OSCR

Serial amplitude-integrated electroencephalography trajectories and brain-injury-associated functional immaturity in extremely preterm infants.

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, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Markdown · 214 lines · 9.3 KB · MIT

  1. # Serial aEEG maturation trajectories and discrimination of brain injury in extremely preterm infants
  3. Single-centre retrospective longitudinal cohort study of **extremely preterm infants (gestational age <28 weeks)** with serial bedside amplitude-integrated EEG (aEEG) Burdjalov scoring at postmenstrual age (PMA) 26, 28, 30, 32, 34, and 36 weeks. The analysis compares maturation trajectories and discrimination between the **reference / no major brain injury** group and the **brain injury** group.
  5. This repository provides a reproducible statistical analysis Jupyter notebook, a publication-style figure script, and exported tables (CSV) and figures (PNG/PDF).
  7. ---
  9. ## Study overview
  11. | Item | Description |
  12. |------|-------------|
  13. | Design | Single-centre, retrospective, longitudinal repeated measures |
  14. | Participants | 156 extremely preterm infants (GA <28 weeks) |
  15. | Longitudinal records | **723** aEEG assessments (after cleaning; 731 non-missing records before removal of 8 exact duplicates) |
  16. | Groups | Reference / no major brain injury **n = 109**; Brain injury **n = 47** |
  17. | Time points | PMA 26, 28, 30, 32, 34, 36 weeks (≥3 time points per infant) |
  18. | Primary marker | Burdjalov total score and four subscores (continuity, sleep–wake cycling, lower-border amplitude, bandwidth) |
  19. | Primary outcome | Imaging-defined brain injury |
  21. ### Research questions
  23. 1. How do Burdjalov total and subdomain scores change across PMA 26–36 weeks?
  24. 2. Do maturation trajectories differ between the brain injury and reference groups (main effect and PMA × group interaction)?
  25. 3. At which PMA window does the total score best discriminate brain injury (ROC/AUC)?
  26. 4. Which subscores show persistent between-group differences across PMA windows?
  28. ### Statistical framework (locked version v3.2)
  30. - **Primary model**: Random-intercept linear mixed-effects model (LMM) with PMA centred at 32 weeks
  31. `burdjalov_total ~ pma_c * group + (1 | patient_id)`
  32. - **Random-slope model**: Explored as a covariance-structure diagnostic only; not used for inference when non-converged; no misleading AIC/BIC/LRT comparisons
  33. - **Sensitivity analyses**: Complete case, ≥4 PMA time points, categorical-PMA model, adjustment for GA/birth weight/sex, GEE (exchangeable correlation + robust SE), **SGA-adjusted models** (WS/T 800—2022 primary; Fenton 2025 sensitivity)
  34. - **SGA**: Derived from birth weight, sex, and completed gestational age using external percentile tables (primary: WS/T 800—2022; one infant at 23+6 weeks outside the primary standard range)
  36. ---
  38. ## Repository structure
  40. ```
  41. .
  42. ├── README.md # This file
  43. ├── requirements.txt # Python dependencies (recommended versions)
  44. ├── make_figures_v3.py # Standalone publication-style figure script
  45. ├── notebooks/
  46. │ ├── _build_notebook.py # Programmatic generator for the analysis notebook
  47. │ └── aEEG_analysis.ipynb # Main analysis notebook (v3.2 locked-candidate)
  48. ├── data/ # Input data (de-identify before public release)
  49. │ ├── aEEG_cleaned_longitudinal_dataset.xlsx
