Pith. sign in

REVIEW 3 major objections 5 minor 71 references

Three-dimensional end-to-end deep learning for brain MRI analysis

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A simple 3D convolutional network outperforms DenseNet and Swin Transformer at predicting age and sex from brain MRI, and generalizes better across cohorts.

desk verdict A useful multi-cohort benchmark for 3D T1 MRI age/sex prediction; the sex result is solid, but the age-superiority claim hinges on an unspecified bias-correction fit that needs clarification. read the letter →

arxiv 2506.23916 v1 pith:R5DJ32SV submitted 2025-06-30 cs.CV

Radhika Juglan , Marta Ligero , Zunamys I. Carrero , Asier Rabasco , Tim Lenz , Leo Misera , Gregory Patrick Veldhuizen , Paul Kuntke
show 4 more authors
Hagen H. Kitzler Sven Nebelung Daniel Truhn Jakob Nikolas Kather
This is my paper · ORCID
classification cs.CV
keywords deeplearningbrainMRIagepredictionsexclassificationexternalvalidationgeneralizabilityconvolutionalneuralnetworkSwinTransformer
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper tests whether heavier, more expressive 3D deep-learning architectures are better than a lightweight convolutional network at reading demographic traits from T1-weighted brain MRI. It claims they are not: the Simple Fully Connected Network (SFCN) beats DenseNet and Swin Transformer on internal and external cohorts for both age regression and sex classification. The paper wants to establish that simpler convolutional models generalize better across imaging sites and populations, and that external validation should be standard before trusting such models. This matters because brain-age and sex predictions underpin many proposed clinical biomarkers, and choosing the right default architecture affects how those biomarkers behave in new data.

What carries the argument

The load-bearing object is SFCN (Simple Fully Connected Network), a lightweight 3D convolutional architecture: a stack of 3D convolution, batch-normalization, and ReLU layers, followed by global average pooling and a fully connected head for a scalar age or binary sex output. It is compared against Monai's DenseNet121 and a 3D Swin Transformer under identical fixed training hyperparameters (batch size 4, learning rate $10^{{-5}}$, early stopping), on brain images preprocessed via registration to standard space, brain extraction, intensity normalization, and center-cropping to $180^{3}$ voxels. The comparison protocol, including a held-out center for internal testing and three external cohorts, is what carries the generalizability claim; the bias-correction step in age evaluation is what carries the reported MAE values.

What would settle it

Re-run the age prediction with bias-correction coefficients estimated only on the training or validation split and re-measure external MAE; if SFCN's advantage over DenseNet and Swin shrinks or disappears, the headline generalizability gap is an artifact of post-hoc label fitting. A separate check would retrain DenseNet and Swin with their own tuned learning rates and batch sizes and see whether they match SFCN on external cohorts.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is a consistent ordering of architectures: across four cohorts, SFCN, a shallow 3D convolutional network with batch normalization and ReLU, achieved an AUC of 1.00 [1.00-1.00] for sex classification on the internal UK test set and 0.85-0.91 on three external sets, while DenseNet lagged at 0.82-0.87 and Swin Transformer dropped to 0.67-0.80 outside the training distribution. For age prediction, SFCN reached a mean absolute error of 2.66 years (r=0.89) internally and 4.98-5.81 years (r=0.55-0.70) externally; both DenseNet and Swin produced MAEs near 6-7 years with near-zero correlation on external data. Statistical tests with Bonferroni correction support SFCN over Swin Transformer, and the paper argues the attention-based model is overparameterized for the available training data. The authors additionally find no substantial age or sex bias in SFCN's performance, and they observe that its attention maps and volume correlations align with known neuroanatomy.

Load-bearing premise

The reported age errors are trustworthy only if the linear correction that shifts predicted ages toward true ages is fitted on data that is separate from the test set where the errors are measured, and the paper does not state which set is used for that fit.

