Pith. sign in

REVIEW 5 major objections 6 minor 23 references

BrainMAP: Multimodal Graph Learning For Efficient Brain Disease Localization

T0 review · 5 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper claims that focusing brain-graph learning on atlas-defined disease-relevant subgraphs cuts computational overhead by more than half, and that a two-stage attention-gated fusion of fMRI and DTI preserves or improves…

desk verdict A sensible multimodal brain-graph pipeline undone by a likely label leak in the atlas-mask selection, which is not nested inside the cross-validation; the accuracy claims need re-running before they can be trusted. read the letter →

arxiv 2506.11178 v1 pith:AZAN5OZV submitted 2025-06-12 cs.CV cs.LGcs.NE

Nguyen Linh Dan Le , Jing Ren , Ciyuan Peng , Chengyao Xie , Bowen Li , Feng Xia This is my paper · ORCID
classification cs.CVcs.LGcs.NE
keywords BraindiseaseGraphlearningLightweightmodelLocalizationMultimodalfusionfMRIDTIAtlas-guidedsubgraphfiltering
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

BrainMAP is a multimodal graph-learning framework for detecting Alzheimer's and Parkinson's disease from paired fMRI and DTI brain graphs. It tries to establish that localizing disease-relevant brain subgraphs before training is not just an interpretability aid but an efficiency mechanism: an atlas-guided filter keeps only ROIs whose pooled features separate patient groups better than the full 90-node graph, and a two-stage fusion then combines the surviving functional and structural signals. The reported results are 82.3% accuracy and 91.5% AUC on ADNI and 86.2% accuracy and 93.8% AUC on PPMI, with per-epoch runtimes of 0.0243 s and 0.0023 s, respectively, well below multimodal baselines. If correct, this is a practical path toward running multimodal brain-graph diagnosis on resource-constrained clinical hardware while outputting a concrete disease-localization map.

What carries the argument

The load-bearing mechanism is the Atlas-Guided Subgraph Filtering (AGSF) module paired with the Attention-Gated Interaction Fusion (AGIF) module. AGSF uses the AAL atlas's predefined subsystems, pools each subject's node features within each mask, trains a random forest on those pooled features against the disease labels, and forms the reduced ROI set $U$ as the union of masks whose mean accuracy exceeds the all-nodes baseline. AGIF then projects the filtered fMRI and DTI features into a shared space, computes scaled-dot-product cross-node attention from each fMRI parcel to the DTI tracts, and for each parcel forms agreement, disparity, gated-mixture, and raw-modality terms that are concatenated into a single embedding. SVD-based feature distillation, adopted from the paper's reference [19], removes the largest and smallest loading components and applies Bernoulli dropout before the embedding is fed to a sparse GCN. Because all subsequent attention and convolution steps run on $|U|$ nodes instead of all $N=90$ ROIs, the complexity of the dominant cross-node attention term drops from $O(N^2D)$ to $O(|U|^2D)$.

What would settle it

Re-run the whole pipeline with atlas-mask selection nested inside each cross-validation fold, so masks are chosen only from training subjects and then evaluated on the held-out fold; if ADNI ACC/AUC falls materially below 82.3%/91.5% or PPMI below 86.2%/93.8%, the headline numbers depend on label information used at selection time rather than on the subgraph mechanism itself.

Watch

Extended reading notes

Core claim

The paper's central claim is that the brain regions driving neurodegenerative pathology can be found cheaply enough to be used as a filter rather than as a post-hoc explanation. BrainMAP defines candidate subsystems from the AAL atlas, scores each one by random-forest accuracy on region-pooled features, and keeps the union of subsystems that beat the whole-brain baseline; the resulting subgraph is the only graph the model sees. On the retained nodes, cross-node attention aligns each fMRI parcel with DTI tracts, a per-node gate blends the modalities, and agreement and disparity terms are concatenated with the gated mixture into a 5D-per-node embedding that is distilled by SVD before a lightweight GCN classifies it. With this pipeline the paper reports 82.3% ACC / 91.5% AUC on ADNI and 86.2% ACC / 93.8% AUC on PPMI, and credits the subgraph filter with reducing memory by over 45% and runtime by over 25% relative to its own no-filter variant, while the fusion module is what preserves accuracy.

Load-bearing premise

The load-bearing premise is that using the disease labels to select which brain regions to keep does not bias the later five-fold accuracy evaluation of the classifier, because the selection step is described outside the cross-validation loop.