  50. │ ├── aEEG_analysis_ready_long.csv
  51. │ └── SGA_reference_updated_WS_T_800_2022_Fenton_2025.xlsx
  52. ├── results/ # Analysis outputs (CSV, Methods text, etc.)
  53. └── figures/ # Analysis and manuscript figures (PNG + PDF)
  54. ```
  56. ### Key code files
  58. | File | Role |
  59. |------|------|
  60. | `notebooks/aEEG_analysis.ipynb` | **Main pipeline**: data QC, SGA, descriptive statistics, LMM/GEE, ROC, sensitivity analyses, export tables and figures |
  61. | `notebooks/_build_notebook.py` | Notebook builder; run after editing analysis logic to regenerate `.ipynb` |
  62. | `make_figures_v3.py` | **Standalone figure script**: journal-style main and supplementary figures (reads the same `data/` and `results/` as the notebook) |
  64. ### Main notebook workflow (top to bottom)
  66. 1. Configuration and paths (`NOTEBOOK_VERSION`, `B8_ACTION`, SGA reference file)
  67. 2. Load cleaned data and SGA reference table
  68. 3. Variable mapping and QC (score ranges, B8 anomaly handling, total-score recalculation, missingness)
  69. 4. SGA classification (WS/T 800—2022 + Fenton 2025)
  70. 5. Table 1 baseline characteristics; Tables 2–4 by PMA
  71. 6. LMM for total and subscores; model comparison (random intercept vs exploratory random slope)
  72. 7. ROC by PMA; sensitivity analyses (including SGA adjustment)
  73. 8. Export to `results/` and `figures/`; Methods paragraph and manuscript bullets
  74. 9. Final locking checklist (`qc_locking_checklist.csv`)
  76. ### Outputs from `make_figures_v3.py`
  78. | File | Content |
  79. |------|---------|
  80. | `fig2_trajectory_combined` | **Combined trajectory figure**: Panel A total score + Panels B–E subdomains (single manuscript figure) |
  81. | `fig4_pma_forest` | PMA-specific between-group forest plot (from `table5c_group_effect_by_pma.csv`) |
  82. | `fig5_roc_curves` | ROC curves by PMA (AUC locked to `table7_roc_by_pma.csv`) |
  83. | `fig6_reference_template` | Reference-group maturation heatmap (colours normalised to theoretical subdomain maxima) |
  84. | `figS1_missingness` | Variable-level missingness |
  85. | `figS2_spaghetti` | Individual trajectories with group-median overlay |
  86. | `figS3_timepoints_distribution` | Distribution of distinct PMA time points per infant |
  88. > **Figure 1** (cohort/STROBE flow diagram) is prepared separately by the authors and is not generated by `make_figures_v3.py`.
  90. ---
  92. ## Environment and dependencies
  94. ### Python version
  96. Developed and validated on:
  98. - **Python 3.12.3** (recommended **≥3.10**)
  100. ### Core Python packages (versions tested locally)
  102. | Package | Version | Use |
  103. |---------|---------|-----|
  104. | pandas | 2.3.3 | Data I/O and wrangling |
  105. | numpy | 1.26.4 | Numerical computing |
  106. | scipy | 1.16.3 | Non-parametric tests, distributions |
  107. | statsmodels | 0.14.5 | LMM, GEE |
  108. | scikit-learn | 1.7.2 | ROC/AUC |
  109. | matplotlib | 3.9.2 | Plotting |
  110. | seaborn | 0.13.2 | Statistical plot styling |
  111. | openpyxl | 3.1.5 | Excel I/O |
  112. | nbformat | — | Notebook generation (`_build_notebook.py` only) |
  113. | jupyter / nbconvert | — | Notebook execution (optional) |
  114. | Pillow | — | Image export helper (optional TIFF conversion) |
  116. Install:
  118. ```bash
  119. pip install -r requirements.txt
  120. ```
  122. ---
  124. ## Quick start
  126. ### 1. Clone the repository and prepare data