Editorial extensions

If this is right

  • For sex classification, architecture choice barely matters within the training distribution (all models hit AUC 1.00), but it determines robustness cross-cohort: the simple CNN stays at 0.85-0.91 where Swin drops to 0.67.
  • Age prediction from T1 MRI degrades by roughly 2-3 years of MAE across imaging sites even for the best model, so external-site error ranges should be reported alongside internal results in future brain-age studies.
  • Attention-based architectures like Swin may need substantially more training data or per-model hyperparameter tuning before they can beat a well-designed lightweight CNN on 3D volumetric medical images.
  • The consistency between model-attention heatmaps and gray-matter volume correlations suggests SFCN is using neuroanatomical features, not image artifacts, to make its predictions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An implication the authors leave implicit: because the same fixed batch size and learning rate were used for all three architectures, the comparison likely penalizes the larger models more than their capacity deserves; a tuned Swin could close part of the age-prediction gap.
  • A testable extension is to run the same three architectures on downstream clinical targets, such as brain-age delta in multiple sclerosis, to see whether SFCN's simplicity advantage persists beyond healthy demographic prediction.
  • Because the code and external cohorts are public, these data could serve as a reusable out-of-distribution benchmark for future 3D brain MRI models, letting the field track whether new architectures actually generalize rather than merely overfit the UK cohort.
  • The volume-correlation check (predictions matching ground-truth correlations with regional volumes) offers a cheap sanity test for any new brain-MRI model: if a new architecture's prediction gradients do not correlate with the same regions, it is likely exploiting site-specific artifacts.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper compares three 3D deep learning architectures—SFCN, DenseNet121, and Swin Transformer—for sex classification and age regression from T1-weighted brain MRI. Models are trained on 34,918 UK Biobank participants from three centers and tested on the held-out Newcastle center (12,472 participants) and on three external cohorts (DLBS, PPMI, IXI). The paper reports SFCN achieving essentially perfect internal sex AUC and 0.85-0.91 external AUC, and lower age MAE than DenseNet and Swin Transformer after a linear bias-correction step, with DeLong and Wilcoxon tests, subgroup analyses, and explainability analyses. The manuscript concludes that simpler convolutional networks generalize better across datasets than denser or attention-based architectures.

Significance. If the results hold, the paper is a useful contribution to the architectural-choice debate in medical image analysis: it provides large-scale UK Biobank training, three external validation cohorts, confidence intervals, pairwise statistical tests, open code (https://github.com/jrad9921/RadBrainDL), and reproducible preprocessing using standard tools. The sex-classification results are strong and consistently reported. The main caveat is that the age-prediction half of the central claim depends on a bias-correction procedure whose fitting set is not specified; until that is clarified, the age-based superiority and generalizability conclusion is not fully established.

