OSCR

Bias and generalizability of brain age prediction models: A multi-cohort evaluation with anatomical and interpretability insights.

Code ↔ Paper

9 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 9 matches · 6 of them tie a paragraph to a whole file, not to given lines: weak matches, whose lines are not tinted
  1. [1] § Materials and Methods › Data pre-processing ↔ preprocessing/BAN_preproc_pipeline.sh, the whole file · a weak match · score 0.94 · N4 bias field, antsRegistrationSyN, skull stripping, rigid registration, preprocessing pipeline, FreeSurfer
  2. [2] § Materials and Methods › Data pre-processing ↔ preprocessing/DBN_preproc_pipeline_(synthstrip).sh, the whole file · a weak match · score 0.94 · N4 bias field, antsRegistrationSyN, skull stripping, rigid registration, preprocessing pipeline, FreeSurfer
  3. [3] § Materials and Methods › Data pre-processing ↔ preprocessing/DBN_preproc_pipeline_(synthstrip).sh, the whole file · a weak match · score 0.61 · skull stripping, DeepBrainNet, FreeSurfer, SynthStrip, pipeline
  4. [4] § Materials and Methods › Data pre-processing ↔ preprocessing/preprocessing_FLIRT_vs_ANTs/FLIRT_affine.sh, the whole file · a weak match · score 0.60 · SynthStrip, skull stripping, FreeSurfer
  5. [5] § Results › Interpretability ↔ results/LRP_saliency_maps/plot_maps.ipynb, lines 74–125 · score 0.59 · coronal views, relevance maps, UNSAM_LC, sagittal, slices, LRP
  6. [6] § Results › Bias and robustness › Age- and dataset-related bias ↔ preprocessing/preprocessing_FLIRT_vs_ANTs/FLIRT_affine.sh, the whole file · a weak match · score 0.55 · SynthStrip, FSL, FreeSurfer, preprocessing, bias, brain
  7. [7] § Results › Bias and robustness › Age- and dataset-related bias ↔ preprocessing/preprocessing_FLIRT_vs_ANTs/FLIRT_rigid.sh, the whole file · a weak match · score 0.55 · SynthStrip, FSL, FreeSurfer, preprocessing, bias, brain
  8. [8] § Materials and Methods › Datasets ↔ utils/Image_quality_metrics/plot_iqms.ipynb, lines 185–200 · score 0.53 · quality metrics, image quality, EFC, CNR, error
  9. [9] § Materials and Methods › Model evaluation strategy › Accuracy evaluation ↔ results/scatter+box.ipynb, lines 93–151 · score 0.51 · chronological age, predicted brain age, identity, scatter, CN

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Shell · 37 lines · 1.1 KB · MIT · 2 matches

  1. #!/bin/bash
  3. INPUT_DIR="/data/Lautaro/Documentos/BrainAgeCOVID/DATOS/Raw_T1/FULL_ADNI_images"
  4. OUTPUT_DIR="/data/Lautaro/Documentos/BrainAgeCOVID/DATOS/Preprocessed/DeepBrainNet/ADNI"
  5. mkdir -p "$OUTPUT_DIR"
  7. MNI_TEMPLATE="/usr/local/fsl/data/standard/MNI152_T1_1mm_brain.nii.gz"
  9. shopt -s nullglob
  11. for nii in "$INPUT_DIR"/*.nii "$INPUT_DIR"/*.nii.gz; do
  12. base=$(basename "$nii")
  13. base_noext="${base%.nii.gz}"
  14. base_noext="${base_noext%.nii}"
  16. echo "🔄 Preprocessing: $base"
  18. # 1. N4 Bias Field Correction
  19. N4_OUT="${OUTPUT_DIR}/${base_noext}_n4.nii.gz"
  20. N4BiasFieldCorrection -i "$nii" -o "$N4_OUT"
  22. # 2. Skull stripping
  23. STRIP_OUT="${OUTPUT_DIR}/${base_noext}_brain.nii.gz"
  24. MASK_OUT="${OUTPUT_DIR}/${base_noext}_mask.mgz"
  25. mri_synthstrip -i "$N4_OUT" -o "$STRIP_OUT" -m "$MASK_OUT"
  27. # 3. Rigid registration with ANTs
  28. OUT_PREFIX="${OUTPUT_DIR}/${base_noext}_ANTS"
  29. antsRegistrationSyN.sh -d 3 \
  30. -f "$MNI_TEMPLATE" \
  31. -m "$STRIP_OUT" \
  32. -o "$OUT_PREFIX" \
  33. -t a \
  34. -n 8
  35. done
  37. echo "✅ Procesamiento completado en: $OUTPUT_DIR"