  128. Place cleaned data files in `data/` (see file list above). **Do not upload raw clinical data with identifiers** to a public repository; use de-identified data or synthetic examples where appropriate.
  130. ### 2. Run the main analysis notebook
  132. ```bash
  133. # Optional: regenerate the notebook from _build_notebook.py
  134. python notebooks/_build_notebook.py
  136. # Run top-to-bottom (requires Jupyter)
  137. jupyter notebook notebooks/aEEG_analysis.ipynb
  138. ```
  140. Or execute headlessly with nbconvert:
  142. ```bash
  143. jupyter nbconvert --to notebook --execute notebooks/aEEG_analysis.ipynb \
  144. --ExecutePreprocessor.timeout=600
  145. ```
  147. Outputs are written to `results/` and `figures/`.
  149. ### 3. Generate publication-style figures
  151. ```bash
  152. python make_figures_v3.py
  153. ```
  155. Figures are saved to `figures/` (PDF + 600 dpi PNG).
  157. ### 4. Key configuration (notebook header)
  159. | Parameter | Description |
  160. |-----------|-------------|
  161. | `NOTEBOOK_VERSION` | Currently `v3.2_locked_candidate` |
  162. | `B8_ACTION` | Handling of the B8/PMA 36 out-of-range record: `mean_imputed_pma36` (default), `source_corrected_to_2`, `exclude_record`, `flag_only_not_locked` |
  163. | `SGA_REFERENCE_FILE` | Path to SGA percentile reference table |
  165. ---
  167. ## Main output files
  169. ### Primary tables (`results/`)
  171. - `table1_baseline.csv` — Baseline characteristics (including WS/T 800—2022 SGA)
  172. - `table2_records_by_pma.csv` — Record counts by PMA
  173. - `table3_total_by_pma.csv` / `table4_subscores_by_pma.csv` — Descriptive statistics by PMA
  174. - `table5_lmm_total.csv` — Primary LMM fixed effects
  175. - `table5c_group_effect_by_pma.csv` — Group effects at each PMA from the categorical-PMA model
  176. - `table7_roc_by_pma.csv` — ROC/AUC by PMA
  177. - `table_sensitivity_lmm.csv` — Sensitivity analysis summary
  179. ### QC and supplementary (`results/`)
  181. - `qc_sga_classification.csv`, `qc_locking_checklist.csv`
  182. - `qc_out_of_range_before_action.csv`, `qc_score_correction_log.csv`
  183. - `supplementary_table_S2_timepoints_per_infant.csv`
  184. - `supplementary_table_S3_sga_definition.csv`
  185. - `methods_paragraph.txt`, `manuscript_bullets.txt`
  187. ---
  189. ## Data and ethics
  191. - Data originate from clinical NICU aEEG and chart records. **Identifiable individual-level data must not be shared without institutional review board approval and appropriate data-use agreements.**
  192. - For public GitHub release, upload analysis code and de-identified or aggregated outputs only, or provide a data-access statement. Do not upload raw collection workbooks (e.g. original hospital identifiers).
  194. ---
  196. ## Version history (summary)
  198. | Version | Highlights |
  199. |---------|------------|
  200. | v3.2 | SGA (WS/T 800—2022 + Fenton 2025), configurable B8 QC, locking checklist, random slope exploratory only |
  201. | v3.1 | Consistent narrative for non-converged random-slope model; sensitivity table and Methods aligned |
  202. | v3 figures | `make_figures_v3.py`: journal style, combined Fig 2+3, forest Fig 4, normalised Fig 6 |
  204. ---
  206. ## Citation and licence
  208. Add the formal citation once the manuscript is published. Choose a code licence (e.g. MIT) and add a `LICENSE` file before public release.
  210. ---
  212. ## Contact
  214. (Add corresponding author or maintainer email / ORCID before uploading to GitHub.)

