Pith. sign in

REVIEW 4 major objections 4 minor 76 references

Biological Brain Age Estimation using Sex-Aware Adversarial Variational Autoencoder with Multimodal Neuroimages

T0 review · 4 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read A sex-aware adversarial variational autoencoder that disentangles shared and modality-specific MRI features estimates brain age with 2.72-year mean absolute error, beating prior OpenBHB methods.

desk verdict A competent architecture but the headline MAE is measured on a different test set than the baselines, so the SOTA claim is unsupported. read the letter →

arxiv 2412.05632 v1 pith:D2ZBVE7D submitted 2024-12-07 cs.CV cs.AI

Abd Ur Rehman , Azka Rehman , Muhammad Usman , Abdullah Shahid , Sung-Min Gho , Aleum Lee , Tariq M. Khan , Imran Razzak This is my paper · ORCID
classification cs.CVcs.AI
keywords brainageestimationmultimodalMRIvariationalautoencoderadversariallearningdisentangledrepresentationsex-awaremodelOpenBHBdatasetneurodegenerativediseasebiomarker
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

Brain age—how old the brain looks on imaging rather than chronological age—is a candidate biomarker for neurodegenerative disease, but fusing structural and functional MRI often adds noise and hurts accuracy. This paper tries to establish that a sex-aware adversarial variational autoencoder (SA-AVAE) can fuse sMRI and fMRI by splitting the latent code into shared and modality-specific parts, so the functional modality contributes signal without contaminating the structural one. The paper reports a mean absolute error of 2.722 years on its multimodal evaluation, lower than the four prior OpenBHB-based methods it compares against, and shows the multimodal model beats its own sMRI-only counterpart even when that counterpart trains on ten times more data. If the claim holds, multimodal brain-age estimation becomes accurate enough to be worth deploying in clinical screening for early neurodegeneration.

What carries the argument

The central object is the Sex-Aware Adversarial Variational Autoencoder (SA-AVAE), a paired-encoder architecture that decomposes each modality's latent vector $z_i$ into $\mathrm{Shared}(z_i)$ and $\mathrm{Dist}(z_i)$, concatenates them with sex for the regressor, and is trained with five loss families: adversarial loss aligning shared codes to a prior, variational KL loss on distinct codes, reconstruction loss, cross-modality reconstruction, and the shared-distinct distance ratio $L_D = L_{\mathrm{Shared}}^D / L_{\mathrm{Dist}}^D$. This ratio loss is what enforces the contract that shared codes converge while distinct codes diverge, and the sex input is what lets the regressor model male- and female-specific aging trajectories.

What would settle it

Re-run SA-AVAE and all Table III baselines on one common held-out test set from the official OpenBHB internal and external splits, and compare MAEs; if SA-AVAE no longer beats the 3.250-year prior best, or if its multimodal MAE is not better than its sMRI-only MAE on matched data, the paper's central claim is refuted.

Watch

Extended reading notes

Core claim

On its own terms, the paper claims that the SA-AVAE is the most accurate brain-age estimator among compared methods on the OpenBHB dataset. The architecture takes sMRI and fMRI feature vectors, encodes each into a latent space split into a shared code (modality-invariant) and a distinct code (modality-specific), and feeds the concatenated codes plus sex into a regressor that outputs biological age. Adversarial alignment regularizes the shared codes, variational KL losses regularize the distinct codes, and two extra terms—cross-modality reconstruction and a shared-distinct distance ratio loss—push the split to be meaningful. The reported outcome is an overall MAE of $2.722 \pm 1.351$ years, RMSE of $3.039$, and $R^2$ of $0.936$, with the sex-aware version outperforming the same model without sex, the multitask sex variant, and all ablated autoencoder variants.

Load-bearing premise

The headline 2.722-year MAE is computed on a small multimodal set assembled from two fMRI datasets (roughly 320-381 scans), while the four methods it is compared against in Table III were evaluated on OpenBHB test splits; the central claim of superior accuracy collapses if those test sets are not comparable enough for a direct MAE comparison.

Editorial extensions

If this is right

  • The multimodal SA-AVAE reaches 2.722 years MAE, below the best listed OpenBHB comparator at 3.250 years, which would make it the most accurate published method on this benchmark.
  • Injecting sex information directly into the regressor improves both accuracy and balance across male and female subgroups while using fewer parameters than multitask sex prediction.
  • Fusion of sMRI and fMRI helps rather than hurts: the multimodal model beats the unimodal sMRI model (2.722 vs 2.906 MAE) even though the unimodal model was trained on roughly ten times more scans.
  • The disentanglement losses—especially cross-reconstruction and the shared-distinct distance ratio—are what the paper credits for letting fMRI add useful signal without the noise penalty that defeats simple multimodal fusion.
  • Age-group breakdowns show MAE below about 3.1 years in every group from under 25 to 45-55, which the paper reads as robustness across the adult lifespan.

Reading between the lines

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

  • Because the paper evaluates only healthy controls (a limitation it states), the 2.72-year error should not yet be read as a clinical diagnostic accuracy for Alzheimer's or Parkinson's; a testable next step is running SA-AVAE on patient cohorts and checking whether brain-age gap separates patients from controls.
  • The architecture's shared/distinct split is not specific to sMRI versus fMRI; the same loss design could be applied to other paired brain measurements, such as T1 with diffusion MRI or with PET, and the shared-distinct distance ratio would be the component to isolate in an ablation.
  • The reported comparison in Table III mixes test sets: the multimodal result comes from roughly 320-381 scans drawn from two fMRI datasets, while the listed prior methods were evaluated on OpenBHB splits. Re-running all methods on one matched test set would settle whether the 0.5-year advantage is real or a test-set artifact.
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

4 major / 4 minor

Summary. The paper proposes SA-AVAE, a sex-aware adversarial variational autoencoder for biological brain age estimation from sMRI and fMRI. The architecture disentangles latent features into shared and modality-specific codes using adversarial, variational, cross-reconstruction, and shared-distinct distance ratio losses, and feeds the concatenated codes plus sex information into a regressor. Experiments are reported on OpenBHB for the unimodal setting and on a smaller multimodal sample assembled from two additional sources. The paper claims state-of-the-art performance with a multimodal MAE of 2.722 years, robustness across age and sex groups, and a consistent advantage of multimodal over unimodal input.

