- Rating
- 3
- Confidence
- 4
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes CORE to combine site-specific deconfounding, cross-site selection of label-associated connections, dynamic FC summaries, and gated line-graph learning.
Critical strengths
- LOSO evaluation across four datasets directly addresses cross-site distribution shifts.
- The performance is good.
Critical weaknesses
- The scaffold’s contribution is insufficiently isolated. The ablations do not separately test supervised scaffold selection, sign consistency, bootstrap stability, and reliability weighting. Add simple classifiers on the same scaffold and a parameter-matched nongraph model to establish the line graph’s added value.
- Atlas robustness should be stated narrowly. Separate training and evaluation on AAL and CC200 show that the method works with either parcellation. They do not establish transfer to an unseen atlas.
Questions for authors
- How sensitive is aggregate correction to covariate scaling and the pseudoinverse convention?
- Existing work explicitly explores demographic-enhanced brain network models [1-3]. Why these works are not included for discussion and comparison?
[1] Classification of brain disorders in rs-fmri via local-to-global graph neural networks. TMI 2022 [2] Knowledge-driven multigraph convolutional network for brain network analysis and potential biomarker discovery. MIA 2025 [3] BrainPrompt+: Multi-Level Brain Prompt Learning for Knowledge-Guided Neurological Disorder Identification. TMI 2026
9 comments
Response to Reviewer fiGY (2/6)
| Variant | ABIDE AUC↑ | ABIDE ACC↑ | REST-meta-MDD AUC↑ | REST-meta-MDD ACC↑ |
|---|---|---|---|---|
| Scaffold selection | ||||
| Random scaffold | 58.17±0.41 | 58.31±0.77 | 58.42±0.68 | 57.95±0.83 |
| Unsupervised, highest-variance edges | 60.06±0.42 | 59.42±0.80 | 60.37±0.40 | 59.88±0.56 |
| Single contrast on pooled source subjects | 62.94±0.45 | 62.41±0.56 | 63.95±0.65 | 60.91±0.75 |
| w/o sign consistency, | 63.81±0.40 | 62.19±1.07 | 64.12±0.51 | 62.65±0.49 |
| w/o both, | 62.71±0.33 | 61.38±0.46 | 63.30±0.62 | 61.67±0.73 |
| Top consensus contrasts, same size | 64.45±0.48 | 63.21±0.99 | 64.38±0.66 | 62.97±0.69 |
| Unseen-site correction | ||||
| Uniform aggregation | 65.13±0.83 | 64.80±1.27 | 65.52±0.62 | 63.62±0.70 |
| Classifier on the CORE scaffold | ||||
| Logistic regression, | 60.92±0.43 | 60.95±0.92 | 62.20±0.61 | 60.41±0.46 |
| Linear SVM | 61.05±0.78 | 60.31±0.82 | 62.49±0.39 | 60.18±0.81 |
| Random forest | 59.83±0.88 | 60.12±1.42 | 62.37±0.45 | 59.08±0.97 |
| MLP, parameter-matched | 62.24±0.47 | 61.73±1.14 | 63.31±0.52 | 60.36±0.60 |
| Transformer, parameter-matched | 63.02±0.35 | 62.84±0.69 | 63.18±0.32 | 61.79±0.52 |
| ROI-graph GINE, parameter-matched | 63.58±0.40 | 62.54±0.59 | 64.11±0.49 | 61.40±0.87 |
| CORE w/o LG | 64.00±0.66 | 63.48±0.74 | 63.96±0.59 | 61.70±0.57 |
| Full model | ||||
| CORE | 65.21±0.78 | 64.95±1.11 | 65.58±0.62 | 63.74±0.79 |
As shown in the above table, supervised selection and cross-site consensus are both needed. A random scaffold falls 5.79 to 7.16 points below CORE, and an unsupervised scaffold of the same size 3.86 to 5.53 points below. Selecting edges by one contrast on the pooled source subjects lowers AUC by 1.63 to 2.27 points and ACC by 2.54 to 2.83 points. Taking the top consensus contrasts of the same size without the sign and bootstrap criteria still falls 0.76 to 1.74 points below CORE, which shows that the criteria select better edges than the contrast magnitude alone. Removing sign consistency more than doubles the scaffold and lowers accuracy by 1.09 to 2.76 points. Without sign consistency, bootstrap stability still removes most edges whose sign is carried by only four of the seven sites, and removing it as well lowers accuracy by a further 0.81 to 1.10 points. Uniform aggregation stays within the variation across seeds of reliability weighting, consistent with its role in the correction only.
