Different scaling of linear models and deep learning in UKBiobank brain images versus machine-learning datasets
Open accessPublished version of a preprint
Abstract
Recently, deep learning has unlocked unprecedented success in various domains, especially using images, text, and speech. However, deep learning is only beneficial if the data have nonlinear relationships and if they are exploitable at available sample sizes. We systematically profiled the performance of deep, kernel, and linear models as a function of sample size on UKBiobank brain images against established machine learning references. On MNIST and Zalando Fashion, prediction accuracy consistently improves when escalating from linear models to shallow-nonlinear models, and further improves with deep-nonlinear models. In contrast, using structural or functional brain scans, simple linear models perform on par with more complex, highly parameterized models in age/sex prediction across increasing sample sizes. In sum, linear models keep improving as the sample size approaches ~10,000 subjects. Yet, nonlinearities for predicting common phenotypes from typical brain scans remain largely inaccessible to the examined kernel and deep learning methods.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Cortes, C., & Vapnik, V. (1995) · Machine Learning · doi:10.1007/bf00994018 · Cited by 59,421 · see citing works
A Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707 · see citing works
Kuhn, M., & Johnson, K. (2013) · doi:10.1007/978-1-4614-6849-3 · Cited by 6,788 · see citing works
Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation
Kamnitsas, K., Ledig, C., Newcombe, V. F. J., Simpson, J. P., Kane, A. D., Menon, D. K., Rueckert, D., & Glocker, B. (2017) · Medical Image Analysis · doi:10.1016/j.media.2016.10.004 · Cited by 3,498 · see citing works
Brain tumor segmentation with Deep Neural Networks
Havaei, M., Davy, A., Warde-Farley, D., Biard, A., Courville, A., Bengio, Y., Pal, C., Jodoin, P.-M., & Larochelle, H. (2017) · Medical Image Analysis · doi:10.1016/j.media.2016.05.004 · Cited by 3,278 · see citing works
Functional connectome fingerprinting: Identifying individuals using patterns of brain connectivity
Finn, E. S., Shen, X., Scheinost, D., Rosenberg, M. D., Huang, J., Chun, M. M., Papademetris, X., & Constable, R. T. (2015) · Nature Neuroscience · doi:10.1038/nn.4135 · Cited by 2,747 · see citing works
Multimodal population brain imaging in the UK Biobank prospective epidemiological study
Miller, K. L., Alfaro-Almagro, F., Bangerter, N. K., Thomas, D. L., Yacoub, E., Xu, J., Bartsch, A. J., Jbabdi, S., Sotiropoulos, S. N., Andersson, J. L. R., et al. (2016) · Nature Neuroscience · doi:10.1038/nn.4393 · Cited by 2,020 · see citing works
Bag of Tricks for Image Classification with Convolutional Neural Networks
He, T., Zhang, Z., Zhang, H., Zhang, Z., Xie, J., & Li, M. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00065 · Cited by 1,680 · see citing works
Kather, J. N., Pearson, A. T., Halama, N., Jäger, D., Krause, J., Loosen, S. H., Marx, A., Boor, P., Tacke, F., Neumann, U. P., et al. (2019) · Nature Medicine · doi:10.1038/s41591-019-0462-y · Cited by 1,458 · see citing works
Self-citationEfficiency of Functional Brain Networks and Intellectual Performance
van den Heuvel, M. P., Stam, C. J., Kahn, R. S., & Hulshoff Pol, H. E. (2009) · Journal of Neuroscience · doi:10.1523/jneurosci.1443-09.2009 · Cited by 1,078 · see citing works
Cole, J. H., Poudel, R. P. K., Tsagkrasoulis, D., Caan, M. W. A., Steves, C., Spector, T. D., & Montana, G. (2017) · NeuroImage · doi:10.1016/j.neuroimage.2017.07.059 · Cited by 872 · cited 2× · see citing works
“Second, deep learning approaches have repeatedly been found to perform worse or indistinguishably well compared to simpler baseline models when predicting demographic or behavioral phenotypes 13,40 .”
