% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@ARTICLE{Singh:282290,
author = {Singh, Devesh and Brima, Yusuf and Levin, Fedor and Becker,
Martin and Hiller, Bjarne and Hermann, Andreas and
Villar-Munoz, Irene and Beichert, Lukas and Bernhardt,
Alexander and Buerger, Katharina and Butryn, Michaela and
Dechent, Peter and Düzel, Emrah and Ewers, Michael and
Fliessbach, Klaus and Freiesleben, Silka D and Glanz, Wenzel
and Hetzer, Stefan and Janowitz, Daniel and Görß, Doreen
and Kilimann, Ingo and Kimmich, Okka and Laske, Christoph
and Levin, Johannes and Lohse, Andrea and Luesebrink, Falk
and Munk, Matthias and Perneczky, Robert and Peters, Oliver
and Preis, Lukas and Priller, Josef and Prudlo, Johannes and
Prychynenko, Diana and Rauchmann, Boris Stephan and
Rostamzadeh, Ayda and Roy-Kluth, Nina and Scheffler, Klaus
and Schneider, Anja and Droste Zu Senden, Louise and Schott,
Björn H and Spottke, Annika and Synofzik, Matthis and
Wiltfang, Jens and Jessen, Frank and Weber, Marc-André and
Teipel, Stefan J and Dyrba, Martin},
title = {{A}n unsupervised {XAI} framework for dementia detection
with context enrichment.},
journal = {Scientific reports},
volume = {15},
number = {1},
issn = {2045-2322},
address = {[London]},
publisher = {Springer Nature},
reportid = {DZNE-2025-01261},
pages = {39554},
year = {2025},
abstract = {Explainable Artificial Intelligence (XAI) methods enhance
the diagnostic efficiency of clinical decision support
systems by making the predictions of a convolutional neural
network's (CNN) on brain imaging more transparent and
trustworthy. However, their clinical adoption is limited due
to limited validation of the explanation quality. Our study
introduces a framework that evaluates XAI methods by
integrating neuroanatomical morphological features with
CNN-generated relevance maps for disease classification. We
trained a CNN using brain MRI scans from six cohorts: ADNI,
AIBL, DELCODE, DESCRIBE, EDSD, and NIFD (N = 3253),
including participants that were cognitively normal, with
amnestic mild cognitive impairment, dementia due to
Alzheimer's disease and frontotemporal dementia. Clustering
analysis benchmarked different explanation space
configurations by using morphological features as
proxy-ground truth. We implemented three post-hoc
explanations methods: (i) by simplifying model decisions,
(ii) explanation-by-example, and (iii) textual explanations.
A qualitative evaluation by clinicians (N = 6) was performed
to assess their clinical validity. Clustering performance
improved in morphology enriched explanation spaces,
improving both homogeneity and completeness of the clusters.
Post hoc explanations by model simplification largely
delineated converters and stable participants, while
explanation-by-example presented possible cognition
trajectories. Textual explanations gave rule-based
summarization of pathological findings. Clinicians'
qualitative evaluation highlighted challenges and
opportunities of XAI for different clinical applications.
Our study refines XAI explanation spaces and applies various
approaches for generating explanations. Within the context
of AI-based decision support system in dementia research we
found the explanations methods to be promising towards
enhancing diagnostic efficiency, backed up by the clinical
assessments.},
keywords = {Humans / Dementia: diagnosis / Dementia: diagnostic imaging
/ Magnetic Resonance Imaging: methods / Male / Female / Aged
/ Cognitive Dysfunction: diagnostic imaging / Cognitive
Dysfunction: diagnosis / Alzheimer Disease: diagnostic
imaging / Alzheimer Disease: diagnosis / Neural Networks,
Computer / Artificial Intelligence / Brain: diagnostic
imaging / Brain: pathology / Aged, 80 and over /
Alzheimer’s disease (Other) / Brain volumetry (Other) /
Explainable artificial intelligence (XAI) (Other) /
Frontotemporal dementia (Other) / Magnetic resonance imaging
(Other) / Neurodegenerative diseases (Other) / Qualitative
evaluation (Other)},
cin = {AG Teipel / AG Hermann / Clinical Research (Munich) / AG
Düzel / Patient Studies (Bonn) / AG Peters / Clinical
Research (Bonn) / AG Gasser / AG Dichgans / AG Priller / AG
Spottke / AG Schneider / AG Fischer / AG Wiltfang / AG
Jessen},
ddc = {600},
cid = {I:(DE-2719)1510100 / I:(DE-2719)1511100 /
I:(DE-2719)1111015 / I:(DE-2719)5000006 / I:(DE-2719)1011101
/ I:(DE-2719)5000000 / I:(DE-2719)1011001 /
I:(DE-2719)1210000 / I:(DE-2719)5000022 / I:(DE-2719)5000007
/ I:(DE-2719)1011103 / I:(DE-2719)1011305 /
I:(DE-2719)1410002 / I:(DE-2719)1410006 /
I:(DE-2719)1011102},
pnm = {353 - Clinical and Health Care Research (POF4-353) / 352 -
Disease Mechanisms (POF4-352)},
pid = {G:(DE-HGF)POF4-353 / G:(DE-HGF)POF4-352},
typ = {PUB:(DE-HGF)16},
pubmed = {pmid:41224940},
pmc = {pmc:PMC12612133},
doi = {10.1038/s41598-025-26227-2},
url = {https://pub.dzne.de/record/282290},
}