Editorial extensions

If this is right

  • Multimodal brain-graph diagnosis can run orders of magnitude faster than whole-brain multimodal baselines, with ADNI per-epoch time at 0.0243 s and PPMI at 0.0023 s.
  • The same atlas-scoring procedure doubles as a localization output: on ADNI it highlights subcortical, limbic, olfactory, and auditory regions, consistent with early Alzheimer's involvement.
  • Removing the subgraph filter leaves accuracy essentially unchanged on ADNI but more than doubles memory use; removing the fusion module drops accuracy by about 10 points, so the two modules separate the efficiency and accuracy contributions.
  • Because the fusion module computes per-node agreement, disparity, and gating terms, the model itself produces a per-region confidence signal, which supports clinical interpretation beyond a single diagnosis.

Reading between the lines

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

  • Beyond the paper, the localization mechanism can be tested independently of the fusion design: if randomly chosen equal-sized ROI subsets match BrainMAP's accuracy, then the atlas scoring, not the subgraph content, is the source of the gain.
  • Beyond the paper, the same atlas-scoring procedure transfers to other brain atlases, other modalities, or regression targets such as cognitive scores, since its only requirement is a labeled signal per candidate ROI.
  • Beyond the paper, the reported per-epoch times imply the remaining bottleneck is the $O(|U|^2D)$ cross-node attention, so scaling to larger or smaller graphs would need sparse attention or node sampling, a follow-up the paper does not address.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 6 minor

Summary. The paper proposes BrainMAP, a multimodal graph learning framework for brain disease classification and localization. It first uses an atlas-guided subgraph filtering (AGSF) step that selects a subset of ROIs by comparing random-forest accuracies computed on labeled data against a whole-brain baseline, then fuses fMRI and DTI features via cross-node attention and gated interaction, applies SVD-based feature distillation, and classifies with a lightweight GCN. On ADNI and PPMI, the paper reports higher accuracy and AUC than most baselines and claims over 50% reduction in computational overhead. The central evaluation protocol has a potential label-leakage issue in the atlas-mask selection step, and the main results are reported without error bars, with a selective efficiency comparison, so the headline claims are not yet supported.

Significance. If the headline results survived a properly nested cross-validation, BrainMAP would be a useful practical contribution: it combines localization with efficiency, uses standard atlas-based ROIs, and reports fast per-epoch training. The paper also includes an ablation study and parameter sensitivity checks, which are positive features. However, because the primary accuracy and efficiency claims rest on a non-nested feature-selection procedure and on point estimates without variability, the significance of the work cannot be assessed from the current evidence. No code or detailed experimental logs are provided, further limiting reproducibility.

