Performance reserves in brain-imaging-based phenotype prediction
Autor(in); Institution
Schulz, Marc-Andre; Charité – Universitätsmedizin Berlin (corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health), Department of Psychiatry and Psychotherapy, Berlin, GERMANY; Bernstein Center for Computational Neuroscience, Berlin, GERMANY
Bzdok, Danilo; McConnell Brain Imaging Centre (BIC), Montreal Neurological Institute (MNI), Faculty of Medicine, McGill University, Montreal, QC, CANADA; Department of Biomedical Engineering, Faculty of Medicine, McGill University, Montreal, QC, CANADA; Mila - Quebec Artificial Intelligence Institute, Montreal, QC, CANADA
Haufe, Stefan; 8.4, Mathematische Modellierung und Datenanalyse, PTB-Berlin; Bernstein Center for Computational Neuroscience, Berlin, GERMANY; Computer Science Department, Technische Universität Berlin, Berlin, GERMANY
Haynes, John-Dylan; Bernstein Center for Computational Neuroscience, Berlin, GERMANY; Charité – Universitätsmedizin Berlin (corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health), Department of Neurology, Berlin Center for Advanced Neuroimaging, Berlin, GERMANY
Ritter, Kerstin; Charité – Universitätsmedizin Berlin (corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health), Department of Neurology, Berlin Center for Advanced Neuroimaging, Berlin, GERMANY; Bernstein Center for Computational Neuroscience, Berlin, GERMANY
This study examines the impact of sample size on predicting cognitive and mental health phenotypes from brain imaging via machine learning. Our analysis shows a 3- to 9-fold improvement in prediction performance when sample size increases from 1,000 to 1 M participants. However, despite this increase, the data suggest that prediction accuracy remains worryingly low and far from fully exploiting the predictive potential of brain imaging data. Additionally, we find that integrating multiple imaging modalities boosts prediction accuracy, often equivalent to doubling the sample size. Interestingly, the most informative imaging modality often varied with increasing sample size, emphasizing the need to consider multiple modalities. Despite significant performance reserves for phenotype prediction, achieving substantial improvements may necessitate prohibitively large sample sizes, thus casting doubt on the practical or clinical utility of machine learning in some areas of neuroimaging.
Kostenfreier Zugang
Open Access Gold
Themenbereich der Metrologie
Metrologie in der Medizin
Zitierung
Schulz, M.-A., Bzdok, D., Haufe, S., Haynes, J.-D., & Ritter, K. (2024). Performance reserves in brain-imaging-based phenotype prediction. Cell Reports, 43(1), 1–14. https://doi.org/10.1016/j.celrep.2023.113597
