Details

Autor(en) / Beteiligte

• Leonardsen, Esten H.
• Peng, Han
• Kaufmann, Tobias
• Agartz, Ingrid
• Andreassen, Ole A.
• Celius, Elisabeth Gulowsen
• Espeseth, Thomas
• Harbo, Hanne F.
• Høgestøl, Einar A.
• Lange, Ann-Marie de
• Marquand, Andre F.
• Vidal-Piñeiro, Didac
• Roe, James M.
• Selbæk, Geir
• Sørensen, Øystein
• Smith, Stephen M.
• Westlye, Lars T.
• Wolfers, Thomas
• Wang, Yunpeng

Titel

Deep neural networks learn general and clinically relevant representations of the ageing brain

Ist Teil von

• NeuroImage (Orlando, Fla.), 2022-08, Vol.256, p.119210-119210, Article 119210

Ort / Verlag

United States: Elsevier Inc

Erscheinungsjahr

2022

Link zum Volltext

Quelle

ScienceDirect

Beschreibungen/Notizen

• •Brain age CNNs achieve state-of-the-art performance in a large, multisite dataset.•A regression-based architecture outperform others in generalizing to new scanners.•Deviations in brain age associate with plausible biological and lifestyle variables.•Encoded representations are better predictors for disorder than the brain age delta. The discrepancy between chronological age and the apparent age of the brain based on neuroimaging data — the brain age delta — has emerged as a reliable marker of brain health. With an increasing wealth of data, approaches to tackle heterogeneity in data acquisition are vital. To this end, we compiled raw structural magnetic resonance images into one of the largest and most diverse datasets assembled (n=53542), and trained convolutional neural networks (CNNs) to predict age. We achieved state-of-the-art performance on unseen data from unknown scanners (n=2553), and showed that higher brain age delta is associated with diabetes, alcohol intake and smoking. Using transfer learning, the intermediate representations learned by our model complemented and partly outperformed brain age delta in predicting common brain disorders. Our work shows we can achieve generalizable and biologically plausible brain age predictions using CNNs trained on heterogeneous datasets, and transfer them to clinical use cases.