ReadMe.md at commit ac7c29c, under MIT · at the source

Overview

Authors: Yong Zhang1, Dekai Xu1, Guoyong Luo1, Yongping Xu1, Yujing Yang1, Yong Ji1
  1. Neonatal Intensive Care Unit, Children’s Hospital of Shanxi Province, Taiyuan, Shanxi, China
Institutions: Shanxi Provincial Children's Hospital (China)
Journal: Frontiers in pediatrics, volume 14, article 1912111
Dates: received 17 June 2026; accepted 12 August 2026; published online 26 August 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3389/fped.2026.1912111 · PMID 42718659 · PMCID PMC13553765 · OpenAlex W7204236925
Open access: gold, a free copy (OpenAlex)
Status: empty repository
Categories: EEG (modality), human (organism), other condition (population), developmental (subfield)
Methods: Spectral & time-frequency, Connectivity, Statistics
Keywords: amplitude-integrated electroencephalography, brain injury, Burdjalov maturation score, extremely preterm infants, intraventricular hemorrhage, longitudinal cohort, postmenstrual age, white-matter injury
Topic: Neonatal and fetal brain pathology (Pediatrics, Perinatology and Child Health, Medicine), according to OpenAlex
Citations: not cited yet (Europe PMC); 38 references in the paper

Abstract

Objective: To characterize postmenstrual-age (PMA)-specific Burdjalov maturation trajectories in extremely preterm infants and to evaluate how imaging-defined major brain injury was associated with these trajectories and with apparent within-cohort discrimination between imaging-defined groups across PMA windows.

Methods: We conducted a single-center retrospective cohort study of 156 infants born before 28 weeks’ gestation, comprising 109 infants without major brain injury (reference group) and 47 with brain injury, defined as grade III–IV intraventricular hemorrhage or white-matter injury. These infants contributed 723 analyzable serial amplitude-integrated electroencephalography (aEEG) recordings across PMA 26 to 36 weeks. Burdjalov scores were analyzed using linear mixed-effects models, categorical-PMA contrasts, a patient-clustered ordinal model, complete time-window sensitivity analyses, and joint patient-level bootstrap comparisons of PMA-specific areas under the curve.

Results: The Burdjalov total score increased by 0.92 points per week of PMA. Infants with brain injury scored on average 1.45 points lower at PMA 32 weeks (95% confidence interval −1.65 to −1.25). Although the continuous-scale model produced a positive PMA-by-group interaction, its magnitude and statistical significance varied across analysis windows and it was not reproduced by the ordinal model (p = 0.979). Categorical-PMA estimates showed a non-monotonic pattern across the full observation period: the imprecise difference at PMA 26 weeks was followed by the largest observed difference at PMA 28 weeks and progressively smaller differences thereafter. At PMA 36 weeks, 87.4% of reference-group and 27.3% of brain-injury recordings reached the maximum score. The largest observed area under the curve was at PMA 32 weeks (0.886, 95% confidence interval 0.824 to 0.940), but joint patient-level bootstrap comparisons did not distinguish PMA 32 from PMA 28, 30, or 36 weeks.

Conclusions: Imaging-defined major brain injury was associated with persistently lower Burdjalov scores. The unstable linear interaction on this bounded scale should not be interpreted as biological recovery or catch-up maturation. The retrospective within-cohort discrimination estimates do not establish predictive, diagnostic, screening, or clinical decision utility.

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

Repository

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

MdXuDeKai/AEEG

License: MIT
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: ac7c29c7b31b308fec02977d7c7de61acf44153c, 17 June 2026
Size: 2 files, 0 scripts
Software Heritage: not archived
Found in: “Data availability statement”
Holds: README, environment (requirements.txt)
Not found: license file, CITATION.cff, tests, continuous integration, documentation
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
1 file
At the source: github.com/MdXuDeKai/AEEG

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:

  • 1 repository 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.