major comments (3)
  1. [Methods, 'Bias Correction'; Results, age prediction; Table 2] The bias-correction step is not sufficiently specified, and this is load-bearing for the age-prediction half of the central claim. The Methods state that for each model, predicted age values were regressed against their corresponding true age labels and that the resulting coefficients were applied to adjust predictions, after which MAE and Pearson r were recomputed; they do not state whether the regression is fitted on training/validation predictions or on the same test set whose labels are later used to compute MAE and r. If the latter, then the age MAEs in Table 2 and in the Results (e.g., SFCN 2.66 years in UKB, 4.97-5.80 years external) and the pairwise Wilcoxon comparisons are not held-out estimates, because each evaluation would absorb test-label information through the fitted slope and intercept. Please specify the fitting set explicitly; if the correction was fitted on test labels, the age results and the model comparisons must be recomputed with a correction fitted only on training/validation predictions, or using a properly nested procedure. If the authors followed Smith et al. (ref 62), where the correction is normally fitted on training data, a sentence confirming this and stating the resulting coefficients would resolve the concern.
  2. [Methods, 'Training Procedures'; Discussion; Abstract] The architecture comparison uses a single learning rate and batch size for all models, and this weakens the reported superiority of SFCN. The Methods state that a batch size of 4 and a learning rate of 10e-05 were fixed across all models, while also stating that hyperparameters were tuned on the validation set; these statements are not reconciled, and no per-model tuning results are reported. The Discussion acknowledges the fixed-parameter limitation, but the abstract and conclusion present the finding unconditionally ('simpler convolutional networks outperform...'). Because DenseNet and Swin Transformer may require different learning rates or batch sizes, the head-to-head comparison cannot rule out that the gap is partly an artifact of the common training protocol. Please provide per-model tuning results or explicitly qualify the central claim as holding under a fixed common protocol.
  3. [Results, age prediction; Table 2; Table 1] The reported age results contain internal inconsistencies that must be reconciled. The Results text gives UKB SwinTransformer MAE=4.86 and r=0.60, while Table 2 reports MAE=4.22 (4.16-4.28) and r=0.71 (0.70-0.72); similarly, DLBS is listed as n=132 in Table 1 but as n=108 in the Results section. Because the age comparisons are central to the paper's conclusion, the text, tables, and figures should be checked and the final numbers used consistently.
minor comments (5)
  1. [Methods, 'Training Procedures'] The expression '10e-05' should be written as 1e-5 (or 10^-5) if that is the intended learning rate; as written, 10e-05 equals 1e-4, which is a different value.
  2. [Abstract; Results, 'Minimal bias...'] The Abstract states that 'No significant demographic subgroup biases were detected', but the Results do not report statistical comparisons for the subgroup differences; the age/sex subgroup MAEs in Fig. 3 are presented descriptively. Please either add appropriate tests or soften the claim.
  3. [Results, 'DL reveals consistent task-specific attention patterns'; Fig. 4 caption] The text refers to 'attention heatmaps' for SFCN, while the Fig. 4 caption correctly identifies the method as Grad-CAM; these are gradient-based saliency maps, not attention weights, so the terminology should be made consistent.
  4. [Methods, 'Training Procedures'] The sample sizes of the training and validation subsets after the 2:1 split are not reported; please state them for reproducibility.
  5. [Abstract] The phrase 'Bonferroni corrections confirmed SFCN's superiority over Swin Transformer across most cohorts (p<0.017, for three comparisons)' is imprecise: p<0.017 is the Bonferroni-corrected threshold for three pairwise comparisons, and the Results report different significance patterns for sex and age; the Abstract should reflect the Results accurately.

Circularity Check

1 steps flagged · score 4.0 of 10

Age-prediction MAEs are computed after a linear bias correction whose fitting set is unspecified; if the correction is fit on the same test labels used for evaluation, the age-based half of the central claim is partly in-sample.

  1. fitted input called prediction [Methods, 'Bias Correction' section]
    "For each model, predicted age values were regressed against their corresponding true age labels. A linear regression model was fitted to these data points to establish a relationship between the predicted and actual ages. The resulting regression coefficients were then applied to adjust the original predictions, producing bias-corrected age estimates. After correction, MAE was calculated to quantify predictive accuracy, and Pearson's correlation coefficient (r) was computed to evaluate the linear relationship between the corrected predictions and true labels."

    The paper does not state whether this regression is fit on training/validation predictions or on the same test set whose true labels are subsequently used to compute the reported MAE and Pearson r. If the fit is on the test set, the bias-corrected predictions are in-sample linear adjustments: the regression coefficients are chosen to minimize residual error against the very labels used for evaluation. The reported age MAEs (UKB 2.66; DLBS 4.98; PPMI 5.13; IXI 5.81) and the pairwise Wilcoxon comparisons are then not demonstrated to be held-out estimates, so the age-based pillar of the abstract's claim would reduce partly to a fitted quantity rather than an independent prediction.