Significance. If the evaluation were valid, the contribution would be meaningful: the disentanglement objective is clearly specified, sex conditioning is a sensible design choice, and the ablation study shows monotonic improvement when adversarial, variational, and sex-aware components are added. The paper also reports useful implementation details such as architecture sizes, optimizer settings, and training time. However, the central comparison is undermined by the fact that the headline multimodal MAE is measured on a small external sample while the state-of-the-art baselines are evaluated on OpenBHB test splits. As a result, the claimed superiority and the practical significance of the framework are not currently established.

major comments (4)
  1. [Section V-B, Table III] The state-of-the-art comparison is not valid because the proposed method and the baselines are evaluated on different test sets. Section IV-A states that multimodal experiments use 381 scans from references [67] and [68], and Section V-D later says 320 sMRI+fMRI scans were used. These sources are not the OpenBHB internal/external test splits used by the baseline methods in Table III, and OpenBHB is a T1-weighted structural MRI dataset without fMRI. Therefore the MAE of 2.722 cannot be directly compared with the OpenBHB-based MAE values of Aqil et al., Ahmed et al., Cheshmi et al., and Träuble et al. The paper should either re-evaluate all methods on a common held-out set under the same protocol or drop the claim of outperforming state-of-the-art methods.
  2. [Section V-D, Table V] The unimodal-versus-multimodal comparison is confounded by dataset differences. The unimodal SA-AVAE is trained on 3,200 OpenBHB sMRI scans, while the multimodal SA-AVAE uses 320 scans from a different multimodal source. The observed improvement from 2.906 to 2.722 years cannot be attributed to multimodality because training-set size, data domain, and modality composition differ simultaneously. Additionally, Section IV-A reports a total of 381 multimodal scans, while Table V and the surrounding text use 320; this inconsistency must be resolved and the experimental protocol specified precisely.
  3. [Section III-B, Section V-A] The paper's evidence for disentanglement is largely circular. The adversarial, variational, cross-reconstruction, and distance-ratio losses are designed to enforce separation of shared and distinct codes, and the ablation study in Table II only reports downstream age-prediction metrics. No quantitative evaluation of the latent space is provided, such as similarity of shared codes across modalities, separation of distinct codes, or reconstruction diagnostics. Since disentanglement is a core claimed contribution, the manuscript should report a direct measure of the learned representations, not only the final MAE.
  4. [Section I, Section III-B, Eq. (17)] The paper lists a 'comprehensive strategy for fine-tuning loss weight parameters' as a contribution, but no loss-weight sensitivity analysis or final values of mu_1...mu_5 and eta_1...eta_4 are reported. These weights are empirically determined free parameters, and the robustness of the method to their choice is never examined. At minimum, the final weights and a small sensitivity study should be included, otherwise this claimed contribution is unsupported.
minor comments (4)
  1. [Section V-D] The text says the MAE for sMRI-only input ranged from 3.52 to 2.72 years and that multimodal fusion yielded values between 3.59 and 2.72 years; this is inconsistent with Table V, where the unimodal SA-AVAE reports MAE 2.906 and the multimodal reports 2.722. The ranges should be reconciled or the description clarified.
  2. [Section III-B, Eq. (7)] The adversarial loss appears to be written with the roles of real and generated samples reversed. As written, the discriminator receives prior samples as real data and generated shared codes as fake data, which conflicts with the standard adversarial alignment described in the text and with the objective of matching the aggregated posterior to the prior.
  3. [Throughout] The terms 'sex' and 'gender' are used interchangeably in several places (for example, Figure 5 and Table V), although the paper's stated variable is biological sex. The terminology should be made consistent.
  4. [Section IV-A] It is not clearly stated whether the multimodal subsets from references [67] and [68] are part of the OpenBHB dataset or independent external datasets. The distinction matters because the abstract and conclusion claim evaluation on OpenBHB, while the multimodal experiments appear to use separate sources.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the brain-age predictions are empirical outputs of a fitted regressor, and the paper's ablations compare externally measured MAE values rather than restating definitions.

full rationale

The paper's central claim is an empirical performance comparison: SA-AVAE is trained with a composite objective (Eq. 17) that includes regression, reconstruction, adversarial, and variational terms, and the reported MAE/RMSE/R2 values are measured on held-out or external test data rather than derived from the loss definitions. The disentanglement losses (Eqs. 7-15) enforce a particular latent structure, but the claim that this structure improves brain-age prediction is supported by ablation tables (Tables II, IV, V, VI) that report predictive errors, not by a tautology. The comparison against M-AVAE in Table VI cites the authors' prior work [65], [76], but those citations act as an ablation baseline within the same paper's controlled variants, not as an external theorem or uniqueness result that forces the conclusion; the comparison is empirical and could in principle have gone the other way. The paper does not import a "uniqueness theorem" from its own prior work, nor does it rename a known result under new coordinates. The most serious concern visible in the manuscript is that the headline multimodal MAE of 2.722 is computed on 320-381 scans from datasets [67] and [68] (Section IV-A, Section V-D, Table V) while the state-of-the-art baselines in Table III are evaluated on OpenBHB splits, so the headline comparison may be invalidated by differing test sets. That is a correctness and experimental-design risk, not a circular derivation: the number 2.722 is still an empirically fitted prediction, not an input restated as an output. Accordingly, no circular step meeting the required evidence standard is present.