Cohort characteristics and aEEG data availability

Of 274 screened infants, 98 did not contribute analyzable longitudinal data: life-sustaining treatment was withdrawn at the family's request in 49, 8 died despite full intensive care, and 41 did not accumulate three qualifying recordings. In 14 of these 41 infants, data collection closed before a third PMA window was reached; in the remaining 27, the reason could not be reliably reconstructed. The remaining 176 infants underwent formal eligibility assessment. After exclusion of 20 infants (necrotizing enterocolitis during the monitoring period, n = 2; intracranial infection during the monitoring period, n = 2; clinical or electrographic seizures, n = 1; incomplete clinical or imaging data, n = 15), 156 infants formed the final analytic cohort (Figure 1 and Supplementary Table S19). The cohort comprised 109 reference/no major brain injury infants and 47 brain-injury infants, including 21 with grade III–IV intraventricular hemorrhage and 26 with white-matter injury. After removal of 8 exact-duplicate records from 731 eligible longitudinal recording entries, the analytic dataset comprised 723 recordings: 27 at PMA 26 weeks, 137 at 28 weeks, 139 at 30 weeks, 152 at 32 weeks, 148 at 34 weeks, and 120 at 36 weeks (Supplementary Table S2).

Baseline perinatal characteristics did not differ significantly between the brain-injury and reference groups (all p > 0.05; Table 1), including gestational age, birth weight, sex, antenatal corticosteroid exposure, mode of delivery, assisted reproduction, and the small-for-gestational-age proportion by the WS/T 800–2022 classification (p = 0.296).

Within the reference group, 9/109 infants (8.3%) had normal imaging, 3/109 (2.8%) had grade I intraventricular hemorrhage, and 97/109 (89.0%) had grade II intraventricular hemorrhage (Supplementary Table S18). Postnatal systemic sepsis was recorded in 43/109 reference infants (39.4%) and 19/47 brain-injury infants (40.4%); maternal infection/chorioamnionitis was recorded in 6/109 (5.5%) and 3/47 (6.4%), respectively (both Fisher exact p = 1.000).

The reliability sample comprised 219 recordings from 45 infants. For the total score, inter-rater exact agreement was 91.8%, agreement within one point was 98.6%, weighted κ was 0.994 (95% cluster-bootstrap CI 0.990 to 0.997), and ICC(A,1) was 0.994. Inter-rater reliability results for all subscores are reported in Supplementary Table S17.

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

Data availability statement

The analysis code supporting the findings of this study is publicly available at https://github.com/MdXuDeKai/AEEG. De-identified aggregate analytic outputs are available from the corresponding authors upon reasonable request. Individual-level data are not publicly available because of institutional privacy restrictions and Chinese health-data regulations. Further inquiries can be directed to the corresponding authors.

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

Software and data availability

All analyses were performed in Python 3.12.3 using statsmodels 0.14.5 for the linear mixed-effects model and the generalized estimating equations (26), scikit-learn 1.7.2 for the receiver-operating characteristic analysis (27), pandas 2.3.3, and numpy 1.26.4. The analysis code is publicly available at https://github.com/MdXuDeKai/AEEG. De-identified aggregate analytic outputs are available from the corresponding author upon reasonable request; individual-level data are not publicly available because of institutional privacy restrictions and Chinese health-data regulations.

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, 27 September 2026: the first record

Recorded: type, language, journal, volume, pages, dates, 6 authors, 8 keywords, 35 references.

Cite

This paper

Zhang, Y., Xu, D., Luo, G., Xu, Y., Yang, Y., & Ji, Y. (2026). Serial amplitude-integrated electroencephalography trajectories and brain-injury-associated functional immaturity in extremely preterm infants. Frontiers in pediatrics, 14, 1912111. https://doi.org/10.3389/fped.2026.1912111

BibTeX

@article{zhang2026serial,
author = {Zhang, Yong and Xu, Dekai and Luo, Guoyong and Xu, Yongping and Yang, Yujing and Ji, Yong},
title = {{Serial amplitude-integrated electroencephalography trajectories and brain-injury-associated functional immaturity in extremely preterm infants}},
journal = {Frontiers in pediatrics},
year = {2026},
month = aug,
volume = {14},
pages = {1912111},
publisher = {Frontiers Media SA},
issn = {2296-2360},
doi = {10.3389/fped.2026.1912111},
url = {https://doi.org/10.3389/fped.2026.1912111},
pmid = {42718659},
pmcid = {PMC13553765}
}