full rationale

There is no mathematical derivation chain in this paper, so the main circularity risk is empirical. The sex classification results, the heatmap explainability, and the volume-correlation analyses are self-contained benchmark outputs and do not reduce to their inputs by construction. The bias-correction step is the one place where a fitted parameter feeds directly into a reported prediction metric: predicted ages are regressed against true age labels, the coefficients are applied, and MAE/r are then recomputed. Because the fitting set is not specified, the reported age metrics cannot be confirmed as fully held-out; if the correction is fit on each test set, the age comparison is partially in-sample and the SFCN-versus-others age gap is not a clean external generalization result. This is a genuine ambiguity rather than a demonstrated equivalence, and the sex-classification half of the central claim remains independently supported, so the circularity score is moderate rather than high. The fixed learning-rate and batch-size limitation is explicitly acknowledged in the Discussion, and the paper's self-citations (e.g., to prior transformer work by overlapping authors) are background context rather than load-bearing derivations. The numerical inconsistencies (UKB Swin MAE 4.22 in Table 2 vs 4.86 in the text; DLBS n=132 in Table 1 vs 108 in Results) further reduce confidence in the age numbers but are not themselves circularity.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The paper is an empirical benchmark rather than a derivation, so the ledger contains experimental choices and domain assumptions rather than mathematical axioms. The main risk items are the unspecified bias-correction fitting set, fixed hyperparameters, and the assumption that preprocessed T1 images from different scanners carry comparable biological signal.

free parameters (4)
  • Bias-correction linear regression coefficients (slope and intercept, per model) = not reported
    Fitted by regressing predicted age against true age in the Bias Correction step; if fitted on the test set, they are test-label-fitted parameters used to compute reported MAE.
  • Learning rate = 10e-05
    Single value fixed across SFCN, DenseNet, and Swin Transformer; the architecture comparison depends on this choice being fair.
  • Batch size = 4
    Fixed across all models; affects convergence and final performance.
  • Age range restriction = 40-70 years
    External cohorts are restricted to the UKB age range; the generalization claim does not cover younger or older adults.
assumptions (4)
  • domain assumption T1-weighted brain MRI after preprocessing retains enough age- and sex-related signal for prediction
    The predictive setup assumes informative signal survives registration, brain extraction, cropping, and z-normalization.
  • domain assumption External cohorts are comparable enough to UKB after MNI registration for held-out evaluation
    Generalizability claims rest on the premise that scanner and site differences do not dominate the biological signal; the authors note performance drops.
  • standard math DeLong and Wilcoxon signed-rank tests with Bonferroni correction are valid for these paired comparisons
    Statistical significance claims rely on standard paired ROC and regression comparison tests, which assume appropriate pairing and subject independence.
  • domain assumption Bias correction fitted on one set can be applied to another without leaking labels
    Methods do not specify the fitting set; valid age MAE requires the correction to be fit in a training-only or label-free way.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Three-dimensional end-to-end deep learning for brain MRI analysis." pith.science (2026). https://pith.science/paper/R5DJ32SV

@misc{pith2026250623916,
  author       = {Pith},
  title        = {Pith review of: Three-dimensional end-to-end deep learning for brain MRI analysis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R5DJ32SV}},
  note         = {Machine review of arXiv:2506.23916}
}
read the original abstract

Deep learning (DL) methods are increasingly outperforming classical approaches in brain imaging, yet their generalizability across diverse imaging cohorts remains inadequately assessed. As age and sex are key neurobiological markers in clinical neuroscience, influencing brain structure and disease risk, this study evaluates three of the existing three-dimensional architectures, namely Simple Fully Connected Network (SFCN), DenseNet, and Shifted Window (Swin) Transformers, for age and sex prediction using T1-weighted MRI from four independent cohorts: UK Biobank (UKB, n=47,390), Dallas Lifespan Brain Study (DLBS, n=132), Parkinson's Progression Markers Initiative (PPMI, n=108 healthy controls), and Information eXtraction from Images (IXI, n=319). We found that SFCN consistently outperformed more complex architectures with AUC of 1.00 [1.00-1.00] in UKB (internal test set) and 0.85-0.91 in external test sets for sex classification. For the age prediction task, SFCN demonstrated a mean absolute error (MAE) of 2.66 (r=0.89) in UKB and 4.98-5.81 (r=0.55-0.70) across external datasets. Pairwise DeLong and Wilcoxon signed-rank tests with Bonferroni corrections confirmed SFCN's superiority over Swin Transformer across most cohorts (p<0.017, for three comparisons). Explainability analysis further demonstrates the regional consistency of model attention across cohorts and specific to each task. Our findings reveal that simpler convolutional networks outperform the denser and more complex attention-based DL architectures in brain image analysis by demonstrating better generalizability across different datasets.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

71 extracted references · 68 canonical work pages

  1. [1]