major comments (5)
  1. [Section IV-A, Algorithm 1; Section V-B] The atlas-mask selection is performed on the full labeled dataset before the five-fold split. Algorithm 1 computes ACC[k] from RFscore(R(k), Y) with the full label vector Y and uses ACC[k] > baseline to define S and U; Section V-B only states that predictive performance is evaluated with five-fold cross-validation and does not state that U is re-estimated inside each training fold. If U is fixed using all subjects, the held-out test folds have already influenced which ROIs are retained, so the reported ACC/AUC in Table I (82.3/91.5 on ADNI, 86.2/93.8 on PPMI) are optimistically biased and do not estimate generalization to unseen subjects. This leak is load-bearing because the abstract and conclusion claim that accuracy is preserved without being compromised. The ablation in Table II does not repair this issue: the w/o AGSF row uses R0 and is leak-free, but the BrainMAP row is not. The authors must repeat the whole pipeline with the AGSF selection nested in each training fold and report the resulting unbiased estimates.
  2. [Section V-B, Table I] The main classification results report fold averages without standard deviations, confidence intervals, or statistical tests. Table I lists no variability for ACC or AUC, so the reader cannot judge whether BrainMAP's differences from baselines, for example 82.3 versus 81.2 for Cross-GNN on ADNI, are meaningful. Table II provides ± values, but the main table does not. Please report mean ± standard deviation over folds for all methods and, ideally, pairwise significance tests.
  3. [Section IV-C; Section V-B] The feature distillation step computes SVD and class prototypes on 'training rows,' but the manuscript does not specify whether these are the training rows within each fold or rows from the full dataset. If the SVD and prototype statistics are estimated on the full data before splitting, this is a second form of label leakage because test rows contribute to the prototypes for their class labels, and the Bernoulli-drop distribution also depends on the labels. Please state explicitly that all data-dependent transformations, including AGSF selection, SVD, prototypes, and masking, are fit only on training folds and applied to test folds.
  4. [Section V-C, Table I] The efficiency claim is selective. Table I shows that GAT achieves per-epoch times of 0.0009 s on both datasets, roughly 27 times faster than BrainMAP's 0.0243 s on ADNI, with lower memory (1.1 GB versus 1.5 GB). The abstract's 'more than 50% reduction in computational overhead' and 'outperforms state-of-the-art methods in computational efficiency' are not true against all compared methods; they hold only against multimodal baselines and against BrainMAP's own w/o AGSF variant. Please specify the intended comparator class, report all methods' parameter counts, and clarify whether times are per epoch or per sample.
  5. [Section V-F, Figures 4 and 5] The localization claim is central to the paper's title and contribution, yet Section V-F provides only qualitative visual examples from ADNI and no quantitative evaluation of whether the selected regions match known disease effects or are stable across cross-validation folds. Please add a quantitative localization evaluation, such as overlap with independently established disease-related ROI sets, consistency of selected ROIs across folds, or comparison with a null selection procedure.
minor comments (6)
  1. [Section III vs. Section V-B] The symbol K is used for both the number of candidate masks in Section III and the feature-group removal count in Section V-B; please rename one of them to avoid ambiguity.
  2. [Section IV-B, Eq. (4)] The denominator in the attention-weight equation is formatted as 'N′s X e^...', which is visually ambiguous; please rewrite it as an explicit summation over j′.
  3. [Algorithm 1, line 9] The union operation in 'U ← S k∈S Rk' is missing the union symbol; it should be 'U ← ∪_{k∈S} R_k'.
  4. [Section V-E] The ablation subsection refers to 'GIF' when the module is called AGIF elsewhere; please use consistent naming.
  5. [Section V-B, Table I] The paper reports a single AUC value for three-class datasets (ADNI: NC/MCI/AD; PPMI: NC/prodromal/PD); please specify the multiclass AUC averaging scheme used.
  6. [Section V-D, Figures 2 and 3] The parameter studies do not state whether the reported accuracy values are computed on training, validation, or test folds; please clarify the protocol for these sweeps.

Circularity Check

1 steps flagged · score 6.0 of 10

Atlas-mask selection uses the full label vector to choose the subgraph, and the five-fold evaluation is not described as nesting that selection, so the reported ACC/AUC and the 'disease-relevant' localization are partly fitted to the labels.

  1. fitted input called prediction [Section IV-A (Algorithm 1) and Section V-B]
    "To evaluate each candidate subgraph’s discriminative power, we repeat Random Forest training S times on (R(k), Y) and record the average accuracy ACCk ... S={k >0 : ACCk >ACC0}. Taking the union of these masks yields: U=... Predictive performance (ACC, AUC) and efficiency (epoch runtime (seconds), peak memory) are evaluated using five-fold cross-validation, with fold averages reported."

    Algorithm 1 selects the ROI set U by comparing random-forest accuracies ACCk computed on the full label vector Y, and then builds the filtered feature matrix X̂ from U. Section V-B reports five-fold cross-validated ACC/AUC but does not state that this mask selection is re-run inside each training fold. As written, the held-out labels influence which ROIs are retained before the splits, so the reported accuracy and AUC estimate the fit of U to the labels rather than generalization to unseen subjects. The claimed 'disease-relevant subgraph' is likewise the subgraph that best classifies the training labels, presented as an independent localization finding.

full rationale

The central circularity is a selection-evaluation leak. Algorithm 1's Step 4 uses the full label vector Y to choose U via ACCk = RFscore(R(k), Y); U then determines X̂ for all subjects and the GCN is trained and evaluated on that filtered representation. The paper never says the atlas-mask selection is nested inside the five-fold cross-validation of Section V-B, so the reproducibility reading is that U is chosen once on all labels before splitting. Under that reading, the test folds' labels contribute to the choice of subgraph, making the reported 82.3% ACC / 91.5% AUC on ADNI and 86.2% ACC / 93.8% AUC on PPMI optimistically biased as estimates of generalization. The efficiency claim (more than 50% reduction in computation) is less affected because it compares runtimes under the same pipeline, but the accuracy side of 'without compromising predictive accuracy' is exactly what the leaked selection calls into question. The feature-distillation module is adopted from the authors' prior work [19]; this is a legitimate citation of a published method and is not load-bearing circularity on its own. The most material issue is the unfitted selection loop, which warrants a score of 6: the predictive and localization claims partially reduce to fitting the subgraph to the labels.