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

The central performance claim rests on several unstated or understated choices: loss weights, latent dimensions, feature counts, and the assumption that external multimodal data can stand in for OpenBHB benchmarks. These are not independently fixed by theory.

free parameters (4)
  • Loss weights mu1-mu5 (eq. 17) = not specified
    Empirically determined trade-off parameters balancing adversarial, variational, reconstruction, regression, and distance-ratio losses. The paper gives no values, so the reported result depends on unstated chosen weights.
  • Loss weights eta1-eta4 (eq. 19) = not specified
    Weights for the single-modality objective, also empirically determined without reported values.
  • Latent dimensions (shared, distinct, total) = shared=50, distinct=70, total=120
    Chosen empirically, per Section IV-B. The split between shared and distinct capacity is a modeling choice that affects the disentanglement and final performance.
  • Feature counts m1, m2 from Random Forest = not specified
    The number of selected features per modality is not reported, yet it sets the model input dimensionality and affects what information is available to the encoders.
assumptions (4)
  • domain assumption Random Forest feature selection is performed on training data only, without leakage from the test split.
    The paper describes using a filter method with Random Forest (Section IV-A) but does not state whether selection is nested within cross-validation or applied to the full dataset before splitting. Leakage would inflate reported accuracy.
  • domain assumption The two external fMRI datasets [67], [68] are representative of the same population and age range as the OpenBHB benchmark, so results can be compared across datasets.
    Used in Section IV-A and V-B to justify comparing the multimodal model's MAE on external scans with OpenBHB-based methods. This is the load-bearing assumption behind the central comparison.
  • domain assumption Sex information concatenated to the regressor input improves age estimation by capturing sex-specific aging patterns.
    The paper assumes this, citing [8], [14], and relying on it for the SA-AVAE design (Equation 6, Section III-A).
  • domain assumption The shared and distinct latent splits are meaningful and can be learned by the proposed losses.
    The loss functions in Section III-B are designed to enforce this split, and the paper's claims of disentanglement rest on these assumptions rather than on an independent measure.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Biological Brain Age Estimation using Sex-Aware Adversarial Variational Autoencoder with Multimodal Neuroimages." pith.science (2026). https://pith.science/paper/D2ZBVE7D

@misc{pith2026241205632,
  author       = {Pith},
  title        = {Pith review of: Biological Brain Age Estimation using Sex-Aware Adversarial Variational Autoencoder with Multimodal Neuroimages},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/D2ZBVE7D}},
  note         = {Machine review of arXiv:2412.05632}
}
read the original abstract

Brain aging involves structural and functional changes and therefore serves as a key biomarker for brain health. Combining structural magnetic resonance imaging (sMRI) and functional magnetic resonance imaging (fMRI) has the potential to improve brain age estimation by leveraging complementary data. However, fMRI data, being noisier than sMRI, complicates multimodal fusion. Traditional fusion methods often introduce more noise than useful information, which can reduce accuracy compared to using sMRI alone. In this paper, we propose a novel multimodal framework for biological brain age estimation, utilizing a sex-aware adversarial variational autoencoder (SA-AVAE). Our framework integrates adversarial and variational learning to effectively disentangle the latent features from both modalities. Specifically, we decompose the latent space into modality-specific codes and shared codes to represent complementary and common information across modalities, respectively. To enhance the disentanglement, we introduce cross-reconstruction and shared-distinct distance ratio loss as regularization terms. Importantly, we incorporate sex information into the learned latent code, enabling the model to capture sex-specific aging patterns for brain age estimation via an integrated regressor module. We evaluate our model using the publicly available OpenBHB dataset, a comprehensive multi-site dataset for brain age estimation. The results from ablation studies and comparisons with state-of-the-art methods demonstrate that our framework outperforms existing approaches and shows significant robustness across various age groups, highlighting its potential for real-time clinical applications in the early detection of neurodegenerative diseases.

Figures

Figures reproduced from arXiv: 2412.05632 by the authors.

Figure 1
Figure 1. Visualization of structural magnetic resonance imaging (sMRI) and functional magnetic resonance imaging (fMRI) in male and female patients across various age groups. sMRI reveals the anatomical details of the brain, whereas fMRI depicts brain activity by measuring changes in blood flow; this is shown on a color scale where warmer colors typically indicate higher levels of activity. TABLE I: Bird’s-eye view compariso… view at source ↗
Figure 2
Figure 2. It consists of two autoencoder networks: the primary [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 2
Figure 2. Architecture of the proposed Multimodal Sex-Aware Adversarial Variational Autoencoder (SA-AVAE) for predicting biological brain age, utilizing sMRI as a compulsory modality and fMRI as an optional input to enhance prediction performance. B. Feature Disentanglement Strategy The proposed SA-AVAE framework introduces a robust fea￾ture disentanglement strategy, leveraging adversarial learning, variational constraints, s… view at source ↗
Figures from the paper (5 more)
Figure 3
Figure 3. Figure 3: This balanced sex distribution is crucial for minimiz [PITH_FULL_IMAGE:figures/full_fig_p007_3.png]
Figure 3
Figure 3. Figure 3: Age distribution of male and female participants in the training and validation sets of the OpenBHB dataset[66] , shown in (a) and (b), respectively. Several implementation strategies were critical for opti￾mizing the performance of the SA-AVAE framework. During traini…
Figure 4
Figure 4. Figure 4: Illustration of architectural details of our proposed Sex-Aware Adversarial Variational Autoencoder. TABLE II: Performance comparison of different models evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Determination (R2 ) fo…
Figure 5
Figure 5. Figure 5: Comparison of predictions for different multi-modal models: (a) AAE, (b) VAE, (c) AVAE, (d) SA-AVAE. Each graph plots predicted brain age versus chronological brain age, with gender and confidence intervals indicated. Notably, the SA-AVAE model outperforms other varian…
Figure 6
Figure 6. Figure 6: Mean Absolute Error (MAE) for brain age estimation obtained from unimodal regressors using sMRI, fMRI, and a multimodal regressor combining both sMRI and fMRI. F. Limitations and Future Work The proposed SA-AVAE framework leverages two neu￾roimaging modalities, structu…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