    & O’Brien, J

    Chouliaras, L. & O’Brien, J. T. The use of neuroimaging techniques in the early and differential diagnosis of dementia. Mol. Psychiatry 28 , 4084–4097 (2023)

    work page 2023

  2. [2]

    H., Tini, K

    Tedyanto, E. H., Tini, K. & Pramana, N. A. K. Magnetic Resonance Imaging in acute ischemic stroke. Cureus 14 , e27224 (2022)

    work page 2022

  3. [3]

    B., Fox, N

    Frisoni, G. B., Fox, N. C., Jack, C. R., Jr, Scheltens, P. & Thompson, P. M. The clinical use of structural MRI in Alzheimer disease. Nat. Rev. Neurol. 6 , 67–77 (2010)

    work page 2010

  4. [4]

    Vemuri, P. et al. Antemortem differential diagnosis of dementia pathology using structural MRI: Differential-STAND. Neuroimage 55 , 522–531 (2011)

    work page 2011

  5. [5]

    & Politis, M

    Yousaf, T., Dervenoulas, G., Valkimadi, P.-E. & Politis, M. Neuroimaging in Lewy body dementia. J. Neurol. 266 , 1–26 (2019)

    work page 2019

  6. [6]

    Whitwell, J. L. et al. MRI correlates of neurofibrillary tangle pathology at autopsy: a voxel-based morphometry study. Neurology 71 , 743–749 (2008)

    work page 2008

  7. [7]

    R., Jr et al

    Jack, C. R., Jr et al. Antemortem MRI findings correlate with hippocampal neuropathology in typical aging and dementia. Neurology 58 , 750–757 (2002)

    work page 2002

  8. [8]

    Harper, L. et al. Patterns of atrophy in pathologically confirmed dementias: a voxelwise analysis. J. Neurol. Neurosurg. Psychiatry 88 , 908–916 (2017)

    work page 2017

Show all 71 references
  1. [9]

    Cash, D. M. et al. Patterns of gray matter atrophy in genetic frontotemporal dementia: results from the GENFI study. Neurobiol. Aging 62 , 191–196 (2018)

    2018

  2. [10]

    Ridha, B. H. et al. Tracking atrophy progression in familial Alzheimer’s disease: a serial MRI study. Lancet Neurol. 5 , 828–834 (2006)

    2006

  3. [11]

    Hill, D. L. G. et al. Coalition Against Major Diseases/European Medicines Agency biomarker qualification of hippocampal volume for enrichment of clinical trials in predementia stages of Alzheimer’s disease. Alzheimers. Dement. 10 , 421–9.e3 (2014)

    2014

  4. [12]

    Rohrer, J. D. et al. Presymptomatic cognitive and neuroanatomical changes in genetic frontotemporal 25 dementia in the Genetic Frontotemporal dementia Initiative (GENFI) study: a cross-sectional analysis. Lancet Neurol. 14 , 253–262 (2015)

    2015

  5. [13]

    Nicastro, N. et al. Cortical complexity analyses and their cognitive correlate in Alzheimer’s disease and frontotemporal dementia. J. Alzheimers. Dis. 76 , 331–340 (2020)

    2020

  6. [14]

    Du, A.-T. et al. Different regional patterns of cortical thinning in Alzheimer’s disease and frontotemporal dementia. Brain 130 , 1159–1166 (2007)

    2007

  7. [15]

    Choi, M. et al. Comparison of neurodegenerative types using different brain MRI analysis metrics in older adults with normal cognition, mild cognitive impairment, and Alzheimer’s dementia. PLoS One 14 , e0220739 (2019)

    2019

  8. [16]

    Kälin, A. M. et al. Subcortical shape changes, hippocampal atrophy and cortical thinning in future Alzheimer’s disease patients. Front. Aging Neurosci. 9 , 38 (2017)

    2017

  9. [17]

    Ibarretxe-Bilbao, N. et al. Progression of cortical thinning in early Parkinson’s disease: Cortical Thinning in Early PD. Mov. Disord. 27 , 1746–1753 (2012)

    2012

  10. [18]

    & Agosta, F

    Filippi, M. & Agosta, F. MRI of non-Alzheimer’s dementia: current and emerging knowledge. Curr. Opin. Neurol. 31 , 405–414 (2018)

    2018

  11. [19]

    Ferretti, M. T. et al. Sex and gender differences in Alzheimer’s disease: current challenges and implications for clinical practice: Position paper of the Dementia and Cognitive Disorders Panel of the European Academy of Neurology: Position paper of the Dementia and Cognitive ...

