% 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{Wiersch:1026172,
author = {Wiersch, Lisa and Friedrich, Patrick and Hamdan, Sami and
Komeyer, Vera and Hoffstaedter, Felix and Patil, Kaustubh R.
and Eickhoff, Simon B. and Weis, Susanne},
title = {{S}ex classification from functional brain connectivity:
{G}eneralization to multiple datasets},
journal = {Human brain mapping},
volume = {45},
number = {6},
issn = {1065-9471},
address = {New York, NY},
publisher = {Wiley-Liss},
reportid = {FZJ-2024-03325},
pages = {e26683},
year = {2024},
note = {Funding informationDeutsche Forschungsgemeinschaft
(DFG),Grant/Award Numbers: 491111487,431549029; National
Institute of MentalHealth, Grant/Award Number:R01-MH074457;
the Helmholtz PortfolioTheme “Supercomputing and Modeling
for theHuman Brain”; European Union's Horizon2020 Research
and Innovation Programme,Grant/Award Number: 945539},
abstract = {Machine learning (ML) approaches are increasingly being
applied to neuroimaging data. Studies in neuroscience
typically have to rely on a limited set of training data
which may impair the generalizability of ML models. However,
it is still unclear which kind of training sample is best
suited to optimize generalization performance. In the
present study, we systematically investigated the
generalization performance of sex classification models
trained on the parcelwise connectivity profile of either
single samples or compound samples of two different sizes.
Generalization performance was quantified in terms of mean
across-sample classification accuracy and spatial
consistency of accurately classifying parcels. Our results
indicate that the generalization performance of parcelwise
classifiers (pwCs) trained on single dataset samples is
dependent on the specific test samples. Certain datasets
seem to "match" in the sense that classifiers trained on a
sample from one dataset achieved a high accuracy when tested
on the respected other one and vice versa. The pwCs trained
on the compound samples demonstrated overall highest
generalization performance for all test samples, including
one derived from a dataset not included in building the
training samples. Thus, our results indicate that both a
large sample size and a heterogeneous data composition of a
training sample have a central role in achieving
generalizable results.Keywords: big data; generalizability;
machine learning; neuroimaging; resting‐state functional
connectivity; sex classification.},
cin = {INM-7},
ddc = {610},
cid = {I:(DE-Juel1)INM-7-20090406},
pnm = {5251 - Multilevel Brain Organization and Variability
(POF4-525) / DFG project 491111487 -
Open-Access-Publikationskosten / 2022 - 2024 /
Forschungszentrum Jülich (OAPKFZJ) (491111487)},
pid = {G:(DE-HGF)POF4-5251 / G:(GEPRIS)491111487},
typ = {PUB:(DE-HGF)16},
pubmed = {38647035},
UT = {WOS:001206019500001},
doi = {10.1002/hbm.26683},
url = {https://juser.fz-juelich.de/record/1026172},
}