76 extracted references · 70 canonical work pages

  1. [67]

    Neuroimaging predictors of creativity in healthy adults,

    A. Sunavsky and J. Poppenk, “Neuroimaging predictors of creativity in healthy adults,” Neuroimage, vol. 206, p. 116292, 2020

    work page 2020

  2. [68]

    The “narratives

    S. A. Nastase, Y .-F. Liu, H. Hillman, A. Zadbood, L. Hasenfratz, N. Keshavarzian, J. Chen, C. J. Honey, Y . Yeshurun, M. Regev et al. , “The “narratives” fmri dataset for evaluating models of naturalistic language comprehension,” Scientific data, vol. 8, no. 1, p. 250, 2021

    work page 2021

  3. [1]

    Population ageing in a globalized world: Risks and dilemmas?

    H. T. Khan, “Population ageing in a globalized world: Risks and dilemmas?” Journal of evaluation in clinical practice , vol. 25, no. 5, pp. 754–760, 2019. 13

    work page 2019

  4. [2]

    Ageing and parkinson’s disease: why is advancing age the biggest risk factor?

    A. Reeve, E. Simcox, and D. Turnbull, “Ageing and parkinson’s disease: why is advancing age the biggest risk factor?” Ageing research reviews, vol. 14, pp. 19–30, 2014

    work page 2014

  5. [3]

    Aging: an important factor for the pathogenesis of neurodegenerative diseases,

    T. Farooqui and A. A. Farooqui, “Aging: an important factor for the pathogenesis of neurodegenerative diseases,” Mechanisms of ageing and development, vol. 130, no. 4, pp. 203–215, 2009

    work page 2009

  6. [4]

    T1- weighted mri-driven brain age estimation in alzheimer’s disease and parkinson’s disease,

    I. Beheshti, S. Mishra, D. Sone, P. Khanna, and H. Matsuda, “T1- weighted mri-driven brain age estimation in alzheimer’s disease and parkinson’s disease,” Aging and disease , vol. 11, no. 3, p. 618, 2020

    work page 2020

  7. [5]

    Can dementia be prevented? brain aging in a population-based context,

    M. N. Haan and R. Wallace, “Can dementia be prevented? brain aging in a population-based context,” Annu. Rev. Public Health, vol. 25, no. 1, pp. 1–24, 2004

    work page 2004

  8. [6]

    Longitudinal changes in individual brainage in healthy aging, mild cognitive impairment, and alzheimer’s disease,

    K. Franke and C. Gaser, “Longitudinal changes in individual brainage in healthy aging, mild cognitive impairment, and alzheimer’s disease,” GeroPsych, 2012

    work page 2012

Show all 76 references
  1. [7]

    Differences between chronological and brain age are related to educa- tion and self-reported physical activity,

    J. Steffener, C. Habeck, D. O’Shea, Q. Razlighi, L. Bherer, and Y . Stern, “Differences between chronological and brain age are related to educa- tion and self-reported physical activity,” Neurobiology of aging, vol. 40, pp. 138–144, 2016

    2016

  2. [8]

    Predicting age using neuroimaging: innova- tive brain ageing biomarkers,

    J. H. Cole and K. Franke, “Predicting age using neuroimaging: innova- tive brain ageing biomarkers,” Trends in neurosciences, vol. 40, no. 12, pp. 681–690, 2017

    2017

  3. [9]

    Quantification of the biological age of the brain using neuroimaging,

    J. H. Cole, K. Franke, and N. Cherbuin, “Quantification of the biological age of the brain using neuroimaging,” Biomarkers of human aging , pp. 293–328, 2019

    2019

  4. [10]

    S. M. Plis, M. F. Amin, A. Chekroud, D. Hjelm, E. Damaraju, H. J. Lee, J. R. Bustillo, K. Cho, G. D. Pearlson, and V . D. Calhoun, “Reading the (functional) writing on the (structural) wall: Multimodal fusion of brain structure and function via a deep neural network based tran...

    2018

  5. [11]

    Morphological brain age prediction using multi- view brain networks derived from cortical morphology in healthy and disordered participants,

    J. Corps and I. Rekik, “Morphological brain age prediction using multi- view brain networks derived from cortical morphology in healthy and disordered participants,” Scientific reports, vol. 9, no. 1, p. 9676, 2019

    2019

  6. [12]

    Importance of multimodal mri in characterizing brain tissue and its potential application for individual age prediction,

    A. Cherubini, M. E. Caligiuri, P. Peran, U. Sabatini, C. Cosentino, and F. Amato, “Importance of multimodal mri in characterizing brain tissue and its potential application for individual age prediction,” IEEE J. Biomed. Health Informat. , vol. 20, no. 5, pp. 1232–1239, Sep 2016

    2016

  7. [13]

    Imaging structural and functional brain development in early childhood,

    J. H. Gilmore, R. C. Knickmeyer, and W. Gao, “Imaging structural and functional brain development in early childhood,” Nature Rev. Neurosci., vol. 19, no. 3, pp. 123–137, 2018

    2018

  8. [14]

    Estimating the age of healthy subjects from t1-weighted mri scans using kernel methods: exploring the influence of various parameters,

    K. Franke, G. Ziegler, S. Kl ¨oppel, C. Gaser, A. D. N. Initiative et al. , “Estimating the age of healthy subjects from t1-weighted mri scans using kernel methods: exploring the influence of various parameters,” Neuroimage, vol. 50, no. 3, pp. 883–892, 2010

    2010

  9. [15]

    Invertible modeling of bidirectional relationships in neuroimaging with normalizing flows: application to brain aging,

    M. Wilms, J. J. Bannister, P. Mouches, M. E. MacDonald, D. Rajashekar, S. Langner, and N. D. Forkert, “Invertible modeling of bidirectional relationships in neuroimaging with normalizing flows: application to brain aging,” IEEE Transactions on Medical Imaging , vol. 41, no. 9,...

