001005113 001__ 1005113
001005113 005__ 20231027114355.0
001005113 0247_ $$2doi$$a10.1016/j.neuroimage.2023.119947
001005113 0247_ $$2ISSN$$a1053-8119
001005113 0247_ $$2ISSN$$a1095-9572
001005113 0247_ $$2Handle$$a2128/34026
001005113 0247_ $$2pmid$$a36801372
001005113 0247_ $$2WOS$$aWOS:000954924800001
001005113 0247_ $$2datacite_doi$$a10.34734/FZJ-2023-01312
001005113 037__ $$aFZJ-2023-01312
001005113 082__ $$a610
001005113 1001_ $$0P:(DE-Juel1)177823$$aMore, Shammi$$b0
001005113 245__ $$aBrain-age prediction: A systematic comparison of machine learning workflows
001005113 260__ $$aOrlando, Fla.$$bAcademic Press$$c2023
001005113 3367_ $$2DRIVER$$aarticle
001005113 3367_ $$2DataCite$$aOutput Types/Journal article
001005113 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1689245059_18704
001005113 3367_ $$2BibTeX$$aARTICLE
001005113 3367_ $$2ORCID$$aJOURNAL_ARTICLE
001005113 3367_ $$00$$2EndNote$$aJournal Article
001005113 520__ $$aThe difference between age predicted using anatomical brain scans and chronological age, i.e., the brain-age delta, provides a proxy for atypical aging. Various data representations and machine learning (ML) algorithms have been used for brain-age estimation. However, how these choices compare on performance criteria important for real-world applications, such as; (1) within-dataset accuracy, (2) cross-dataset generalization, (3) test-retest reliability, and (4) longitudinal consistency, remains uncharacterized. We evaluated 128 workflows consisting of 16 feature representations derived from gray matter (GM) images and eight ML algorithms with diverse inductive biases. Using four large neuroimaging databases covering the adult lifespan (total N = 2953, 18–88 years), we followed a systematic model selection procedure by sequentially applying stringent criteria. The 128 workflows showed a within-dataset mean absolute error (MAE) between 4.73–8.38 years, from which 32 broadly sampled workflows showed a cross-dataset MAE between 5.23–8.98 years. The test-retest reliability and longitudinal consistency of the top 10 workflows were comparable. The choice of feature representation and the ML algorithm both affected the performance. Specifically, voxel-wise feature spaces (smoothed and resampled), with and without principal components analysis, with non-linear and kernel-based ML algorithms performed well. Strikingly, the correlation of brain-age delta with behavioral measures disagreed between within-dataset and cross-dataset predictions. Application of the best-performing workflow on the ADNI sample showed a significantly higher brain-age delta in Alzheimer's and mild cognitive impairment patients compared to healthy controls. However, in the presence of age bias, the delta estimates in the patients varied depending on the sample used for bias correction. Taken together, brain-age shows promise, but further evaluation and improvements are needed for its real-world application.
