OSCR

Robust Multi-Site ADHD Classification via GraphSAGE-Based Functional Connectivity Modeling from rs-fMRI.

Code ↔ Paper

The paper beside its authors' code: matches between them have not been computed for this paper yet.

Paper

Loaded from Europe PMC by your browser; OSCR does not store it: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Markdown · 113 lines · 5 KB · no license

  2. # Robust Multi-Site ADHD Classification via GraphSAGE-Based Functional Connectivity Modeling from rs-fMRI
  4. [![Python](https://img.shields.io/badge/Python-3.8%2B-blue.svg)](https://www.python.org/)
  5. [![PyTorch](https://img.shields.io/badge/PyTorch-2.0%2B-orange.svg)](https://pytorch.org/)
  7. ## Title
  8. **Robust Multi-Site ADHD Classification via GraphSAGE-Based Functional Connectivity Modeling from rs-fMRI**
  10. ### Authors
  11. - **Rabab BOUSMAHA**¹* (✉)
  12. - Khouloud MERIBAI¹
  13. - Nardjes BOUCHEMAL²,³
  14. - Naila BOUCHEMAL⁴
  15. - Galina IVANOVA⁵
  17. ---
  19. ### Affiliations
  20. ¹ LabRi Laboratory, Ecole Supérieure en Informatique, Sidi Bel Abbes 22000, Algeria
  21. ² LIRE Laboratory, Abdelhamid Mehri Constantine 2 University, Constantine 25000, Algeria
  22. ³ LISI Laboratory of Intelligent Systems and Informatics, Mila 43000, Algeria
  23. ⁴ LyRIDS ECE Ecole d’Ingénieurs de Paris, 75015 Paris, France
  24. ⁵ Faculty of Electrical Engineering, Electronics and Automation, University of Ruse “Angel Kanchev”, 7017 Ruse, Bulgaria
  26. **Corresponding Author:** Rabab BOUSMAHA — [email hidden]
  28. ## Abstract
  30. Attention Deficit Hyperactivity Disorder (ADHD) is a heterogeneous neurodevelopmental disorder whose diagnosis is mainly based on behavioral assessment and is often delayed due to clinical complexity and limited availability of specialists. Resting-state functional magnetic resonance imaging (rs-fMRI) provides a valuable source of information for supporting automated and objective diagnosis.
  32. This work proposes a **graph-based deep learning framework** for ADHD classification from rs-fMRI that combines **Phase-Locking Value (PLV)** functional connectivity estimation with **GraphSAGE** representation learning. The model effectively captures both regional brain activity and inter-regional interactions while remaining robust to noise and inter-site variability.
  34. The proposed framework was evaluated on the publicly available **ADHD-200** dataset across multiple acquisition sites as well as on a combined multi-site dataset. The model achieved **Accuracy = 0.89**, **AUC = 0.96**, and **Specificity = 0.96** on the combined dataset, outperforming several existing methods.
  36. ---
  38. ## Key Features
  40. - Phase-Locking Value (PLV) based functional connectivity
  41. - 3-layer GraphSAGE with LayerNorm and Dropout
  42. - Data augmentation (Gaussian noise + time shift) applied **only on training subjects**
  43. - Stratified Cross-Validation (5 outer folds)
  44. - Early Stopping + Learning Rate Scheduler
  45. - Optimized decision threshold with minimum recall constraint
  46. - Fully reproducible code
  48. ---
  50. ## Dataset
  52. - **ADHD-200** Multi-site rs-fMRI dataset: The data used in this study are publicly available from the ADHD-200 Preprocessed initiative (Athena pipeline): http://preprocessed-connectomes-project.org/adhd200/
  54. - Due to size constraints, the dataset files (.npy)
  55. are hosted on Google Drive and can be accessed here :
  56. [Download Dataset](https://drive.google.com/drive/folders/1ycOw3TS5qSSJPrFZuF3S4J2oUKSzKGO5?usp=sharing)
  58. - You only need to change the paths of the `.npy` files in the notebook
  60. ---
  61. ## Prerequisites & Packages
  63. ### Required Python Version
  64. - **Python ≥ 3.8**
  66. ### Main Packages
  68. | Package | Version Recommendation |
  69. |-------------------------|------------------------|
  70. | torch | ≥ 2.0 |
  71. | torch-geometric | Latest |
  72. | numpy | ≥ 1.21 |
  73. | pandas | ≥ 1.5 |
  74. | scikit-learn | ≥ 1.2 |
  75. | scipy | ≥ 1.10 |
  76. | networkx | ≥ 3.0 |
  77. | pywavelets (pywt) | ≥ 1.4 |
  78. | tqdm | Latest |
  79. | matplotlib & seaborn | Latest |
  80. | joblib | Latest |
  82. ---
  83. ## How to Run
  85. 1. Clone or download the repository
  86. 2. Open the Jupyter notebook:
  87. ```bash
  88. jupyter notebook ADHD_PLV_GraphSAGE.ipynb
  89. 3. Update the dataset paths in the notebook:Python
  91. TS_NPY = "path/to/timeseries_NeuroIMAGE2.npy"
  92. LAB_NPY = "path/to/labels_NeuroIMAGE2.npy"
  94. 4. Run all cells
  96. Results, best model, and plots will be automatically saved in the Results/ folder.
  98. ## 📁 Project Structure
  100. ├── ADHD_PLV_GraphSAGE.ipynb # Main notebook
  101. ├── README.md # Project documentation
  102. └── Results/ # Generated automatically after running
  103. ├── best_model_weights.pt # Best model weights
  104. ├── scaler.pkl # Fitted scaler
  105. ├── confusion_matrix_global.png # Confusion matrix plot
  106. ├── roc_curve_global.png # ROC curve plot
  107. ├── precision_recall_curve.png # Precision-Recall curve plot
  108. └── final_results_metadata.json # Final results & metrics
  110. ## 📄 Citation
  112. > This paper is currently **under review** at *MDPI Bioengineering* (Manuscript ID: `bioengineering-4271001`).
  113. > A full citation will be provided upon publication.