    2020

  12. [20]

    R., Irvine, K

    Laws, K. R., Irvine, K. & Gale, T. M. Sex differences in cognitive impairment in Alzheimer’s disease. World J. Psychiatry 6 , 54–65 (2016)

    2016

  13. [21]

    Mazure, C. M. & Swendsen, J. Sex differences in Alzheimer’s disease and other dementias. Lancet Neurol. 15 , 451–452 (2016)

    2016

  14. [22]

    Lopez-Lee, C., Torres, E. R. S., Carling, G. & Gan, L. Mechanisms of sex differences in Alzheimer’s disease. Neuron 112 , 1208–1221 (2024)

    2024

  15. [23]

    R., Jr et al

    Jack, C. R., Jr et al. Age, sex, and APOE ε4 effects on memory, brain structure, and β- amyloid across 26 the adult life span. JAMA Neurol. 72 , 511–519 (2015)

    2015

  16. [24]

    Aoyama, S. et al. Sex differences in brainstem structure volumes in patients with schizophrenia. Schizophrenia (Heidelb.) 9 , 16 (2023)

    2023

  17. [25]

    Hu, X. et al. Sex-specific alterations of cortical morphometry in treatment-naïve patients with major depressive disorder. Neuropsychopharmacology 47 , 2002–2009 (2022)

    2022

  18. [26]

    Lai, M.-C. et al. Biological sex affects the neurobiology of autism. Brain 136 , 2799–2815 (2013)

    2013

  19. [27]

    Floris, D. L. et al. The link between autism and sex-related neuroanatomy, and associated cognition and gene expression. Am. J. Psychiatry 180 , 50–64 (2023)

    2023

  20. [28]

    Han, L. K. M. et al. Brain aging in major depressive disorder: results from the ENIGMA major depressive disorder working group. Mol. Psychiatry 26 , 5124–5139 (2021)

    2021

  21. [29]

    Han, L. K. M. et al. Contributing factors to advanced brain aging in depression and anxiety disorders. Transl. Psychiatry 11 , 402 (2021)

    2021

  22. [30]

    W., Downar, J., Gunning, F

    Dunlop, K., Victoria, L. W., Downar, J., Gunning, F. M. & Liston, C. Accelerated brain aging predicts impulsivity and symptom severity in depression. Neuropsychopharmacology 46 , 911–919 (2021)

    2021

  23. [31]

    & Langs, G

    Nenning, K.-H. & Langs, G. Machine learning in neuroimaging: from research to clinical practice. Radiologie (Heidelb.) 62 , 1–10 (2022)

    2022

  24. [32]

    Liem, F. et al. Predicting brain-age from multimodal imaging data captures cognitive impairment. Neuroimage 148 , 179–188 (2017)

    2017

  25. [33]

    & Alzheimer’s Disease Neuroimaging Initiative

    Franke, K., Ziegler, G., Klöppel, S., Gaser, C. & Alzheimer’s Disease Neuroimaging Initiative. Estimating the age of healthy subjects from T1-weighted MRI scans using kernel methods: exploring the influence of various parameters. Neuroimage 50 , 883–892 (2010)

    2010

  26. [34]

    Ritchie, S. J. et al. Sex differences in the adult human brain: Evidence from 5216 UK biobank participants. Cereb. Cortex 28 , 2959–2975 (2018)

    2018

  27. [35]

    Yu, Y. et al. Brain-age prediction: Systematic evaluation of site effects, and sample age range and size. Hum. Brain Mapp. 45 , e26768 (2024). 27

    2024

  28. [36]