DBN_preproc_pipeline_(synthstrip).sh at commit 6ba6f92, under MIT · at the source

Overview

Authors: Lautaro J. Aguzin Parrilli1, Martin A. Belzunce1,2,3
ORCID iDs: Martin A. Belzunce
  1. Centro Universitario de Imágenes Médicas (CEUNIM), Escuela de Ciencia y Tecnología, Universidad Nacional de Gral. San Martín, Buenos Aires, Argentina
  2. Instituto de Ciencias Físicas (ICIFI UNSAM-CONICET), Escuela de Ciencia y Tecnología, Universidad Nacional de Gral. San Martín (UNSAM), Buenos Aires, Argentina
  3. Consejo Nacional de Investigaciones Científicas y Tecnológicas (CONICET), Buenos Aires, Argentina
Institutions: National University of General San Martín (Argentina); Consejo Nacional de Investigaciones Científicas y Técnicas (Argentina)
Journal: Imaging neuroscience (Cambridge, Mass.), volume 4, article IMAG.a.1164
Dates: received 21 September 2025; accepted 11 February 2026; published online 12 March 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1162/imag.a.1164 · PMID 41836919 · PMCID PMC12983579 · OpenAlex W7130340323
Open access: diamond, a free copy (OpenAlex)
Status: code verified
Categories: structural MRI / diffusion (modality), human (organism), Alzheimer's / dementia (population)
Methods: Connectivity, Statistics, Machine learning, fMRI & imaging, Preprocessing
Keywords: brain age prediction, brain age gap (BAG), deep learning, neuroimaging, aging, interpretability
Topic: Functional Brain Connectivity Studies (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: Fondo de Innovación Tecnológica de Buenos Aires, Ministerio de Producción, Ciencia e Innovación Tecnológica de la Provincia de Buenos Aires, Argentina (FITBA-B06); Agencia Nacional de Promoción Científica y Tecnológica, Argentina (PICT-PRH-2022-01); IBRO-Wellcome Neuroscience Capacity Accelerator for Mental Health (NCAMH)
Citations: not cited yet (Europe PMC); 38 references in the paper

Abstract

Brain age prediction from T1-weighted MRI and its associated brain age gap (BAG) has emerged as a promising neuroimaging biomarker for assessing deviations from normative aging. However, the robustness, bias, and interpretability of existing models across external datasets remain poorly understood, limiting clinical translation. In this study, we evaluated four publicly available brain age models (ENIGMA, DeepBrainNet, Pyment, and BrainAgeNeXt) across four independent MRI datasets (ADNI, UNSAM Long COVID, and two OpenNeuro cohorts), comprising 1,634 subjects with diverse demographic and clinical profiles. Models were tested using their original preprocessing pipelines, and performance was assessed using mean absolute error (MAE), mean error (ME), and BAG variability metrics, with additional analyses of biases related to age, dataset, ethnicity, and education. Interpretability was evaluated using Layer-wise Relevance Propagation, and anatomical correlates were explored using BrainChart-derived centile scores. Group-level comparisons were performed between cognitively normal (CN) individuals and patients with Mild Cognitive Impairment (MCI), Alzheimer’s disease (AD), or Long COVID (LC). Models based on 3D convolutional neural networks (Pyment and BrainAgeNeXt) outperformed the DeepBrainNet 2D CNN and the ENIGMA ridge regression model in both accuracy (MAE: 3.9–3.7 vs. 6.2–12.4 years respectively) and stability (ASTD: 3.2–2.9 vs. 4.6–8.3 years). Dataset-specific BAG differences were largely explained by age distributions, whereas ethnicity showed a statistically significant but small effect on BAG in some models. Relevance maps highlighted the lateral ventricles as the most consistently relevant anatomical region, with additional cerebellar contributions emerging in older adults for BrainAgeNeXt. Group-level analyses confirmed elevated BAG in MCI and AD patients compared to CN, while no significant differences were observed in Long COVID participants. These findings suggest that, while BAG is a promising biomarker for group-level analyses, current models are required to address age and demographic biases to enable individual-level clinical application.