README.md at commit 1e5f028, no license · at the source

Overview

Authors: Rabab Bousmaha1, Khouloud Meribai1, Nardjes Bouchemal2,3, Naila Bouchemal4, Galina Ivanova5
ORCID iDs: Rabab Bousmaha, Nardjes Bouchemal, Naila Bouchemal, Galina Ivanova
  1. LabRi Laboratory, Ecole Superieure en Informatique, Sidi Bel Abbes 22000, Algeria
  2. LIRE Laboratory, Abdelhamid Mehri Constantine 2 University, Constantine 25000, Algeria
  3. LISI Laboratory of Intelligent Systems and Informatics, Mila 43000, Algeria
  4. LyRIDS, ECE Paris, 10 rue Sextius Michel, 75015 Paris, France
  5. Faculty of Electrical Engineering, Electronics and Automation, University of Ruse “Angel Kanchev”, 7017 Ruse, Bulgaria
Institutions: École Supérieure en Informatique de Sidi Bel Abbes (Algeria); École Centrale d'Électronique (France); Centre Universitaire de Mila (Algeria); Université Constantine 2 (Algeria); Angel Kanchev University of Ruse (Bulgaria)
Journal: Bioengineering (Basel, Switzerland), volume 13, issue 5, article 586
Dates: received 5 April 2026; accepted 15 May 2026; published online 20 May 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3390/bioengineering13050586 · PMID 42194343 · PMCID PMC13203753 · OpenAlex W7161800706
Open access: gold, a free copy (OpenAlex)
Status: code verified (how the code is found and verified)
Categories: fMRI (modality), ADHD (population)
Methods: Spectral & time-frequency, Connectivity, Smoothing, state filtering, decompositions, Machine learning, Preprocessing, Graphs, Complexity, fMRI & imaging, Physiology & signal measures
Keywords: ADHD, multi-site data, resting-state fMRI, functional connectivity, PLV, GraphSAGE, graph-based deep learning
Topic: Functional Brain Connectivity Studies (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: European Commission (BGRRP-2.013-0001)
Citations: not cited yet (Europe PMC); 85 references in the paper