    2022

  10. [16]

    Predicting brain-age from multimodal imaging data captures cognitive impairment,

    F. Liem, G. Varoquaux, J. Kynast, F. Beyer, S. K. Masouleh, J. M. Huntenburg, L. Lampe, M. Rahim, A. Abraham, R. C. Craddock et al., “Predicting brain-age from multimodal imaging data captures cognitive impairment,” Neuroimage, vol. 148, pp. 179–188, 2017

    2017

  11. [17]

    Multi-channel attention-fusion neural network for brain age estimation: Accuracy, generality, and interpretation with 16,705 healthy mris across lifespan,

    S. He, D. Pereira, J. D. Perez, R. L. Gollub, S. N. Murphy, S. Prabhu, R. Pienaar, R. L. Robertson, P. E. Grant, and Y . Ou, “Multi-channel attention-fusion neural network for brain age estimation: Accuracy, generality, and interpretation with 16,705 healthy mris across lifesp...

    2021

  12. [18]

    Brain age estimation from mri using cascade networks with ranking loss,

    J. Cheng and et al., “Brain age estimation from mri using cascade networks with ranking loss,” IEEE Trans. Med. Imaging, vol. 40, no. 12, pp. 3400–3412, Dec 2021

    2021

  13. [19]

    Age-net: An mri-based iterative framework for brain biological age estimation,

    K. Armanious and et al., “Age-net: An mri-based iterative framework for brain biological age estimation,” IEEE Trans. Med. Imaging, vol. 40, no. 7, pp. 1778–1791, Jul 2021

    2021

  14. [20]

    Multimodal biological brain age prediction using magnetic resonance imaging and angiography with the identification of predictive regions,

    P. Mouches, M. Wilms, D. Rajashekar, S. Langner, and N. D. Forkert, “Multimodal biological brain age prediction using magnetic resonance imaging and angiography with the identification of predictive regions,” Human Brain Mapping , vol. 43, no. 8, pp. 2554–2566, 2022

    2022

  15. [21]

    Graph transformer geometric learning of brain networks using multimodal mr images for brain age estimation,

    H. Cai, Y . Gao, and M. Liu, “Graph transformer geometric learning of brain networks using multimodal mr images for brain age estimation,” IEEE Trans. Med. Imaging , vol. 42, no. 2, pp. 456–466, Feb 2023

    2023

  16. [22]

    Risk prediction of alzheimer’s disease conversion in mild cognitive impaired population based on brain age estimation,

    W. Liu, Q. Dong, S. Sun, J. Shen, K. Qian, and B. Hu, “Risk prediction of alzheimer’s disease conversion in mild cognitive impaired population based on brain age estimation,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 31, pp. 2468–2476, 2023

    2023

  17. [23]

    Disentangled-multimodal adversarial autoencoder: Application to infant age prediction with incomplete multimodal neu- roimages,

    D. Hu and et al., “Disentangled-multimodal adversarial autoencoder: Application to infant age prediction with incomplete multimodal neu- roimages,” IEEE Trans. Med. Imaging , vol. 39, no. 12, pp. 4137–4149, Dec 2020

    2020

  18. [24]

    Estimating brain age based on a uniform healthy population with deep learning and structural magnetic resonance imaging,

    X. Feng, Z. C. Lipton, J. Yang, S. A. Small, F. A. Provenzano, A. D. N. Initiative, F. L. D. N. Initiative et al., “Estimating brain age based on a uniform healthy population with deep learning and structural magnetic resonance imaging,” Neurobiology of aging , vol. 91, pp. 15...

    2020

  19. [25]

    Learning patterns of the ageing brain in mri using deep convolutional networks,

    N. K. Dinsdale, E. Bluemke, S. M. Smith, Z. Arya, D. Vidaurre, M. Jenkinson, and A. I. Namburete, “Learning patterns of the ageing brain in mri using deep convolutional networks,” NeuroImage, vol. 224, p. 117401, 2021

    2021

  20. [26]

    Accurate brain age prediction with lightweight deep neural networks,

    H. Peng, W. Gong, C. F. Beckmann, A. Vedaldi, and S. M. Smith, “Accurate brain age prediction with lightweight deep neural networks,” Medical image analysis , vol. 68, p. 101871, 2021

    2021

  21. [27]

    Global-local transformer for brain age estimation,

    S. He, P. E. Grant, and Y . Ou, “Global-local transformer for brain age estimation,” IEEE Trans. Med. Imaging , vol. 41, no. 1, pp. 213–224, Jan 2022

    2022

  22. [28]

    Structural networks for brain age prediction,

    O. Pina, I. Cumplido-Mayoral, R. Cacciaglia, J. M. Gonz ´alez-de Ech´avarri, J. D. Gispert, and V . Vilaplana, “Structural networks for brain age prediction,” in International Conference on Medical Imaging with Deep Learning. PMLR, 2022, pp. 944–960

    2022

  23. [29]