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

Repositories

Its files are read in the Code ↔ Paper reader above, with 9 matches between paragraphs and lines of code.

vishnubashyam/deepbrainnet

License: none: the authors keep all their rights
State: the link answers, verified on 26 September 2026
Evidence: files inventoried
Commit: 505f4e4fdda6a2f5773bec0c98ef567a7a4ae36e, 4 September 2020
Languages: Python (2), Shell (1)
Size: 11 files, 3 scripts
Software Heritage: not archived
Found in: the text, “DeepBrainNet”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (2 files), pandas (2 files), Keras (1 file), NiBabel (1 file), Pillow (1 file), scikit-learn (1 file)
Availability: 1 check, the latest on 26 September 2026: the link answers
  • 26 September 2026: the link answers
4 files
At the source: github.com/vishnubashyam/DeepBrainNet/tree/master

mabelzunce/brainage-models-benchmark

License: MIT
State: the link answers, verified on 30 September 2026
Evidence: files inventoried
Commit: 6ba6f923d3478e7d5ae1480ee025ad0c189f9ccb, 3 April 2026
Languages: Python (141), Jupyter (15), Shell (13)
Size: 37,872 files, 169 scripts
Software Heritage: not archived
Found in: “Data and Code Availability”
Holds: README, license file
Not found: CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (16 files), Matplotlib (13 files), pandas (13 files), seaborn (9 files), ANTs (7 files), FreeSurfer (7 files), NiBabel (5 files), SciPy (5 files), statsmodels (5 files), FSL (4 files), Plotly (3 files), Nilearn (2 files), PyTorch (2 files), scikit-learn (2 files), Keras (1 file), MONAI (1 file), Pillow (1 file), TensorFlow (1 file)
Availability: 1 check, the latest on 30 September 2026: the link answers
  • 30 September 2026: the link answers
171 files
At the source: github.com/mabelzunce/brainage-models-benchmark

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;
  • 172 scripts, each with its path and the digest of its content;
  • 9 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • 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

Datasets cited

Data and Code Availability

All code used for image preprocessing, model evaluation, and statistical analysis, as well as the resulting brain age predictions for all models and datasets, are publicly available at: https://github.com/mabelzunce/brainage-models-benchmark

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 1, 30 September 2026: the first record

Recorded: type, language, journal, volume, pages, dates, 2 authors, 6 keywords, 3 funders, 38 references.

Cite

This paper

Aguzin Parrilli, L. J., & Belzunce, M. A. (2026). Bias and generalizability of brain age prediction models: A multi-cohort evaluation with anatomical and interpretability insights. Imaging neuroscience (Cambridge, Mass.), 4, IMAG.a.1164. https://doi.org/10.1162/imag.a.1164

BibTeX

@article{aguzinparrilli2026bias,
author = {Aguzin Parrilli, Lautaro J. and Belzunce, Martin A.},
title = {{Bias and generalizability of brain age prediction models: A multi-cohort evaluation with anatomical and interpretability insights}},
journal = {Imaging neuroscience (Cambridge, Mass.)},
year = {2026},
month = mar,
volume = {4},
pages = {IMAG.a.1164},
publisher = {MIT Press},
issn = {2837-6056},
doi = {10.1162/imag.a.1164},
url = {https://doi.org/10.1162/imag.a.1164},
pmid = {41836919},
pmcid = {PMC12983579}
}

