000996722 001__ 996722
000996722 005__ 20231027114354.0
000996722 0247_ $$2doi$$a10.1016/j.neuroimage.2023.119929
000996722 0247_ $$2ISSN$$a1053-8119
000996722 0247_ $$2ISSN$$a1095-9572
000996722 0247_ $$2Handle$$a2128/34042
000996722 0247_ $$2pmid$$a36740029
000996722 0247_ $$2WOS$$aWOS:000943220200001
000996722 037__ $$aFZJ-2023-01142
000996722 082__ $$a610
000996722 1001_ $$0P:(DE-HGF)0$$aWang, Di$$b0
000996722 245__ $$aDeep neural network heatmaps capture Alzheimer’s disease patterns reported in a large meta-analysis of neuroimaging studies
000996722 260__ $$aOrlando, Fla.$$bAcademic Press$$c2023
000996722 3367_ $$2DRIVER$$aarticle
000996722 3367_ $$2DataCite$$aOutput Types/Journal article
000996722 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1677570883_13022
000996722 3367_ $$2BibTeX$$aARTICLE
000996722 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000996722 3367_ $$00$$2EndNote$$aJournal Article
000996722 520__ $$aDeep neural networks currently provide the most advanced and accurate machine learning models to distinguish between structural MRI scans of subjects with Alzheimer's disease and healthy controls. Unfortunately, the subtle brain alterations captured by these models are difficult to interpret because of the complexity of these multi-layer and non-linear models. Several heatmap methods have been proposed to address this issue and analyze the imaging patterns extracted from the deep neural networks, but no quantitative comparison between these methods has been carried out so far. In this work, we explore these questions by deriving heatmaps from Convolutional Neural Networks (CNN) trained using T1 MRI scans of the ADNI data set and by comparing these heatmaps with brain maps corresponding to Support Vector Machine (SVM) activation patterns. Three prominent heatmap methods are studied: Layer-wise Relevance Propagation (LRP), Integrated Gradients (IG), and Guided Grad-CAM (GGC). Contrary to prior studies where the quality of heatmaps was visually or qualitatively assessed, we obtained precise quantitative measures by computing overlap with a ground-truth map from a large meta-analysis that combined 77 voxel-based morphometry (VBM) studies independently from ADNI. Our results indicate that all three heatmap methods were able to capture brain regions covering the meta-analysis map and achieved better results than SVM activation patte
000996722 536__ $$0G:(DE-HGF)POF4-5253$$a5253 - Neuroimaging (POF4-525)$$cPOF4-525$$fPOF IV$$x0
000996722 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
000996722 7001_ $$0P:(DE-HGF)0$$aHonnorat, Nicolas$$b1
000996722 7001_ $$0P:(DE-HGF)0$$aFox, Peter T.$$b2
000996722 7001_ $$0P:(DE-HGF)0$$aRitter, Kerstin$$b3
000996722 7001_ $$0P:(DE-Juel1)131678$$aEickhoff, Simon B.$$b4$$ufzj
000996722 7001_ $$0P:(DE-HGF)0$$aSeshadri, Sudha$$b5
000996722 7001_ $$0P:(DE-HGF)0$$aHabes, Mohamad$$b6$$eCorresponding author
000996722 773__ $$0PERI:(DE-600)1471418-8$$a10.1016/j.neuroimage.2023.119929$$gp. 119929 -$$p119929 -$$tNeuroImage$$v269$$x1053-8119$$y2023
000996722 8564_ $$uhttps://juser.fz-juelich.de/record/996722/files/1-s2.0-S1053811923000770-main.pdf$$yOpenAccess
000996722 909CO $$ooai:juser.fz-juelich.de:996722$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire
000996722 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)131678$$aForschungszentrum Jülich$$b4$$kFZJ
000996722 9101_ $$0I:(DE-HGF)0$$6P:(DE-Juel1)131678$$a HHU Düsseldorf$$b4
000996722 9101_ $$0I:(DE-HGF)0$$6P:(DE-HGF)0$$a University of Texas Health Science Center at San Antonio, San Antonio, Texas$$b6
000996722 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)POF4-5253$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vDecoding Brain Organization and Dysfunction$$x0
000996722 9141_ $$y2023
000996722 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2022-11-12
000996722 915__ $$0StatID:(DE-HGF)1190$$2StatID$$aDBCoverage$$bBiological Abstracts$$d2022-11-12
000996722 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2022-11-12
000996722 915__ $$0StatID:(DE-HGF)0700$$2StatID$$aFees$$d2022-11-12
000996722 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000996722 915__ $$0StatID:(DE-HGF)0561$$2StatID$$aArticle Processing Charges$$d2022-11-12
000996722 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
000996722 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2023-05-02T08:47:40Z
000996722 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2023-05-02T08:47:40Z
000996722 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Anonymous peer review$$d2023-05-02T08:47:40Z
000996722 915__ $$0StatID:(DE-HGF)0420$$2StatID$$aNationallizenz$$d2023-10-21$$wger
000996722 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bNEUROIMAGE : 2022$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)1050$$2StatID$$aDBCoverage$$bBIOSIS Previews$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)1030$$2StatID$$aDBCoverage$$bCurrent Contents - Life Sciences$$d2023-10-21
000996722 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bNEUROIMAGE : 2022$$d2023-10-21
000996722 920__ $$lyes
000996722 9201_ $$0I:(DE-Juel1)INM-7-20090406$$kINM-7$$lGehirn & Verhalten$$x0
000996722 980__ $$ajournal
000996722 980__ $$aVDB
000996722 980__ $$aUNRESTRICTED
000996722 980__ $$aI:(DE-Juel1)INM-7-20090406
000996722 9801_ $$aFullTexts