Assumptions & free parameters 6 free parameters · 3 assumptions · 0 invented entities

The central claim depends on the validity of the label-based mask selection, the AAL atlas subsystem definitions, and the SVD distillation prior work. The mask selection threshold and several architectural hyperparameters are fitted to the data, and the evaluation protocol does not clearly separate selection from evaluation.

free parameters (6)
  • mask selection threshold = ACC_k > ACC_0 (data-driven)
    The union of AAL subsystem masks is chosen when their random forest accuracy exceeds the whole-brain baseline, directly fitting the selected regions to the labels.
  • CNA dimension H = 64 (ADNI), 128 (PPMI)
    Tuned per dataset in the parameter study (Fig. 2).
  • GIAC dimension = 32 (ADNI), 128 (PPMI)
    Tuned per dataset in the parameter study (Fig. 2).
  • SVD feature-group removal K = 3
    Set in Section V-B; removes top and bottom loading features in feature distillation.
  • Masking rate = 0.2
    Bernoulli drop probability for feature masking in Section IV-C; set in Section V-B.
  • GCN hidden widths = 128 (ADNI), 32 (PPMI)
    Chosen per dataset; Fig. 3 shows robustness to width.
assumptions (3)
  • domain assumption AAL atlas parcellation and the predefined subsystem masks R_k correspond to disease-relevant anatomical units.
    The entire AGSF module relies on these masks being meaningful for disease localization (Section III, Section IV-A).
  • ad hoc to paper Random forest accuracy on regional-averaged features is a valid proxy for a subgraph's diagnostic relevance.
    The selection criterion ACC_k > ACC_0 is defined by the authors and is not derived from prior theory.
  • domain assumption SVD-based feature distillation from the authors' prior work [19] preserves diagnostic information.
    The module is adopted from Peng et al. 2024 without independent verification, and the current paper does not analyze its failure modes.

how reviews work

0 comments
Cite this review

Pith. "Pith review of BrainMAP: Multimodal Graph Learning For Efficient Brain Disease Localization." pith.science (2026). https://pith.science/paper/AZAN5OZV

@misc{pith2026250611178,
  author       = {Pith},
  title        = {Pith review of: BrainMAP: Multimodal Graph Learning For Efficient Brain Disease Localization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AZAN5OZV}},
  note         = {Machine review of arXiv:2506.11178}
}
read the original abstract

Recent years have seen a surge in research focused on leveraging graph learning techniques to detect neurodegenerative diseases. However, existing graph-based approaches typically lack the ability to localize and extract the specific brain regions driving neurodegenerative pathology within the full connectome. Additionally, recent works on multimodal brain graph models often suffer from high computational complexity, limiting their practical use in resource-constrained devices. In this study, we present BrainMAP, a novel multimodal graph learning framework designed for precise and computationally efficient identification of brain regions affected by neurodegenerative diseases. First, BrainMAP utilizes an atlas-driven filtering approach guided by the AAL atlas to pinpoint and extract critical brain subgraphs. Unlike recent state-of-the-art methods, which model the entire brain network, BrainMAP achieves more than 50% reduction in computational overhead by concentrating on disease-relevant subgraphs. Second, we employ an advanced multimodal fusion process comprising cross-node attention to align functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI) data, coupled with an adaptive gating mechanism to blend and integrate these modalities dynamically. Experimental results demonstrate that BrainMAP outperforms state-of-the-art methods in computational efficiency, without compromising predictive accuracy.

Figures

Figures reproduced from arXiv: 2506.11178 by the authors.

Figure 1
Figure 1. Overview of BrainMAP, including: (1) Atlas-Guided Subgraph Filtering (AGSF) which extracts key nodes from fMRI [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. BrainMAP classification accuracy with CNA-GIAC [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Accuracy and AUC values of BrainMAP under [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Disease-relevant regions from ADNI fMRI data [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Disease-relevant regions from ADNI DTI data [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references · 20 canonical work pages

  1. [1]

    Braingb: a benchmark for brain network analysis with graph neural networks,

    H. Cui, W. Dai, Y . Zhu, X. Kan, A. A. C. Gu, J. Lukemire, L. Zhan, L. He, Y . Guo, and C. Yang, “Braingb: a benchmark for brain network analysis with graph neural networks,”IEEE transactions on medical imaging, vol. 42, no. 2, pp. 493–506, 2022

    2022

  2. [2]