RIS

TY - JOUR
AU - Aguzin Parrilli, Lautaro J.
AU - Belzunce, Martin A.
TI - Bias and generalizability of brain age prediction models: A multi-cohort evaluation with anatomical and interpretability insights
T2 - Imaging neuroscience (Cambridge, Mass.)
J2 - Imaging Neurosci (Camb)
PY - 2026
DA - 2026/03/12
VL - 4
SP - IMAG.a.1164
SN - 2837-6056
PB - MIT Press
DO - 10.1162/imag.a.1164
UR - https://doi.org/10.1162/imag.a.1164
LA - en
ER -

CSL-JSON

{
"id": "10.1162/imag.a.1164",
"type": "article-journal",
"title": "Bias and generalizability of brain age prediction models: A multi-cohort evaluation with anatomical and interpretability insights",
"container-title": "Imaging neuroscience (Cambridge, Mass.)",
"author": [
{
"family": "Aguzin Parrilli",
"given": "Lautaro J."
},
{
"family": "Belzunce",
"given": "Martin A."
}
],
"container-title-short": "Imaging Neurosci (Camb)",
"volume": "4",
"page": "IMAG.a.1164",
"DOI": "10.1162/imag.a.1164",
"PMID": "41836919",
"PMCID": "PMC12983579",
"ISSN": "2837-6056",
"publisher": "MIT Press",
"URL": "https://doi.org/10.1162/imag.a.1164",
"language": "en",
"issued": {
"date-parts": [
[
2026,
3,
12
]
]
}
}

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:10.1038/s41398-026-04081-8 [code]
Functional system-specific brain aging across the Alzheimer's disease continuum.
Journal: Translational psychiatry
In common: Keras, FreeSurfer, TensorFlow, 8 other tools, Alzheimer's / dementia, structural MRI / diffusion, 3 references
[2] doi:10.1002/hbm.70425 [code]
Brain Age Estimation on T2-FLAIR Scans for Application to Multiple Sclerosis.
Journal: Human brain mapping
In common: MONAI, FSL, PyTorch, 3 other tools, structural MRI / diffusion, 7 references
[3] doi:10.1186/s40708-026-00316-y [code]
Generalizable and explainable deep learning for brain MRI: a multi-cohort evaluation of 3D architectures for age and sex prediction.
Journal: Brain informatics
In common: MONAI, PyTorch, seaborn, 5 other tools, structural MRI / diffusion, 6 references
[4] doi:10.1002/alz.71649 [code]
Postmortem brain MRI reveals differential associations of subcortical and limbic volumes with cortical thinning and neurodegenerative pathologies.
Journal: Alzheimer's & dementia : the journal of the Alzheimer's Association
In common: ANTs, FreeSurfer, FSL, 11 other tools, Alzheimer's / dementia, structural MRI / diffusion
[5] doi:10.21203/rs.3.rs-9326213/v1 [code]
Multi-task fMRI outperforms resting-state fMRI for revealing task-invariant organization of the human brain
Journal: Research Square (preprint)
In common: ANTs, FreeSurfer, FSL, 11 other tools
[6] doi:10.64898/2026.03.09.710558 [code]
Multi-task fMRI outperforms resting-state fMRI for revealing task-invariant organization of the human brain
Journal: bioRxiv (preprint)
In common: ANTs, FreeSurfer, FSL, 11 other tools
[7] doi:10.1371/journal.pbio.3003856 [code]
Aging and metabolism contribute separately to brain-body health.
Journal: PLoS biology
In common: ANTs, FreeSurfer, FSL, 10 other tools, structural MRI / diffusion, 1 reference
[8] doi:10.1162/imag.a.1352 [code]
Brain-age in ultra-low-field MRI: How well does it work?
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: MONAI, ANTs, FreeSurfer, 6 other tools, structural MRI / diffusion, 3 references
[9] doi:10.1038/s41586-026-10631-3 [code]
A prognostic human brain network for diffuse midline glioma.
Journal: Nature
In common: Keras, ANTs, FreeSurfer, 10 other tools
[10] doi:10.1016/j.xcrm.2026.102943 [code]
Parent-of-origin effects in Alzheimer's liability dissociate neurocognitive and cardiovascular traits in at-risk individuals.
Journal: Cell reports. Medicine
In common: ANTs, FreeSurfer, FSL, 10 other tools, Alzheimer's / dementia

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.

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: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

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