BackgroundResultUnderstanding heterogeneity in clinical cohorts using normative models: Beyond case-control studies
Marquand, A. F., Rezek, I., Buitelaar, J., & Beckmann, C. F. (2016) · Biological Psychiatry · doi:10.1016/j.biopsych.2015.12.023 · Cited by 683 · see citing works
Quicksilver: Fast predictive image registration – A deep learning approach
Yang, X., Kwitt, R., Styner, M., & Niethammer, M. (2017) · NeuroImage · doi:10.1016/j.neuroimage.2017.07.008 · Cited by 630 · cited 1× · see citing works
“However, a few successful exceptions have shown advantages of deep learning applied to neuroimaging data over conventional approaches, such as for the speci fi c goals of image segmentation 18,34 – 37 and image registration 38,39 .”
BackgroundHe, T., Kong, R., Holmes, A. J., Nguyen, M., Sabuncu, M. R., Eickhoff, S. B., Bzdok, D., Feng, J., & Yeo, B. T. T. (2020) · NeuroImage · doi:10.1016/j.neuroimage.2019.116276 · Cited by 271 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Heterogeneous aging across multiple organ systems and prediction of chronic disease and mortality
Tian, Y. E., Cropley, V., Maier, A. B., Lautenschlager, N. T., Breakspear, M., & Zalesky, A. (2023) · Nature Medicine · doi:10.1038/s41591-023-02296-6 · Cited by 618 · see citing works
A Synergistic Core for Human Brain Evolution and Cognition
Luppi, A. I., Mediano, P. A. M., Rosas, F. E., Holland, N., Fryer, T. D., O’Brien, J. T., Rowe, J. B., Menon, D. K., Bor, D., & Stamatakis, E. A. (2022) · Nature Neuroscience · doi:10.1038/s41593-022-01070-0 · Cited by 314 · see citing works
Zhou, Q., Chen, Z.-H., Cao, Y.-H., & Peng, S. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00524-2 · Cited by 126 · see citing works
Teghipco, A., Newman-Norlund, R., Fridriksson, J., Rorden, C., & Bonilha, L. (2024) · Communications Medicine · doi:10.1038/s43856-024-00541-8 · Cited by 11 · see citing works
- Authors
- Schulz, M.-A., Yeo, B. T. T., Vogelstein, J. T., Mourao-Miranada, J., Kather, J. N., Kording, K., Richards, B., & Bzdok, D.
- Affiliations
- National University of Singapore (SG)
- Johns Hopkins University (US)
- German Cancer Research Center (DE)
- University of Pennsylvania (US)
- Canadian Institute for Advanced Research (CA)
- Montreal Neurological Institute and Hospital (CA)
- Corresponding author
- Bzdok, Danilo
- ORCID
- Venue
- Nature Communications
- Published
- 2020-08-25
- Volume
- 11
- Issue
- 1
- Pages
- 4238-4238
- ISSN
- 2041-1723 (ISSN-L)
- PMC
- PMC7447816
- OpenAlex
- W3080221164
- Cited by
- 315
- References
- 15
- Citations by year
- 2026: 29 · 2025: 55 · 2024: 57 · 2023: 61 · 2022: 53 · 2021: 45 · 2020: 15
- FWCI
- 16.95
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Nature Communications (publishedVersion, cc-by)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, other-oa)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- UCL Discovery (University College London) (submittedVersion, cc-by)
- National University of Singapore (submittedVersion, cc-by)
- HAL (submittedVersion, other-oa)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- RWTH Publications (RWTH Aachen) (submittedVersion, cc-by)
- MPG.PuRe (Max Planck Society) (submittedVersion)
- RWTH Publications (RWTH Aachen)
- Funders
- Canadian Institute for Advanced Research
- Wellcome Trust (102845/Z/13/Z)
- National Research Foundation
- National Research Foundation Singapore
- Deutsche Forschungsgemeinschaft (DFG, BZ2/2-1, BZ2/4‐1, BZ2/2‐1, BZ2/2-1, BZ2/3-1, IRTG-2150, DFG, BZ2/2-1, BZ2/3-1, and BZ2/4-1; International Research Training Group IRTG2150, BZ2/3-1)
- RWTH Aachen University
- Canada First Research Excellence Fund
- National Institutes of Health
- National Institute on Aging (R01AG068563)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Life Sciences › Neuroscience › Cognitive Neuroscience › Functional Brain Connectivity Studies
- Keywords
- deep learning, linear models, kernel methods, sample size, neuroimaging, UK Biobank, age prediction, sex prediction, nonlinear relationship, functional MRI, structural MRI, MNIST, model performance comparison
- MeSH terms
- Machine Learning, Deep Learning, Brain, United Kingdom, Humans, Phenotype, Linear Models, Biological Specimen Banks, Sample Size, Neuroimaging
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
