Residual Embedding Similarity-Based Network Selection for Predicting Brain Network Evolution Trajectory from a Single Observation

While existing predictive frameworks are able to handle Euclidean structured data (i.e, brain images), they might fail to generalize to geometric non-Euclidean data such as brain networks. Besides, these are rooted the sample selection step in using Euclidean or learned similarity measure between ve...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2020-09
Hauptverfasser: Goktas, Ahmet Serkan, Bessadok, Alaa, Rekik, Islem
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Goktas, Ahmet Serkan
Bessadok, Alaa
Rekik, Islem
description While existing predictive frameworks are able to handle Euclidean structured data (i.e, brain images), they might fail to generalize to geometric non-Euclidean data such as brain networks. Besides, these are rooted the sample selection step in using Euclidean or learned similarity measure between vectorized training and testing brain networks. Such sample connectomic representation might include irrelevant and redundant features that could mislead the training sample selection step. Undoubtedly, this fails to exploit and preserve the topology of the brain connectome. To overcome this major drawback, we propose Residual Embedding Similarity-Based Network selection (RESNets) for predicting brain network evolution trajectory from a single timepoint. RESNets first learns a compact geometric embedding of each training and testing sample using adversarial connectome embedding network. This nicely reduces the high-dimensionality of brain networks while preserving their topological properties via graph convolutional networks. Next, to compute the similarity between subjects, we introduce the concept of a connectional brain template (CBT), a fixed network reference, where we further represent each training and testing network as a deviation from the reference CBT in the embedding space. As such, we select the most similar training subjects to the testing subject at baseline by comparing their learned residual embeddings with respect to the pre-defined CBT. Once the best training samples are selected at baseline, we simply average their corresponding brain networks at follow-up timepoints to predict the evolution trajectory of the testing network. Our experiments on both healthy and disordered brain networks demonstrate the success of our proposed method in comparison to RESNets ablated versions and traditional approaches.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2445806012</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2445806012</sourcerecordid><originalsourceid>FETCH-proquest_journals_24458060123</originalsourceid><addsrcrecordid>eNqNjMEKwjAQRIMgKOo_LHguxLRVz0rFk4p6L9FsJTVNdNMqHv1zo4hnT8Mw702LdUUcj6JpIkSHDbwvOediPBFpGnfZc4teq0YayKoDKqXtCXa60kaSrh_RTHpUsML67ugMOzR4rLWzUDiCDaHSoQZjRlLbH5bdnGk-2J5kGQxHDyjIVSDDtz0ZhPXBI93kG-qzdiGNx8E3e2y4yPbzZXQhd23Q13npGrJhykWSpFM-5iMR_0e9AMRzUZY</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2445806012</pqid></control><display><type>article</type><title>Residual Embedding Similarity-Based Network Selection for Predicting Brain Network Evolution Trajectory from a Single Observation</title><source>Free E- Journals</source><creator>Goktas, Ahmet Serkan ; Bessadok, Alaa ; Rekik, Islem</creator><creatorcontrib>Goktas, Ahmet Serkan ; Bessadok, Alaa ; Rekik, Islem</creatorcontrib><description>While existing predictive frameworks are able to handle Euclidean structured data (i.e, brain images), they might fail to generalize to geometric non-Euclidean data such as brain networks. Besides, these are rooted the sample selection step in using Euclidean or learned similarity measure between vectorized training and testing brain networks. Such sample connectomic representation might include irrelevant and redundant features that could mislead the training sample selection step. Undoubtedly, this fails to exploit and preserve the topology of the brain connectome. To overcome this major drawback, we propose Residual Embedding Similarity-Based Network selection (RESNets) for predicting brain network evolution trajectory from a single timepoint. RESNets first learns a compact geometric embedding of each training and testing sample using adversarial connectome embedding network. This nicely reduces the high-dimensionality of brain networks while preserving their topological properties via graph convolutional networks. Next, to compute the similarity between subjects, we introduce the concept of a connectional brain template (CBT), a fixed network reference, where we further represent each training and testing network as a deviation from the reference CBT in the embedding space. As such, we select the most similar training subjects to the testing subject at baseline by comparing their learned residual embeddings with respect to the pre-defined CBT. Once the best training samples are selected at baseline, we simply average their corresponding brain networks at follow-up timepoints to predict the evolution trajectory of the testing network. Our experiments on both healthy and disordered brain networks demonstrate the success of our proposed method in comparison to RESNets ablated versions and traditional approaches.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Ablation ; Brain ; Embedding ; Euclidean geometry ; Evolution ; Networks ; Similarity ; Topology ; Training</subject><ispartof>arXiv.org, 2020-09</ispartof><rights>2020. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Goktas, Ahmet Serkan</creatorcontrib><creatorcontrib>Bessadok, Alaa</creatorcontrib><creatorcontrib>Rekik, Islem</creatorcontrib><title>Residual Embedding Similarity-Based Network Selection for Predicting Brain Network Evolution Trajectory from a Single Observation</title><title>arXiv.org</title><description>While existing predictive frameworks are able to handle Euclidean structured data (i.e, brain images), they might fail to generalize to geometric non-Euclidean data such as brain networks. Besides, these are rooted the sample selection step in using Euclidean or learned similarity measure between vectorized training and testing brain networks. Such sample connectomic representation might include irrelevant and redundant features that could mislead the training sample selection step. Undoubtedly, this fails to exploit and preserve the topology of the brain connectome. To overcome this major drawback, we propose Residual Embedding Similarity-Based Network selection (RESNets) for predicting brain network evolution trajectory from a single timepoint. RESNets first learns a compact geometric embedding of each training and testing sample using adversarial connectome embedding network. This nicely reduces the high-dimensionality of brain networks while preserving their topological properties via graph convolutional networks. Next, to compute the similarity between subjects, we introduce the concept of a connectional brain template (CBT), a fixed network reference, where we further represent each training and testing network as a deviation from the reference CBT in the embedding space. As such, we select the most similar training subjects to the testing subject at baseline by comparing their learned residual embeddings with respect to the pre-defined CBT. Once the best training samples are selected at baseline, we simply average their corresponding brain networks at follow-up timepoints to predict the evolution trajectory of the testing network. Our experiments on both healthy and disordered brain networks demonstrate the success of our proposed method in comparison to RESNets ablated versions and traditional approaches.</description><subject>Ablation</subject><subject>Brain</subject><subject>Embedding</subject><subject>Euclidean geometry</subject><subject>Evolution</subject><subject>Networks</subject><subject>Similarity</subject><subject>Topology</subject><subject>Training</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNjMEKwjAQRIMgKOo_LHguxLRVz0rFk4p6L9FsJTVNdNMqHv1zo4hnT8Mw702LdUUcj6JpIkSHDbwvOediPBFpGnfZc4teq0YayKoDKqXtCXa60kaSrh_RTHpUsML67ugMOzR4rLWzUDiCDaHSoQZjRlLbH5bdnGk-2J5kGQxHDyjIVSDDtz0ZhPXBI93kG-qzdiGNx8E3e2y4yPbzZXQhd23Q13npGrJhykWSpFM-5iMR_0e9AMRzUZY</recordid><startdate>20200923</startdate><enddate>20200923</enddate><creator>Goktas, Ahmet Serkan</creator><creator>Bessadok, Alaa</creator><creator>Rekik, Islem</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20200923</creationdate><title>Residual Embedding Similarity-Based Network Selection for Predicting Brain Network Evolution Trajectory from a Single Observation</title><author>Goktas, Ahmet Serkan ; Bessadok, Alaa ; Rekik, Islem</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_24458060123</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Ablation</topic><topic>Brain</topic><topic>Embedding</topic><topic>Euclidean geometry</topic><topic>Evolution</topic><topic>Networks</topic><topic>Similarity</topic><topic>Topology</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Goktas, Ahmet Serkan</creatorcontrib><creatorcontrib>Bessadok, Alaa</creatorcontrib><creatorcontrib>Rekik, Islem</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Goktas, Ahmet Serkan</au><au>Bessadok, Alaa</au><au>Rekik, Islem</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Residual Embedding Similarity-Based Network Selection for Predicting Brain Network Evolution Trajectory from a Single Observation</atitle><jtitle>arXiv.org</jtitle><date>2020-09-23</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>While existing predictive frameworks are able to handle Euclidean structured data (i.e, brain images), they might fail to generalize to geometric non-Euclidean data such as brain networks. Besides, these are rooted the sample selection step in using Euclidean or learned similarity measure between vectorized training and testing brain networks. Such sample connectomic representation might include irrelevant and redundant features that could mislead the training sample selection step. Undoubtedly, this fails to exploit and preserve the topology of the brain connectome. To overcome this major drawback, we propose Residual Embedding Similarity-Based Network selection (RESNets) for predicting brain network evolution trajectory from a single timepoint. RESNets first learns a compact geometric embedding of each training and testing sample using adversarial connectome embedding network. This nicely reduces the high-dimensionality of brain networks while preserving their topological properties via graph convolutional networks. Next, to compute the similarity between subjects, we introduce the concept of a connectional brain template (CBT), a fixed network reference, where we further represent each training and testing network as a deviation from the reference CBT in the embedding space. As such, we select the most similar training subjects to the testing subject at baseline by comparing their learned residual embeddings with respect to the pre-defined CBT. Once the best training samples are selected at baseline, we simply average their corresponding brain networks at follow-up timepoints to predict the evolution trajectory of the testing network. Our experiments on both healthy and disordered brain networks demonstrate the success of our proposed method in comparison to RESNets ablated versions and traditional approaches.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2020-09
issn 2331-8422
language eng
recordid cdi_proquest_journals_2445806012
source Free E- Journals
subjects Ablation
Brain
Embedding
Euclidean geometry
Evolution
Networks
Similarity
Topology
Training
title Residual Embedding Similarity-Based Network Selection for Predicting Brain Network Evolution Trajectory from a Single Observation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-03T15%3A51%3A49IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Residual%20Embedding%20Similarity-Based%20Network%20Selection%20for%20Predicting%20Brain%20Network%20Evolution%20Trajectory%20from%20a%20Single%20Observation&rft.jtitle=arXiv.org&rft.au=Goktas,%20Ahmet%20Serkan&rft.date=2020-09-23&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2445806012%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2445806012&rft_id=info:pmid/&rfr_iscdi=true