Skull-stripping induces shortcut learning in MRI-based Alzheimer’s disease classification
Shortcut learning is induced by skull-stripping applied to T1-weighted MRIs. Explainable deep learning and spectral clustering estimate the bias. Highlights the importance of understanding the dataset, image preprocessing and deep learning model, for interpretation and validation.
Authors
- Christian Langkammer (ORCID: https://orcid.org/0000-0002-7097-9707)
- Stefan Ropele (ORCID: https://orcid.org/0000-0002-5559-768X)
- Christian Tinauer (ORCID: https://orcid.org/0000-0003-4355-3898)
- Maximilian Sackl (ORCID: https://orcid.org/0000-0002-1927-7953)
- Reinhold Schmidt
- Rudolf Stollberger
Institutions
- Medical University of Graz (AT)
- Graz University of Technology (AT)
- BioTechMed-Graz (AT)
Publication Details
- Journal
- Insights into Imaging
- Published
- 2025-12-22
- DOI
- https://doi.org/10.1186/s13244-025-02158-4
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- U.S. Department of Defense
- Alzheimer's Association
- Bristol-Myers Squibb
- Eli Lilly and Company
- Pfizer
- Biogen
- University of Southern California
- Meso Scale Diagnostics
- Alzheimer's Disease Neuroimaging Initiative
- BioClinica
- Novartis Pharmaceuticals Corporation
- Northern California Institute for Research and Education
- Austrian Science Fund
- Eisai
- Servier
- H. Lundbeck A/S
- IXICO
- National Institutes of Health
- Genentech
- Canadian Institutes of Health Research