Abstract

Attention Deficit Hyperactivity Disorder (ADHD) is a heterogeneous neurodevelopmental disorder whose diagnosis is mainly based on behavioral assessment and is often delayed due to clinical complexity and limited availability of specialists. Resting-state functional magnetic resonance imaging (rs-fMRI) provides a valuable source of information for supporting automated and objective diagnosis. However, existing studies often do not fully capture the complex interactions of functional connectivity between different brain regions. To address this limitation, this work proposes a graph-based deep learning framework for ADHD classification from rs-fMRI that combines functional connectivity modeling with graph representation learning. The approach used Phase-Locking Value (PLV)-based connectivity estimation and Graph Sample and Aggregate (GraphSAGE) to jointly capture regional brain activity and inter-regional interactions in a scalable and efficient manner. GraphSAGE improves robustness to noise and inter-subject variability by aggregating information from stable local graph neighborhoods. This integration allows the model to learn discriminative connectivity-aware representations while remaining robust to signal variability and adaptable to multi-site data. The proposed framework was evaluated on the publicly available ADHD-200 dataset across multiple acquisition sites as well as on a combined multi-site dataset. The results indicate consistent performance across individual sites and on the combined dataset. The model achieved an Accuracy of 0.89, an AUC of 0.96, and a Specificity of 0.96 on the combined dataset, outperforming several existing methods in this setting. By integrating PLV-based connectivity with GraphSAGE learning, the approach provides an effective and scalable solution for automated ADHD classification from rs-fMRI data, contributing to data-driven approaches for the analysis of neurodevelopmental disorders.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repositories

Its files are read in the Code ↔ Paper reader above.

preprocessed-connectomes-project.org/adhd200

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: the link answers
Software Heritage: not checked
Found in: “Data Availability Statement”
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Availability: 1 check, the latest on 28 September 2026: the link answers (HTTP 200)
  • 28 September 2026: the link answers (HTTP 200)
At the source: preprocessed-connectomes-project.org/adhd200/

Rabab070707/ADHD_PLV_GraphSAGE

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 1e5f028dce808821e2fe8b361e36e64a75396558, 7 June 2026
Size: 2 files, 0 scripts
Software Heritage: not archived
Found in: “Data Availability Statement”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
1 file
At the source: github.com/Rabab070707/ADHD_PLV_GraphSAGE/

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 2 repositories of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 0 scripts, each with its path and the digest of its content;
  • no match between paragraphs and code yet;
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data Availability Statement

The data used in this study are publicly available from the ADHD-200 Preprocessed initiative (Athena pipeline): http://preprocessed-connectomes-project.org/adhd200/ (accessed on 15 February 2025), The code is publicly available at: https://github.com/Rabab070707/ADHD_PLV_GraphSAGE/ (accessed on 13 May 2026).

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

  • Funding: added European Commission: BGRRP-2.013-0001

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 5 authors, 7 keywords, 85 references.

Cite

This paper

Bousmaha, R., Meribai, K., Bouchemal, N., Bouchemal, N., & Ivanova, G. (2026). Robust Multi-Site ADHD Classification via GraphSAGE-Based Functional Connectivity Modeling from rs-fMRI. Bioengineering (Basel, Switzerland), 13(5), 586. https://doi.org/10.3390/bioengineering13050586

BibTeX

@article{bousmaha2026robust,
author = {Bousmaha, Rabab and Meribai, Khouloud and Bouchemal, Nardjes and Bouchemal, Naila and Ivanova, Galina},
title = {{Robust Multi-Site ADHD Classification via GraphSAGE-Based Functional Connectivity Modeling from rs-fMRI}},
journal = {Bioengineering (Basel, Switzerland)},
year = {2026},
month = may,
volume = {13},
number = {5},
pages = {586},
publisher = {Multidisciplinary Digital Publishing Institute (MDPI)},
issn = {2306-5354},
doi = {10.3390/bioengineering13050586},
url = {https://doi.org/10.3390/bioengineering13050586},
pmid = {42194343},
pmcid = {PMC13203753}
}

RIS

TY - JOUR
AU - Bousmaha, Rabab
AU - Meribai, Khouloud
AU - Bouchemal, Nardjes
AU - Bouchemal, Naila
AU - Ivanova, Galina
TI - Robust Multi-Site ADHD Classification via GraphSAGE-Based Functional Connectivity Modeling from rs-fMRI
T2 - Bioengineering (Basel, Switzerland)
J2 - Bioengineering (Basel)
PY - 2026
DA - 2026/05/20
VL - 13
IS - 5
SP - 586
SN - 2306-5354
PB - Multidisciplinary Digital Publishing Institute (MDPI)
DO - 10.3390/bioengineering13050586
UR - https://doi.org/10.3390/bioengineering13050586
LA - en
ER -

