Gewählte Publikation:

SHR Neuro Krebs Kardio Lipid Stoffw Microb

Tinauer, C; Sackl, M; Stollberger, R; Schmidt, R; Ropele, S; Langkammer, C.
Skull-stripping induces shortcut learning in MRI-based Alzheimer's disease classification
INSIGHTS IMAGING. 2025; 16(1): 283 Doi: 10.1186/s13244-025-02158-4 [OPEN ACCESS]
Web of Science PubMed FullText FullText_MUG

 

Führende Autor*innen der Med Uni Graz

Langkammer Christian

Tinauer Christian Gerhard

Co-Autor*innen der Med Uni Graz

Ropele Stefan

Sackl Maximilian

Schmidt Reinhold

Stollberger Rudolf

Altmetrics:

Dimensions Citations:

Plum Analytics:

Scite (citation analytics):

Abstract:

ObjectivesHigh classification accuracy of Alzheimer's disease (AD) from structural MRI has been achieved using deep neural networks, yet the specific image features contributing to these decisions remain unclear. In this study, the contributions of T1-weighted (T1w) gray-white matter texture, volumetric information, and preprocessing-particularly skull-stripping-were systematically assessed.Materials and methodsA dataset of 990 matched T1w MRIs from AD patients and cognitively normal controls from the ADNI database was used. Preprocessing was varied through skull-stripping and intensity binarization to isolate texture and shape contributions. A 3D convolutional neural network was trained on each configuration, and classification performance was compared using exact McNemar tests with discrete Bonferroni-Holm correction. Feature relevance was analyzed using Layer-wise Relevance Propagation, image similarity metrics, and spectral clustering of relevance maps.ResultsDespite substantial differences in image content, classification accuracy, sensitivity, and specificity remained stable across preprocessing conditions. Models trained on binarized images preserved performance, indicating minimal reliance on gray-white matter texture. Instead, volumetric features-particularly brain contours introduced through skull-stripping-were consistently used by the models.ConclusionThis behavior reflects a shortcut learning phenomenon, where preprocessing artifacts act as potentially unintended cues. The resulting Clever Hans effect emphasizes the critical importance of interpretability tools to reveal hidden biases and to ensure robust and trustworthy deep learning in medical imaging.Critical relevance statementWe investigated the mechanisms underlying deep learning-based disease classification using a widely utilized Alzheimer's disease dataset, and our findings reveal a reliance on features induced through skull-stripping, highlighting the need for careful preprocessing to ensure clinically relevant and interpretable models.Key PointsShortcut 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.

Find related publications in this database (Keywords)

Shortcut learning

Preprocessing bias

Alzheimer's disease

Explainable deep learning

© Med Uni Graz • Impressum