Response to Reviewer fiGY (3/6)
On the same scaffold and features, the linear classifiers, the random forest and the parameter-matched MLP fall 2.27 to 5.38 points below CORE. The models that capture interactions between pathways perform better, and the Transformer, the GINE and CORE w/o LG lie within about 1 point of each other. Line-graph message passing adds 1.21 to 1.95 points over the best of these three models, which isolates the contribution of the line graph beyond graph or attention models with the same parameter budget.
The scaffold is also reproducible. Over ten runs per fold with different seeds and validation sites, the selected scaffolds have a mean Jaccard index of 0.84 on ABIDE and 0.87 on REST-meta-MDD.
W2: Atlas robustness should be stated narrowly. Separate training and evaluation on AAL and CC200 show that the method works with either parcellation. They do not establish transfer to an unseen atlas.
A2: Thanks for your comment. We agree with this point. CORE and all baselines were trained and evaluated separately with each parcellation. This result shows that the gains hold when the method is applied with either atlas, and it does not show that a trained model transfers to an unseen atlas. We have revised the Abstract, the Introduction, Sec. 4.2 and the Conclusion to state this scope. For example, the Abstract now ends with "CORE can also be trained and evaluated with a second parcellation of ABIDE," and Sec. 4.2 states that each method is trained and evaluated separately with each parcellation, so the result does not address the transfer of a trained model to an unseen atlas.
Q1: How sensitive is aggregate correction to covariate scaling and the pseudoinverse convention?
A3: Thanks for your question. The aggregate correction depends on the covariate scale and the pseudoinverse convention, and this dependence lies mostly in a shift shared by all subjects of a target site. Rescaling the site-level covariates by 10 moves the correction by 0.41 to 0.79 SD of the edge and lowers ACC by 1.49 to 2.41 points, while AUC changes by at most 1 point across all rescalings, and an identified implementation stays within the variation across seeds of the submitted deconfounder.
Within a source site, TR and sequence length take a single value, so their columns in Eq. 5 are multiples of the intercept column. The source data identify the total site offset and the within-site slopes, but not how is assigned between the intercept and the site-level covariates. The same holds for sex at the two all-male ABIDE sites, SBL and TRINITY. The site-wise residuals in Eq. 6 and the scaffold do not depend on this assignment, while the reliability weights in Eq. 7 and the unseen-site correction in Eq. 8 do. With standardized site-level covariates , the pseudoinverse selects the minimum-norm assignment, under which site contributes to the correction at a target site with covariates . Rescaling changes this assignment, while rescaling a covariate that varies within a site only reparameterizes its slope and leaves the correction unchanged, so the sensitivity comes only from the covariates that are constant within a site.
To characterize where this dependence acts, we have added Proposition 3 to Sec. 3.1. For any aggregate deconfounder, the intercept and the offset of a target site cancel in its diagnostic contrast, and the assignment enters the contrast only through the aggregated slopes of covariates that vary within the target site. It reaches these slopes through the reliability weights and, on ABIDE, through the sex coefficients of SBL and TRINITY. The part of the correction shared by all subjects of the target site moves the operating point of the classifier, so ACC is expected to respond more strongly than AUC.
Response to Reviewer fiGY (1/6)
We are truly grateful for the time you have taken to review our paper and your insightful review. Here we address your comments in the following:
A1: Thanks for your comment. The reliability weights do not enter scaffold selection. The scaffold is extracted from the site-wise residuals of the training sites, and the weights αe enter only the correction of validation and test subjects in Eq. 8, so we evaluate them separately from the selection criteria.
To isolate each part of the scaffold and the added value of the line graph, we replaced one part of CORE at a time and kept all other components fixed. The variants are constructed as follows:
We used ABIDE and REST-meta-MDD with three seeds. Averaged over folds, the scaffold has 412 edges on ABIDE and 486 on REST-meta-MDD, 987 and 1,020 without sign consistency, and 1,334 on both datasets without both criteria. Since τE is fixed at the 80th percentile of the absolute consensus contrasts, removing both criteria keeps the top 20% of the 6,670 AAL-116 edges.
Removing bootstrap stability alone leaves the scaffold unchanged at the default thresholds. With seven training sites and ηE=0.75, an edge that passes sign consistency has at most one source site of the opposite sign. The median of seven resampled sites reverses its sign only if that site is drawn at least four times, which gives πj≈0.99, far above ζE. Bootstrap stability acts only when sign consistency is removed. Edges whose consensus sign is carried by four of the seven sites then have πj≈0.65, and with B=50 about 30% of them pass ζE=0.70.
The table below reports the LOSO AUC and ACC in %. For each seed, we average the metric over all target sites of a dataset, and each entry is the mean ± SD of the three seed-level values.