RIS

TY - JOUR
AU - Zhang, Yong
AU - Xu, Dekai
AU - Luo, Guoyong
AU - Xu, Yongping
AU - Yang, Yujing
AU - Ji, Yong
TI - Serial amplitude-integrated electroencephalography trajectories and brain-injury-associated functional immaturity in extremely preterm infants
T2 - Frontiers in pediatrics
J2 - Front Pediatr
PY - 2026
DA - 2026/08/26
VL - 14
SP - 1912111
SN - 2296-2360
PB - Frontiers Media SA
DO - 10.3389/fped.2026.1912111
UR - https://doi.org/10.3389/fped.2026.1912111
LA - en
ER -

CSL-JSON

{
"id": "10.3389/fped.2026.1912111",
"type": "article-journal",
"title": "Serial amplitude-integrated electroencephalography trajectories and brain-injury-associated functional immaturity in extremely preterm infants",
"container-title": "Frontiers in pediatrics",
"author": [
{
"family": "Zhang",
"given": "Yong"
},
{
"family": "Xu",
"given": "Dekai"
},
{
"family": "Luo",
"given": "Guoyong"
},
{
"family": "Xu",
"given": "Yongping"
},
{
"family": "Yang",
"given": "Yujing"
},
{
"family": "Ji",
"given": "Yong"
}
],
"container-title-short": "Front Pediatr",
"volume": "14",
"page": "1912111",
"DOI": "10.3389/fped.2026.1912111",
"PMID": "42718659",
"PMCID": "PMC13553765",
"ISSN": "2296-2360",
"publisher": "Frontiers Media SA",
"URL": "https://doi.org/10.3389/fped.2026.1912111",
"language": "en",
"issued": {
"date-parts": [
[
2026,
8,
26
]
]
}
}

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:10.1162/imag.a.1324
Spontaneous activation of cortical somatosensory networks depresses their excitability in preterm human neonates.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: EEG, other condition, 2 references
[2] doi:10.3390/brainsci16070678
CP5-Centered Parietal HD-tACS Is Associated with Improved Performance in a Smartphone-Based Shopping Task in Older Adults: A Behavioral and EEG Investigation.
Journal: Brain sciences
In common: EEG, 2 references
[3] doi:10.1016/j.dcn.2026.101775 [code]
Neonatal brain-age models in full- and preterm infants.
Journal: Developmental cognitive neuroscience
In common: developmental, other condition, 1 reference
[4] doi:10.64898/2026.04.01.26349523 [code]
Developmental brain age gap in prematurity and postnatally emerging delay in congenital heart disease
Journal: medRxiv (preprint)
In common: developmental, 1 reference
[5] doi:10.1162/imag.a.1293 [code]
PANDA: Patch-based unsupervised deep learning for brain anomaly detection via age prediction in fetal MRI.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: developmental, 1 reference
[6] doi:10.1111/desc.70187 [code]
Prenatal Volume in the Bilateral Superior Temporal Gyrus Associates With Children's Expressive Vocabulary at 24-36 Months.
Journal: Developmental science
In common: developmental, 1 reference
[7] doi:10.1038/s41597-026-07809-9 [code]
EEG-based brain-computer interface (BCI) dataset for directional word recognition.
Journal: Scientific data
In common: EEG, 1 reference
[8] doi:10.3389/fpsyg.2026.1768430
From neuromyths to neuroscience literacy: a cross-domain study of brain misconceptions among educators in Chilean special schools.
Journal: Frontiers in psychology
In common: developmental, 1 reference
[9] doi:10.1038/s41467-026-72545-y [code]
Personalized machine learning-guided radiation dose escalation in newly diagnosed glioblastoma: prospective pilot study.
Journal: Nature communications
In common: other condition, 1 reference
[10] doi:10.1016/j.isci.2026.116945
Neurodegeneration in the olfactory system in Niemann Pick type C1 disease.
Journal: iScience
In common: other condition, 1 reference

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.