CSL-JSON

{
"id": "10.3390/bioengineering13050586",
"type": "article-journal",
"title": "Robust Multi-Site ADHD Classification via GraphSAGE-Based Functional Connectivity Modeling from rs-fMRI",
"container-title": "Bioengineering (Basel, Switzerland)",
"author": [
{
"family": "Bousmaha",
"given": "Rabab"
},
{
"family": "Meribai",
"given": "Khouloud"
},
{
"family": "Bouchemal",
"given": "Nardjes"
},
{
"family": "Bouchemal",
"given": "Naila"
},
{
"family": "Ivanova",
"given": "Galina"
}
],
"container-title-short": "Bioengineering (Basel)",
"volume": "13",
"issue": "5",
"page": "586",
"DOI": "10.3390/bioengineering13050586",
"PMID": "42194343",
"PMCID": "PMC13203753",
"ISSN": "2306-5354",
"publisher": "Multidisciplinary Digital Publishing Institute (MDPI)",
"URL": "https://doi.org/10.3390/bioengineering13050586",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
20
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1186/s12916-023-02941-4
Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry.
Journal: BMC medicine
In common: 9 references
[2] doi:10.1038/s41398-023-02309-5
Multiple measurement analysis of resting-state fMRI for ADHD classification in adolescent brain from the ABCD study.
Journal: Translational psychiatry
In common: ADHD, fMRI, 6 references
[3] doi:10.1016/j.neuroimage.2023.119944
Identification of overlapping and interacting networks reveals intrinsic spatiotemporal organization of the human brain.
Journal: NeuroImage
In common: fMRI, 6 references
[4] doi:10.3390/neurosci5040033 [code]
Resting-State Functional Connectivity Predicts Attention Problems in Children: Evidence from the ABCD Study.
Journal: NeuroSci
In common: ADHD, fMRI, 4 references
[5] doi:10.1162/imag.a.1254 [code]
The individuality of single-frame functional brain connectivity.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: fMRI, 5 references
[6] doi:10.1007/s11682-022-00708-8
Centrality and interhemispheric coordination are related to different clinical/behavioral factors in attention deficit/hyperactivity disorder: a resting-state fMRI study.
Journal: Brain imaging and behavior
In common: ADHD, fMRI, 3 references
[7] doi:10.1002/hbm.26587 [code]
Cortical timescales and the modular organization of structural and functional brain networks.
Journal: Human brain mapping
In common: fMRI, 5 references
[8] doi:10.1162/imag_a_00113 [code]
EiDA: A lossless approach for dynamic functional connectivity; application to fMRI data of a model of ageing.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: fMRI, 4 references
[9] doi:10.1038/s43856-025-01015-1 [code]
Transformer-based structural connectivity networks for ADHD-related connectivity alterations.
Journal: Communications medicine
In common: ADHD, 3 references
[10] doi:10.1038/s41380-025-02974-6 [code]
Deep graph learning of multimodal brain networks defines treatment-predictive signatures in major depression.
Journal: Molecular psychiatry
In common: fMRI, 4 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Add the badge to the code's README

OSCR: the badge of this paper

If you are an author or a maintainer of this paper's code, you can add this badge. It says what OSCR knows of the paper (its code, its validated tracing map) and links the code to this page. Paste one of these snippets under the README's title. Keep the two marks around it, so that the badge can be replaced in place and is never added twice. OSCR changes nothing in your repository.

Markdown
<!-- oscr-badge:start -->
[![OSCR](https://openscicode.org/badge/doi/10.3390/bioengineering13050586.svg)](https://openscicode.org/badge/doi/10.3390/bioengineering13050586)
<!-- oscr-badge:end -->
reStructuredText
.. oscr-badge:start

.. image:: https://openscicode.org/badge/doi/10.3390/bioengineering13050586.svg
   :target: https://openscicode.org/badge/doi/10.3390/bioengineering13050586
   :alt: OSCR

.. oscr-badge:end
HTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.3390/bioengineering13050586"><img src="https://openscicode.org/badge/doi/10.3390/bioengineering13050586.svg" alt="OSCR"></a>
<!-- oscr-badge:end -->

On GitHub, open the README, choose Edit (the pencil), paste the snippet under the title, then "Commit changes".

If you choose to, you can have the badge kept up to date for you, with oscr badge --write from the command line or with the GitHub Action in your repository's workflows. Every other form and option (a card, AsciiDoc, LaTeX, shields.io) is on the badge page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.

Report a mismatch between the code and the paper · Report a code error (signed in)

Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.

Report a reproduction failure (signed in)

Versions of its code: the releases of its code tied to a version of this paper (the preprint, the accepted manuscript, the version of record…), each with the tracing map versioned with it.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.