    Brain network analyses of diffusion tensor imaging for brain aging,

    S. Xu, X. Yao, L. Han, Y . Lv, X. Bu, G. Huang, Y . Fan, T. Yu, and G. Huang, “Brain network analyses of diffusion tensor imaging for brain aging,” Math Biosci Eng , vol. 18, pp. 6066–78, 2021

    2021

  24. [30]

    Statistical estimation of physiological brain age as a descriptor of senescence rate during adulthood,

    A. Irimia, C. M. Torgerson, S.-Y . M. Goh, and J. D. Van Horn, “Statistical estimation of physiological brain age as a descriptor of senescence rate during adulthood,” Brain imaging and behavior , vol. 9, pp. 678–689, 2015

    2015

  25. [31]

    Importance of multimodal mri in characterizing brain tissue and its potential application for individual age prediction,

    A. Cherubini, M. E. Caligiuri, P. P ´eran, U. Sabatini, C. Cosentino, and F. Amato, “Importance of multimodal mri in characterizing brain tissue and its potential application for individual age prediction,” IEEE journal of biomedical and health informatics , vol. 20, no. 5, pp...

    2016

  26. [32]

    Multimodality neuroimaging brain-age in uk biobank: relationship to biomedical, lifestyle, and cognitive factors,

    J. H. Cole, “Multimodality neuroimaging brain-age in uk biobank: relationship to biomedical, lifestyle, and cognitive factors,”Neurobiology of aging, vol. 92, pp. 34–42, 2020

    2020

  27. [33]

    Base: Brain age standardized evaluation,

    L. Dular and Z. Spiclin, “Base: Brain age standardized evaluation,” NeuroImage, vol. 285, p. 120469, Jan 2024

    2024

  28. [34]

    Residual attention based uncertainty-guided mean teacher model for semi-supervised breast masses segmentation in 2d ultrasonography,

    M. U. Farooq, Z. Ullah, and J. Gwak, “Residual attention based uncertainty-guided mean teacher model for semi-supervised breast masses segmentation in 2d ultrasonography,” Computerized Medical Imaging and Graphics , vol. 104, p. 102173, 2023

    2023

  29. [35]

    Mask-transformer-based networks for teeth segmentation in panoramic radiographs,

    M. Kanwal, M. M. Ur Rehman, M. U. Farooq, and D.-K. Chae, “Mask-transformer-based networks for teeth segmentation in panoramic radiographs,” Bioengineering, vol. 10, no. 7, p. 843, 2023

    2023

  30. [36]

    In- telligent healthcare system for iomt-integrated sonography: Leveraging multi-scale self-guided attention networks and dynamic self-distillation,

    M. Usman, A. Rehman, S. Masood, T. M. Khan, and J. Qadir, “In- telligent healthcare system for iomt-integrated sonography: Leveraging multi-scale self-guided attention networks and dynamic self-distillation,” Internet of Things , vol. 25, p. 101065, 2024

    2024

  31. [37]

    Mesaha-net: Multi-encoders based self- adaptive hard attention network with maximum intensity projections for lung nodule segmentation in ct scan,

    M. Usman, A. Rehman, A. Shahid, S. Latif, S. S. Byon, S. H. Kim, T. M. Khan, and Y . G. Shin, “Mesaha-net: Multi-encoders based self- adaptive hard attention network with maximum intensity projections for lung nodule segmentation in ct scan,” arXiv preprint arXiv:2304.01576, 2023

    2023 arXiv

  32. [38]

    Meds-net: Multi-encoder based self-distilled network with bidirectional maximum intensity projections fusion for lung nodule detection,

    M. Usman, A. Rehman, A. Shahid, S. Latif, and Y .-G. Shin, “Meds-net: Multi-encoder based self-distilled network with bidirectional maximum intensity projections fusion for lung nodule detection,” Engineering Applications of Artificial Intelligence , vol. 129, p. 107597, 2024

    2024

  33. [39]

    Selective deeply supervised multi-scale attention network for brain tumor seg- mentation,

    A. Rehman, M. Usman, A. Shahid, S. Latif, and J. Qadir, “Selective deeply supervised multi-scale attention network for brain tumor seg- mentation,” Sensors, vol. 23, no. 4, p. 2346, 2023

    2023

  34. [40]

    Ldmres-net: A lightweight neural network for efficient medical image segmentation on iot and edge devices,

    S. Iqbal, T. M. Khan, S. S. Naqvi, A. Naveed, M. Usman, H. A. Khan, and I. Razzak, “Ldmres-net: A lightweight neural network for efficient medical image segmentation on iot and edge devices,” IEEE journal of biomedical and health informatics , 2023

    2023

  35. [41]

    Deha-net: a dual-encoder-based hard attention network with an adaptive roi mechanism for lung nodule segmentation,

    M. Usman and Y .-G. Shin, “Deha-net: a dual-encoder-based hard attention network with an adaptive roi mechanism for lung nodule segmentation,” Sensors, vol. 23, no. 4, p. 1989, 2023

    1989

  36. [42]

    Dual-stage deeply supervised attention-based convolutional neural networks for mandibular canal segmentation in cbct scans,

    M. Usman, A. Rehman, A. M. Saleem, R. Jawaid, S.-S. Byon, S.- H. Kim, B.-D. Lee, M.-S. Heo, and Y .-G. Shin, “Dual-stage deeply supervised attention-based convolutional neural networks for mandibular canal segmentation in cbct scans,” Sensors, vol. 22, no. 24, p. 9877, 2022

    2022

  37. [43]

    Ssmd-unet: semi-supervised multi-task decoders network for diabetic retinopathy segmentation,

    Z. Ullah, M. Usman, S. Latif, A. Khan, and J. Gwak, “Ssmd-unet: semi-supervised multi-task decoders network for diabetic retinopathy segmentation,” Scientific Reports, vol. 13, no. 1, p. 9087, 2023. 14