    Abrol, A. et al. Deep learning encodes robust discriminative neuroimaging representations to outperform standard machine learning. Nat. Commun. 12 , 353 (2021)

    2021

  29. [37]

    F., Vedaldi, A

    Peng, H., Gong, W., Beckmann, C. F., Vedaldi, A. & Smith, S. M. Accurate brain age prediction with lightweight deep neural networks. Med. Image Anal. 68 , 101871 (2021)

    2021

  30. [38]

    Singh, S. P. et al. 3D deep learning on medical images: A review. Sensors (Basel) 20 , 5097 (2020)

    2020

  31. [39]

    Wen, J. et al. Convolutional neural networks for classification of Alzheimer’s disease: Overview and reproducible evaluation. Med. Image Anal. 63 , 101694 (2020)

    2020

  32. [40]

    Jonsson, B. A. et al. Brain age prediction using deep learning uncovers associated sequence variants. Nat. Commun. 10 , 5409 (2019)

    2019

  33. [41]

    Feng, X. et al. Estimating brain age based on a uniform healthy population with deep learning and structural magnetic resonance imaging. Neurobiol. Aging 91 , 15–25 (2020)

    2020

  34. [42]

    & Sajedi, H

    Pardakhti, N. & Sajedi, H. Brain age estimation based on 3D MRI images using 3D convolutional neural network. Multimed. Tools Appl. 79 , 25051–25065 (2020)

    2020

  35. [43]

    & van der Smagt, P

    Kayalibay, B., Jensen, G. & van der Smagt, P. CNN-based segmentation of medical imaging data. arXiv [cs.CV] (2017)

    2017

  36. [44]

    L., Chung, S., Wang, Y

    Chen, J., Bayanagari, V. L., Chung, S., Wang, Y. & Lui, Y. W. Deep learning with diffusion MRI as in vivo microscope reveals sex-related differences in human white matter microstructure. Sci. Rep. 14 , 9835 (2024)

    2024

  37. [45]

    Cole, J. H. et al. Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker. Neuroimage 163 , 115–124 (2017)

    2017

  38. [46]

    Akkus, Z., Galimzianova, A., Hoogi, A., Rubin, D. L. & Erickson, B. J. Deep learning for brain MRI segmentation: State of the art and future directions. J. Digit. Imaging 30 , 449–459 (2017)

    2017

  39. [47]

    & Klein, T

    Wachinger, C., Reuter, M. & Klein, T. DeepNAT: Deep convolutional neural network for segmenting neuroanatomy. Neuroimage 170 , 434–445 (2018)

    2018

  40. [48]

    Liu, Z. et al. Swin transformer: Hierarchical vision transformer using shifted windows. in 2021 IEEE/CVF International Conference on Computer Vision (ICCV) 9992–10002 (IEEE, 2021). 28 doi:10.1109/iccv48922.2021.00986

    2021

  41. [49]

    Atabansi, C. C. et al. A survey of Transformer applications for histopathological image analysis: New developments and future directions. Biomed. Eng. Online 22 , 96 (2023)

    2023

  42. [50]

    Wagner, S. J. et al. Transformer-based biomarker prediction from colorectal cancer histology: A large-scale multicentric study. Cancer Cell 41 , 1650–1661.e4 (2023)

    2023

  43. [51]

    https://arxiv.org/html/2407.06686v1

    MRI Volume-Based Robust Brain Age Estimation Using Weight-Shared Spatial Attention in 3D CNNs. https://arxiv.org/html/2407.06686v1

    arXiv

  44. [52]

    Adeli, E. et al. Deep learning identifies morphological determinants of sex differences in the pre-adolescent brain. Neuroimage 223 , 117293 (2020)

    2020

  45. [53]

    He, S., Feng, Y., Grant, P. E. & Ou, Y. Deep relation learning for regression and its application to brain age estimation. IEEE Trans. Med. Imaging 41 , 2304–2317 (2022)

    2022

  46. [54]

    Miller, K. L. et al. Multimodal population brain imaging in the UK Biobank prospective epidemiological study. Nat. Neurosci. 19 , 1523–1536 (2016)

    2016

  47. [55]

