 Paris</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.sorbonne-universite.fr/</ref>
          </desc>
        </org>
        <org type="regroupinstitution" xml:id="struct-441569" status="VALID">
          <idno type="IdRef">02636817X</idno>
          <idno type="ISNI">0000000122597504</idno>
          <idno type="ROR">https://ror.org/02feahw73</idno>
          <orgName>Centre National de la Recherche Scientifique</orgName>
          <orgName type="acronym">CNRS</orgName>
          <date type="start">1939-10-19</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.cnrs.fr/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-198056" status="VALID">
          <idno type="IdRef">15863621X</idno>
          <idno type="ISNI">0000 0001 2176 4817</idno>
          <idno type="ROR">https://ror.org/035xkbk20</idno>
          <orgName>Aix Marseille Université</orgName>
          <orgName type="acronym">AMU</orgName>
          <date type="start">2012-01-01</date>
          <desc>
            <address>
              <addrLine>Aix-Marseille UniversitéJardins du Pharo58 Boulevard Charles Livon13284 Marseille cedex 7</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.univ-amu.fr/</ref>
          </desc>
        </org>
        <org type="regroupinstitution" xml:id="struct-300314" status="VALID">
          <idno type="IdRef">291809022</idno>
          <idno type="ISNI">0000000121880893</idno>
          <idno type="ROR">https://ror.org/0372th171</idno>
          <orgName>Université de Brest</orgName>
          <orgName type="acronym">UBO EPE</orgName>
          <desc>
            <address>
              <addrLine>Université de Bretagne Occidentale - 3 Rue des Archives 29238, Brest</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.univ-brest.fr/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-356603" status="VALID">
          <orgName>Centre Hospitalier Régional Universitaire de Brest</orgName>
          <orgName type="acronym">CHRU Brest</orgName>
          <desc>
            <address>
              <country key="FR"/>
            </address>
          </desc>
        </org>
        <org type="institution" xml:id="struct-481355" status="VALID">
          <idno type="IdRef">202743233</idno>
          <idno type="ROR">https://ror.org/030hj3061</idno>
          <orgName>IMT Atlantique</orgName>
          <orgName type="acronym">IMT Atlantique</orgName>
          <date type="start">2017-01-01</date>
          <desc>
            <address>
              <addrLine>Campus Brest : Technopôle Brest-Iroise CS 8381829238 BREST Cedex 3 -Campus Nantes : 4, rue Alfred Kastler- La chantrerie 44300 NANTES -Campus Rennes :  2 Rue de la Châtaigneraie, 35510 CESSON SEVIGNE</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.imt-atlantique.fr</ref>
          </desc>
          <listRelation>
            <relation active="#struct-302102" type="direct"/>
          </listRelation>
        </org>
        <org type="regroupinstitution" xml:id="struct-302102" status="VALID">
          <idno type="IdRef">192427156</idno>
          <idno type="ISNI">000000012202567X</idno>
          <idno type="ROR">https://ror.org/025vp2923</idno>
          <idno type="Wikidata">Q27962533</idno>
          <orgName>Institut Mines-Télécom [Paris]</orgName>
          <orgName type="acronym">IMT</orgName>
          <date type="start">2012-03-01</date>
          <desc>
            <address>
              <addrLine>19 Place Marguerite Perey, 91120 Palaiseau</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.imt.fr/</ref>
          </desc>
        </org>
        <org type="regrouplaboratory" xml:id="struct-519898" status="VALID">
          <idno type="RNSR">201722456P</idno>
          <orgName>Institut Brestois Santé Agro Matière</orgName>
          <orgName type="acronym">IBSAM</orgName>
          <date type="start">2017-10-12</date>
          <desc>
            <address>
              <addrLine>Université de Brest- UBO29238 Brest</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.univ-brest.fr/menu/recherche-innovation/Laboratoires/Axe+Sant%C3%A9-+Agro-+Mati%C3%A8re/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-300314" type="direct"/>
          </listRelation>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>