    2023

  38. [44]

    Automating mo- tion correction in multishot mri using generative adversarial networks,

    S. Latif, M. Asim, M. Usman, J. Qadir, and R. Rana, “Automating mo- tion correction in multishot mri using generative adversarial networks,” arXiv preprint arXiv:1811.09750 , 2018

    2018 arXiv

  39. [45]

    Evaluation of the feasibility of explainable computer-aided detection of cardiomegaly on chest radiographs using deep learning,

    M. S. Lee, Y . S. Kim, M. Kim, M. Usman, S. S. Byon, S. H. Kim, B. I. Lee, and B.-D. Lee, “Evaluation of the feasibility of explainable computer-aided detection of cardiomegaly on chest radiographs using deep learning,” Scientific reports, vol. 11, no. 1, p. 16885, 2021

    2021

  40. [46]

    Mobile technologies for managing non- communicable diseases in developing countries,

    S. Latif, M. Y . Khan, A. Qayyum, J. Qadir, M. Usman, S. M. Ali, Q. H. Abbasi, and M. A. Imran, “Mobile technologies for managing non- communicable diseases in developing countries,” in Mobile applications and solutions for social inclusion . IGI Global, 2018, pp. 261–287

    2018

  41. [47]

    Mtss-aae: Multi-task semi- supervised adversarial autoencoding for covid-19 detection based on chest x-ray images,

    Z. Ullah, M. Usman, and J. Gwak, “Mtss-aae: Multi-task semi- supervised adversarial autoencoding for covid-19 detection based on chest x-ray images,” Expert Systems with Applications , vol. 216, p. 119475, 2023

    2023

  42. [48]

    Densely attention mech- anism based network for covid-19 detection in chest x-rays,

    Z. Ullah, M. Usman, S. Latif, and J. Gwak, “Densely attention mech- anism based network for covid-19 detection in chest x-rays,” Scientific Reports, vol. 13, no. 1, p. 261, 2023

    2023

  43. [49]

    Using deep autoencoders for facial expression recognition,

    M. Usman, S. Latif, and J. Qadir, “Using deep autoencoders for facial expression recognition,” in 2017 13th International Conference on Emerging Technologies (ICET). IEEE, 2017, pp. 1–6

    2017

  44. [50]

    Cascade multiscale residual attention cnns with adaptive roi for automatic brain tumor segmentation,

    Z. Ullah, M. Usman, M. Jeon, and J. Gwak, “Cascade multiscale residual attention cnns with adaptive roi for automatic brain tumor segmentation,” Information sciences, vol. 608, pp. 1541–1556, 2022

    2022

  45. [51]

    Retrospec- tive motion correction in multishot mri using generative adversarial network,

    M. Usman, S. Latif, M. Asim, B.-D. Lee, and J. Qadir, “Retrospec- tive motion correction in multishot mri using generative adversarial network,” Scientific reports, vol. 10, no. 1, p. 4786, 2020

    2020

  46. [52]

    V olumetric lung nodule segmentation using adaptive roi with multi- view residual learning,

    M. Usman, B.-D. Lee, S.-S. Byon, S.-H. Kim, B.-i. Lee, and Y .-G. Shin, “V olumetric lung nodule segmentation using adaptive roi with multi- view residual learning,” Scientific Reports , vol. 10, no. 1, p. 12839, 2020

    2020

  47. [53]

    Cross lingual speech emotion recognition: Urdu vs. western languages,

    S. Latif, A. Qayyum, M. Usman, and J. Qadir, “Cross lingual speech emotion recognition: Urdu vs. western languages,” in 2018 International conference on frontiers of information technology (FIT) . IEEE, 2018, pp. 88–93

    2018

  48. [54]

    Phonocardiographic sensing using deep learning for abnormal heartbeat detection,

    S. Latif, M. Usman, R. Rana, and J. Qadir, “Phonocardiographic sensing using deep learning for abnormal heartbeat detection,” IEEE Sensors Journal, vol. 18, no. 22, pp. 9393–9400, 2018

    2018

  49. [55]

    Leveraging data science to combat covid-19: A comprehensive review,

    S. Latif, M. Usman, S. Manzoor, W. Iqbal, J. Qadir, G. Tyson, I. Castro, A. Razi, M. N. K. Boulos, A. Weller et al. , “Leveraging data science to combat covid-19: A comprehensive review,” IEEE Transactions on Artificial Intelligence, vol. 1, no. 1, pp. 85–103, 2020

    2020

  50. [56]

    Lssf-net: Lightweight segmentation with self-awareness, spatial atten- tion, and focal modulation,

    H. Farooq, Z. Zafar, A. Saadat, T. M. Khan, S. Iqbal, and I. Razzak, “Lssf-net: Lightweight segmentation with self-awareness, spatial atten- tion, and focal modulation,” Artificial Intelligence in Medicine, vol. 158, 2024

    2024

  51. [57]

    Ad-net: Attention-based dilated convolutional residual network with guided decoder for robust skin lesion segmentation,

    A. Naveed, S. S. Naqvi, T. M. Khan, S. Iqbal, M. Y . Wani, and H. A. Khan, “Ad-net: Attention-based dilated convolutional residual network with guided decoder for robust skin lesion segmentation,” Neural Computing and Applications , pp. 1–23, 2024

    2024

  52. [58]

    Tbconvl-net: A hybrid deep learning architecture for robust medical image segmentation,

    S. Iqbal, T. M. Khan, S. S. Naqvi, A. Naveed, and E. Meijering, “Tbconvl-net: A hybrid deep learning architecture for robust medical image segmentation,” Pattern Recognition, vol. 158, p. 111028, 2025