001005113 536__ $$0G:(DE-HGF)POF4-5251$$a5251 - Multilevel Brain Organization and Variability (POF4-525)$$cPOF4-525$$fPOF IV$$x0
001005113 536__ $$0G:(GEPRIS)432015680$$aDFG project 432015680 - Automatisierte Gehirnalterung-Vorhersage und deren Interpretation $$c432015680$$x1
001005113 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
001005113 7001_ $$0P:(DE-Juel1)180946$$aAntonopoulos, Georgios$$b1
001005113 7001_ $$0P:(DE-Juel1)131684$$aHoffstaedter, Felix$$b2
001005113 7001_ $$0P:(DE-HGF)0$$aCaspers, Julian$$b3
001005113 7001_ $$0P:(DE-Juel1)131678$$aEickhoff, Simon B.$$b4
001005113 7001_ $$0P:(DE-Juel1)172843$$aPatil, Kaustubh R.$$b5$$eCorresponding author
001005113 7001_ $$0P:(DE-HGF)0$$aInitiative, Alzheimer's Disease Neuroimaging$$b6
001005113 773__ $$0PERI:(DE-600)1471418-8$$a10.1016/j.neuroimage.2023.119947$$gVol. 270, p. 119947 -$$p119947 -$$tNeuroImage$$v270$$x1053-8119$$y2023
001005113 8564_ $$uhttps://juser.fz-juelich.de/record/1005113/files/1-s2.0-S1053811923000940-main.pdf$$yOpenAccess
001005113 8564_ $$uhttps://juser.fz-juelich.de/record/1005113/files/Brainage_paper_SMore_manuscript.pdf$$yOpenAccess
001005113 8767_ $$d2023-03-09$$eAPC$$jZahlung erfolgt
001005113 909CO $$ooai:juser.fz-juelich.de:1005113$$pdnbdelivery$$popenCost$$pVDB$$pdriver$$pOpenAPC$$popen_access$$popenaire
001005113 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)177823$$aForschungszentrum Jülich$$b0$$kFZJ
001005113 9101_ $$0I:(DE-HGF)0$$6P:(DE-Juel1)177823$$a HHU Düsseldorf$$b0
001005113 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)180946$$aForschungszentrum Jülich$$b1$$kFZJ
001005113 9101_ $$0I:(DE-HGF)0$$6P:(DE-Juel1)180946$$a HHU Düsseldorf$$b1
001005113 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)131684$$aForschungszentrum Jülich$$b2$$kFZJ
001005113 9101_ $$0I:(DE-HGF)0$$6P:(DE-Juel1)131684$$a HHU Düsseldorf$$b2
001005113 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)131678$$aForschungszentrum Jülich$$b4$$kFZJ
001005113 9101_ $$0I:(DE-HGF)0$$6P:(DE-Juel1)131678$$a HHU Düsseldorf$$b4
001005113 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)172843$$aForschungszentrum Jülich$$b5$$kFZJ
001005113 9101_ $$0I:(DE-HGF)0$$6P:(DE-Juel1)172843$$a HHU Düsseldorf$$b5
001005113 9131_ $$0G:(DE-HGF)POF4-525$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5251$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vDecoding Brain Organization and Dysfunction$$x0
001005113 9141_ $$y2023
001005113 915pc $$0PC:(DE-HGF)0000$$2APC$$aAPC keys set
001005113 915pc $$0PC:(DE-HGF)0001$$2APC$$aLocal Funding
001005113 915pc $$0PC:(DE-HGF)0002$$2APC$$aDFG OA Publikationskosten
001005113 915pc $$0PC:(DE-HGF)0003$$2APC$$aDOAJ Journal
001005113 915__ $$0StatID:(DE-HGF)1190$$2StatID$$aDBCoverage$$bBiological Abstracts$$d2022-11-12
001005113 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2022-11-12
001005113 915__ $$0StatID:(DE-HGF)0700$$2StatID$$aFees$$d2022-11-12
001005113 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001005113 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2022-11-12
001005113 915__ $$0StatID:(DE-HGF)0561$$2StatID$$aArticle Processing Charges$$d2022-11-12
001005113 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
001005113 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2023-05-02T08:47:40Z
001005113 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2023-05-02T08:47:40Z
001005113 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Anonymous peer review$$d2023-05-02T08:47:40Z
001005113 915__ $$0StatID:(DE-HGF)0420$$2StatID$$aNationallizenz$$d2023-10-21$$wger
001005113 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bNEUROIMAGE : 2022$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)1050$$2StatID$$aDBCoverage$$bBIOSIS Previews$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)1030$$2StatID$$aDBCoverage$$bCurrent Contents - Life Sciences$$d2023-10-21
001005113 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bNEUROIMAGE : 2022$$d2023-10-21
001005113 920__ $$lyes
001005113 9201_ $$0I:(DE-Juel1)INM-7-20090406$$kINM-7$$lGehirn & Verhalten$$x0
001005113 980__ $$ajournal
001005113 980__ $$aVDB
001005113 980__ $$aUNRESTRICTED
001005113 980__ $$aI:(DE-Juel1)INM-7-20090406
001005113 980__ $$aAPC
001005113 9801_ $$aAPC
001005113 9801_ $$aFullTexts