    Rh-brainfs: regional heterogeneous multimodal brain networks fusion strategy,

    H. Ye, Y . Zheng, Y . Li, K. Zhang, Y . Kong, and Y . Yuan, “Rh-brainfs: regional heterogeneous multimodal brain networks fusion strategy,” Advances in Neural Information Processing Systems, vol. 36, pp. 59 286–59 303, 2023

    work page 2023

  3. [3]

    Mapping multi- modal brain connectome for brain disorder diagnosis via cross-modal mutual learning,

    Y . Yang, C. Ye, X. Guo, T. Wu, Y . Xiang, and T. Ma, “Mapping multi- modal brain connectome for brain disorder diagnosis via cross-modal mutual learning,”IEEE Transactions on Medical Imaging, vol. 43, no. 1, pp. 108–121, 2023

    work page 2023

  4. [4]

    Multimodal hyperbolic graph learning for alzheimer’s disease detection,

    C. Xie, W. Zhou, C. Peng, A. N. Hoshyar, C. Xu, U. Naseem, and F. Xia, “Multimodal hyperbolic graph learning for alzheimer’s disease detection,” inAustralasian Joint Conference on Artificial Intelligence. Springer, 2024, pp. 390–403

    work page 2024

  5. [5]

    Multicenter and multichannel pooling gcn for early ad diagnosis based on dual-modality fused brain network,

    X. Song, F. Zhou, A. F. Frangi, J. Cao, X. Xiao, Y . Lei, T. Wang, and B. Lei, “Multicenter and multichannel pooling gcn for early ad diagnosis based on dual-modality fused brain network,”IEEE Transactions on Medical Imaging, vol. 42, no. 2, pp. 354–367, 2022

    work page 2022

  6. [6]

    Feature attention graph neural network for estimating brain age and identifying important neural connections in mouse models of genetic risk for alzheimer’s disease,

    H. S. Moon, A. Mahzarnia, J. Stout, R. J. Anderson, Z. Y . Han, J. T. Tremblay, C. T. Badea, and A. Badea, “Feature attention graph neural network for estimating brain age and identifying important neural connections in mouse models of genetic risk for alzheimer’s disease,” Imaging Neuroscience, vol. 2, pp. 1–22, 2024

    work page 2024

  7. [7]

    Interpretable sparsification of brain graphs: Better practices and effective designs for graph neural networks,

    G. Li, M. Duda, X. Zhang, D. Koutra, and Y . Yan, “Interpretable sparsification of brain graphs: Better practices and effective designs for graph neural networks,” inProceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2023, pp. 1223–1234

    work page 2023

  8. [8]

    Adversarial learning based node-edge graph attention networks for autism spectrum disorder identification,

    Y . Chen, J. Yan, M. Jiang, T. Zhang, Z. Zhao, W. Zhao, J. Zheng, D. Yao, R. Zhang, K. M. Kendricket al., “Adversarial learning based node-edge graph attention networks for autism spectrum disorder identification,”IEEE Transactions on Neural Networks and Learning Systems, 2022

    work page 2022

Show all 23 references
  1. [9]

    Braingnn: Interpretable brain graph neural network for fmri analysis,

    X. Li, Y . Zhou, N. Dvornek, M. Zhang, S. Gao, J. Zhuang, D. Scheinost, L. H. Staib, P. Ventola, and J. S. Duncan, “Braingnn: Interpretable brain graph neural network for fmri analysis,”Medical Image Analysis, vol. 74, p. 102233, 2021

    2021

  2. [10]

    Automated anatom- ical labeling of activations in spm using a macroscopic anatomical parcellation of the mni mri single-subject brain,

    N. Tzourio-Mazoyer, B. Landeau, D. Papathanassiou, F. Crivello, O. Etard, N. Delcroix, B. Mazoyer, and M. Joliot, “Automated anatom- ical labeling of activations in spm using a macroscopic anatomical parcellation of the mni mri single-subject brain,”Neuroimage, vol. 15, no. 1,...