    2025

  53. [59]

    Tesl- net: A transformer-enhanced cnn for accurate skin lesion segmentation,

    S. Iqbal, M. Zeeshan, M. Mehmood, T. M. Khan, and I. Razzak, “Tesl- net: A transformer-enhanced cnn for accurate skin lesion segmentation,” arXiv preprint arXiv:2408.09687 , 2024

    2024 arXiv

  54. [60]

    Motion corrected multishot mri reconstruction using generative networks with sensitivity encoding,

    M. Usman, M. U. Farooq, S. Latif, M. Asim, and J. Qadir, “Motion corrected multishot mri reconstruction using generative networks with sensitivity encoding,” arXiv preprint arXiv:1902.07430 , 2019

    1902 arXiv

  55. [61]

    Meds-net: Self-distilled multi-encoders network with bi-direction maximum intensity projections for lung nodule detection,

    M. Usman, A. Rehman, A. Shahid, S. Latif, S. S. Byon, B. D. Lee, S. H. Kim, Y . G. Shin et al. , “Meds-net: Self-distilled multi-encoders network with bi-direction maximum intensity projections for lung nodule detection,” arXiv preprint arXiv:2211.00003 , 2022

    2022 arXiv

  56. [62]

    Ra-net: Region-aware attention network for skin lesion segmentation,

    A. Naveed, S. S. Naqvi, S. Iqbal, I. Razzak, H. A. Khan, and T. M. Khan, “Ra-net: Region-aware attention network for skin lesion segmentation,” Cognitive Computation, pp. 1–18, 2024

    2024

  57. [63]

    Multimodal machine learning: A survey and taxonomy,

    T. Baltrusaitis, C. Ahuja, and L.-P. Morency, “Multimodal machine learning: A survey and taxonomy,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 41, no. 2, pp. 423–443, Feb 2019

    2019

  58. [64]

    Dc-aae: dual channel adversarial autoencoder with multitask learning for kl-grade classifica- tion in knee radiographs,

    M. U. Farooq, Z. Ullah, A. Khan, and J. Gwak, “Dc-aae: dual channel adversarial autoencoder with multitask learning for kl-grade classifica- tion in knee radiographs,” Computers in Biology and Medicine, vol. 167, p. 107570, 2023

    2023

  59. [65]

    Advancing metaverse-based healthcare with multimodal neuroimaging fusion via multi-task adversarial variational autoencoder for brain age estimation,

    M. Usman, A. Rehman, A. Shahid, A. U. Rehman, S.-M. Gho, A. Lee, T. M. Khan, and I. Razzak, “Advancing metaverse-based healthcare with multimodal neuroimaging fusion via multi-task adversarial variational autoencoder for brain age estimation,” IEEE Journal of Biomedical and He...

    2024

  60. [66]

    Openbhb: a large-scale multi-site brain mri data-set for age prediction and debiasing,

    B. Dufumier, A. Grigis, J. Victor, C. Ambroise, V . Frouin, and E. Duch- esnay, “Openbhb: a large-scale multi-site brain mri data-set for age prediction and debiasing,” NeuroImage, vol. 263, p. 119637, 2022

    2022

  61. [69]

    Relief-based feature selection: Introduction and review,

    R. J. Urbanowicz, M. Meeker, W. La Cava, R. S. Olson, and J. H. Moore, “Relief-based feature selection: Introduction and review,” Journal of biomedical informatics, vol. 85, pp. 189–203, 2018

    2018

  62. [70]

    Rapid feature selection based on random forests for high-dimensional data,

    H. Kawakubo and H. Yoshida, “Rapid feature selection based on random forests for high-dimensional data,” Expert Syst. Appl, vol. 40, pp. 6241– 6252, 2012

    2012

  63. [71]

    Robust brain age estimation via regression models and mri-derived features,

    M. Ahmed, U. Sardar, S. Ali, S. Alam, M. Patterson, and I. U. Khan, “Robust brain age estimation via regression models and mri-derived features,” in International Conference on Computational Collective Intelligence. Springer, 2023, pp. 661–674

    2023

  64. [72]

    Confounding factors mitigation in brain age prediction using mri with deformation fields,

    K. Aqil, T. Kulkarni, J. Jayakumar, K. Ram, and M. Sivaprakasam, “Confounding factors mitigation in brain age prediction using mri with deformation fields,” in International Workshop on PRedictive Intelli- gence In MEdicine . Springer, 2023, pp. 58–69

    2023

  65. [73]

    Brain age estimation using structural mri: A clustered federated learn- ing approach,

    S. S. Cheshmi, A. Mahyar, A. Soroush, Z. Rezvani, and B. Farahani, “Brain age estimation using structural mri: A clustered federated learn- ing approach,” in 2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS) . IEEE, 2023, pp. 1–6

    2023

  66. [74]

    Contrastive learning with dynamic localized repulsion for brain age prediction on 3d stiffness maps,

    J. Tr ¨auble, L. Hiscox, C. Johnson, C.-B. Sch ¨onlieb, G. Kamin- ski Schierle, and A. Aviles-Rivero, “Contrastive learning with dynamic localized repulsion for brain age prediction on 3d stiffness maps,” arXiv e-prints, pp. arXiv–2408, 2024

    2024

  67. [75]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 770–778

    2016

  68. [76]

    Multi-task adversarial variational autoen- coder for estimating biological brain age with multimodal neuroimag- ing,

    M. Usman, A. Rehman, A. Shahid, A. U. Rehman, S.-M. Gho, A. Lee, T. M. Khan, and I. Razzak, “Multi-task adversarial variational autoen- coder for estimating biological brain age with multimodal neuroimag- ing,” arXiv preprint arXiv:2411.10100 , 2024

    2024 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.