Table 3

Summary of the studies applying interpretability methods to neuroimaging data which are presented in Subheading 3

StudyData setModalityTaskInterpretability methodSection
Abrol et al. [28]ADNIT1wAD classificationFM visualization, Perturbation3.2, 3.4
Bae et al. [32]ADNIsMRIAD classificationPerturbation3.4
Ball et al. [33]PINGT1wAge predictionWeight visualization, SHAP3.1, 3.5
Biffi et al. [29]ADNIT1wAD classificationFM visualization3.2
Böhle et al. [34]ADNIT1wAD classificationLRP, Guided back-propagation3.3
Burduja et al. [35]RSNACT scanIntracranial Hemorrhage detectionGrad-CAM3.3
Cecotti and Gräser [26]in-houseEEGP300 signals detectionWeight visualization3.1
Dyrba et al. [36]ADNIT1wAD classificationDeconvNet, Deep Taylor decomposition, Gradient ⊙ Input, LRP, Grad-CAM3.3
Eitel and Ritter [37]ADNIT1wAD classificationGradient ⊙ Input, Guided back-propagation, LRP, Perturbation3.3, 3.4
Eitel et al. [38]ADNI, in-houseT1wMultiple Sclerosis detectionGradient ⊙ Input, LRP3.3
Fu et al. [39]CQ500, RSNACT scanDetection of Critical Findings in Head CT scanAttention mechanism3.6
Gutiérrez-Becker and Wachinger [40]ADNIT1wAD classificationPerturbation3.4
Hu et al. [41]ADNI, NIFDT1wAD/CN/FTD classificationGuided back-propagation3.3
Jin et al. [42]ADNI, in-houseT1wAD classificationAttention mechanism3.6
Lee et al. [43]ADNIT1wAD classificationModular transparency3.6
Leming et al. [31]OpenFMRI, ADNI, ABIDE, ABIDE II, ABCD, NDAR ICBM, UK Biobank, 1000FCfMRIAutism classification Sex classification Task vs rest classificationFM visualization, Grad-CAM3.2, 3.3
Magesh et al. [44]PPMISPECTParkinson’s disease detectionLIME3.5
Martinez-Murcia et al. [30]ADNIT1wAD classification Prediction of neuropsychological tests & other clinical variablesFM visualization3.2
Nigri et al. [45]ADNI, AIBLT1wAD classificationPerturbation, Swap test3.4
Oh et al. [27]ADNIT1wAD classificationFM visualization, Standard back-propagation, Perturbation3.2, 3.3, 3.4
Qiu et al. [46]ADNI, AIBL, FHS, NACCT1wAD classificationModular transparency3.6
Ravi et al. [47]ADNIT1wCN/MCI/AD reconstructionModular transparency3.6
Rieke et al. [48]ADNIT1wAD classificationStandard back-propagation, Guided back-propagation, Perturbation, Brain area occlusion3.3, 3.4
Tang et al. [49]UCD-ADC, Brain BankHistologyDetection of amyloid-β pathologyGuided back-propagation, Perturbation3.3, 3.4
Wood et al. [50]ADNIT1wAD classificationModular transparency3.6

Data sets: 1000FC, 1000 Functional Connectomes; ABCD, Adolescent Brain Cognitive Development; ABIDE, Autism Brain Imaging Data Exchange; ADNI, Alzheimer‘s Disease Neuroimaging Initiative; AIBL, Australian Imaging, Biomarkers and Lifestyle; FHS, Framingham Heart Study; ICBM, International Consortium for Brain Mapping; NACC, National Alzheimer’s Coordinating Center; NDAR, National Database for Autism Research; NIFD, frontotemporal lobar degeneration neuroimaging initiative; PING, Pediatric Imaging, Neurocognition and Genetics; PPMI, Parkinson‘s Progres- sion Markers Initiative; RSNA, Radiological Society of North America 2019 Brain CT Hemorrhage data set; UCD-ADC Brain Bank, University of California Davis Alzhei-mer‘s Disease Center Brain Bank

Modalities: CT, computed tomography; EEG, electroencephalography; fMRI, functional magnetic resonance imaging; sMRI, structural magnetic resonance imaging; SPECT, single-photon emission computed tomography; T1w, T1-weighted [magnetic resonance imaging]

Tasks: AD, Alzheimer’s disease; CN, cognitively normal; FTD, frontotemporal dementia; MCI, mild cognitive impairment

Interpretability methods: FM, feature maps; Grad-CAM, gradient-weighted class activation mapping; LIME, local interpretable model-agnostic explanations; LRP, layer-wise relevance; SHAP, SHapley Additive exPlanations

From: Chapter 22, Interpretability of Machine Learning Methods Applied to Neuroimaging

Cover of Machine Learning for Brain Disorders
Machine Learning for Brain Disorders [Internet].
Colliot O, editor.
New York, NY: Humana; 2023.
Copyright 2023, The Author(s)

Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.