    2002

  3. [11]

    Alzheimer’s disease neuroimaging initiative (adni) clinical characterization,

    R. C. Petersen, P. S. Aisen, L. A. Beckett, M. C. Donohue, A. C. Gamst, D. J. Harvey, C. Jack Jr, W. J. Jagust, L. M. Shaw, A. W. Togaet al., “Alzheimer’s disease neuroimaging initiative (adni) clinical characterization,”Neurology, vol. 74, no. 3, pp. 201–209, 2010

    2010

  4. [12]

    The parkin- son progression marker initiative (ppmi),

    K. Marek, D. Jennings, S. Lasch, A. Siderowf, C. Tanner, T. Simuni, C. Coffey, K. Kieburtz, E. Flagg, S. Chowdhuryet al., “The parkin- son progression marker initiative (ppmi),”Progress in neurobiology, vol. 95, no. 4, pp. 629–635, 2011

    2011

  5. [13]

    Deep joint learning of patholog- ical region localization and alzheimer’s disease diagnosis,

    C. Park, W. Jung, and H.-I. Suk, “Deep joint learning of patholog- ical region localization and alzheimer’s disease diagnosis,”Scientific reports, vol. 13, no. 1, p. 11664, 2023

    2023

  6. [14]

    Explainable fmri- based brain decoding via spatial temporal-pyramid graph convolutional network,

    Z. Ye, Y . Qu, Z. Liang, M. Wang, and Q. Liu, “Explainable fmri- based brain decoding via spatial temporal-pyramid graph convolutional network,”Human Brain Mapping, vol. 44, no. 7, pp. 2921–2935, 2023

    2023

  7. [15]

    Bpi-gnn: In- terpretable brain network-based psychiatric diagnosis and subtyping,

    K. Zheng, S. Yu, L. Chen, L. Dang, and B. Chen, “Bpi-gnn: In- terpretable brain network-based psychiatric diagnosis and subtyping,” NeuroImage, vol. 292, p. 120594, 2024

    2024

  8. [16]

    An adaptive multi-graph neural network with multimodal feature fusion learning for mdd detection,

    T. Xing, Y . Dou, X. Chen, J. Zhou, X. Xie, and S. Peng, “An adaptive multi-graph neural network with multimodal feature fusion learning for mdd detection,”Scientific Reports, vol. 14, no. 1, p. 28400, 2024

    2024

  9. [17]

    4d multimodal co-attention fusion network with latent contrastive align- ment for alzheimer’s diagnosis,

    Y . Wei, Y . Zhang, X. Xiao, T. Wang, X. Wang, and V . D. Calhoun, “4d multimodal co-attention fusion network with latent contrastive align- ment for alzheimer’s diagnosis,”arXiv preprint arXiv:2504.16798, 2025

    2025 arXiv

  10. [18]

    Integrated brain connectivity analysis with fmri, dti, and smri powered by interpretable graph neural networks,

    G. Qu, Z. Zhou, V . D. Calhoun, A. Zhang, and Y .-P. Wang, “Integrated brain connectivity analysis with fmri, dti, and smri powered by interpretable graph neural networks,”Medical Image Analysis, p. 103570, 2025

    2025

  11. [19]

    Adaptive brain network augmentation based on group-aware graph learning,

    C. Peng, M. Liu, C. Meng, S. Yu, and F. Xia, “Adaptive brain network augmentation based on group-aware graph learning,” inThe Second Tiny Papers Track at ICLR 2024, 2024

    2024

  12. [20]

    Gretna: a graph theoretical network analysis toolbox for imaging connectomics,

    J. Wang, X. Wang, M. Xia, X. Liao, A. Evans, and Y . He, “Gretna: a graph theoretical network analysis toolbox for imaging connectomics,” Frontiers in Human Neuroscience, vol. V olume 9 - 2015, 2015

    2015

  13. [21]

    Panda: a pipeline toolbox for analyzing brain diffusion images,

    Z. Cui, S. Zhong, P. Xu, Y . He, and G. Gong, “Panda: a pipeline toolbox for analyzing brain diffusion images,”Frontiers in human neuroscience, vol. 7, p. 42, 2013

    2013

  14. [22]

    Thalamic nuclei changes in early and late onset alzheimer’s disease,

    G. Forno, M. Saranathan, J. Contador, N. Guillen, N. Falg `as, A. Tort- Merino, M. Balasa, R. Sanchez-Valle, M. Hornberger, and A. Llad ´o, “Thalamic nuclei changes in early and late onset alzheimer’s disease,” Current Research in Neurobiology, vol. 4, p. 100084, 2023

    2023

  15. [23]

    The subcortical default mode network and alzheimer’s disease: a systematic review and meta-analysis,

    S. Seoane, M. van den Heuvel, ´A. Acebes, and N. Janssen, “The subcortical default mode network and alzheimer’s disease: a systematic review and meta-analysis,”Brain Communications, vol. 6, no. 2, p. fcae128, 2024

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.