    Alfaro-Almagro, F. et al. Image processing and Quality Control for the first 10,000 brain imaging datasets from UK Biobank. Neuroimage 166 , 400–424 (2018)

    2018

  48. [56]

    Pérez-García, F. et al. A self-supervised learning strategy for postoperative brain cavity segmentation simulating resections. Int. J. Comput. Assist. Radiol. Surg. 16 , 1653–1661 (2021)

    2021

  49. [57]

    Pérez-García, F. et al. Simulation of brain resection for cavity segmentation using self-supervised and semi-supervised learning. in Medical Image Computing and Computer Assisted Intervention – MICCAI 2020 115–125 (Springer International Publishing, Cham, 2020). doi:10.1007/97...

    2020 doi

  50. [58]

    Isensee, F. et al. Automated brain extraction of multisequence MRI using artificial neural networks. Hum. Brain Mapp. 40 , 4952–4964 (2019)

    2019

  51. [59]

    & Ourselin, S

    Pérez-García, F., Sparks, R. & Ourselin, S. TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning. Comput. Methods Programs Biomed. 208 , 106236 (2021). 29

    2021

  52. [60]

    Cardoso, M. J. et al. MONAI: An open-source framework for deep learning in healthcare. arXiv [cs.LG] (2022)

    2022

  53. [61]

    & Weinberger, K

    Huang, G., Liu, Z., Van Der Maaten, L. & Weinberger, K. Q. Densely connected convolutional networks. in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2261–2269 (IEEE, 2017). doi:10.1109/cvpr.2017.243

    2017 doi

  54. [62]

    M., Vidaurre, D., Alfaro-Almagro, F., Nichols, T

    Smith, S. M., Vidaurre, D., Alfaro-Almagro, F., Nichols, T. E. & Miller, K. L. Estimation of brain age delta from brain imaging. Neuroimage 200 , 528–539 (2019)

    2019

  55. [63]

    Kaufmann, T. et al. Common brain disorders are associated with heritable patterns of apparent aging of the brain. Nat. Neurosci. 22 , 1617–1623 (2019)

    2019

  56. [64]

    Bellantuono, L. et al. Predicting brain age with complex networks: From adolescence to adulthood. Neuroimage 225 , 117458 (2021)

    2021

  57. [65]

    & Bengio, S

    Raghu, M., Zhang, C., Kleinberg, J. & Bengio, S. Transfusion: Understanding transfer learning for medical imaging. arXiv [cs.CV] (2019)

    2019

  58. [66]

    Cole, J. H. et al. Brain age predicts mortality. Mol. Psychiatry 23 , 1385–1392 (2018)

    2018

  59. [67]

    & Alzheimer’s Disease Neuroimaging Initiative

    Franke, K., Ristow, M., Gaser, C. & Alzheimer’s Disease Neuroimaging Initiative. Gender-specific impact of personal health parameters on individual brain aging in cognitively unimpaired elderly subjects. Front. Aging Neurosci. 6 , 94 (2014)

    2014

  60. [68]

    Bittner, N. et al. When your brain looks older than expected: combined lifestyle risk and BrainAGE. Brain Struct. Funct. 226 , 621–645 (2021)

    2021

  61. [69]

    & Gaser, C

    Nenadić, I., Dietzek, M., Langbein, K., Sauer, H. & Gaser, C. BrainAGE score indicates accelerated brain aging in schizophrenia, but not bipolar disorder. Psychiatry Res. Neuroimaging 266 , 86–89 (2017)

    2017

  62. [70]

    & Mechelli, A

    Baecker, L., Garcia-Dias, R., Vieira, S., Scarpazza, C. & Mechelli, A. Machine learning for brain age prediction: Introduction to methods and clinical applications. EBioMedicine 72 , 103600 (2021)

    2021

  63. [71]

    Besson, P., Parrish, T., Katsaggelos, A. K. & Bandt, S. K. Geometric deep learning on brain shape predicts sex and age. Comput. Med. Imaging Graph. 91 , 101939 (2021). 30 Supplementary Material Table S1. Dataset Information Dataset Full cohort name MRI Type MRI Sequence Source...

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.