OSCR

Hybrid Vi+ECNN framework for advanced ADHD diagnostic accuracy in medical imaging.

Code ↔ Paper

10 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 10 matches · 1 of them tie a paragraph to a whole file, not to given lines: a weak match, whose lines are not tinted
  1. [1] § Proposed model › Hybrid ViT+ECNN model ↔ Combined.ipynb, lines 119–157 · score 0.71 · linearly projected, classification layer, CLS, Patch16, pretrained, token
  2. [2] § Experimental results › Hyperparameter settings ↔ Comments response .ipynb, lines 227–299 · score 0.65 · AdamW, cross validation, weight decay, stratified, memory, metrics
  3. [3] § Experimental results › Hyperparameter settings ↔ Comments__response_ v2.ipynb, lines 227–299 · score 0.65 · AdamW, cross validation, weight decay, stratified, memory, metrics
  4. [4] § Proposed model › Data preprocessing ↔ Combined.ipynb, lines 24–30 · score 0.62 · Quantile Histogram Equalization, median filtering, QHED
  5. [5] § Proposed model › Data preprocessing ↔ ECNN.ipynb, lines 23–29 · score 0.62 · Quantile Histogram Equalization, median filtering, QHED
  6. [6] § Proposed model › Hybrid ViT+ECNN model ↔ Combined.ipynb, lines 119–157 · score 0.61 · fully connected layer, ReLU, concatenation, module, ViT, ECNN
  7. [7] § Proposed model › Data augmentation ↔ ViT.ipynb, the whole file · a weak match · score 0.56 · cosine annealing learning, scheduler, cross, loss, splits, validation
  8. [8] § Proposed model › Data augmentation ↔ Combined.ipynb, lines 185–188 · score 0.52 · cosine annealing learning, scheduler, cross, loss, Model
  9. [9] § Experimental results › Structural ablation study for module necessity validation ↔ Comments__response_ v2.ipynb, lines 663–782 · score 0.51 · structural ablation, necessity, score, recall, configuration, precision
  10. [10] § Experimental results › Structural ablation study for module necessity validation ↔ Comments__response_ v2.ipynb, lines 663–782 · score 0.51 · structural ablation, QHED preprocessing, necessity, fusion, configuration, modules

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

Jupyter notebook · 260 lines · 9.6 KB · no license · 4 matches

  1. # %%
  2. import os
  3. import numpy as np
  4. import torch
  5. import torch.nn as nn
  6. from torch.optim import AdamW
  7. from torch.optim.lr_scheduler import CosineAnnealingLR
  8. from torchvision import datasets, transforms
  9. from torch.utils.data import DataLoader
  10. from transformers import ViTForImageClassification
  11. from PIL import Image, ImageEnhance, ImageFilter
  13. # --- Preprocessing Functions ---
  14. def apply_pst(image, alpha=0.1, beta=0.1):
  15. """Apply Phase Stretch Transform (PST) to the image."""
  16. image = np.array(image, dtype=np.float32) / 255.0
  17. fft_image = np.fft.fft2(image)
  18. magnitude = np.abs(fft_image)
  19. phase = np.angle(fft_image)
  20. phase_stretched = phase + alpha * np.tanh(beta * phase)
  21. pst_image = np.abs(np.fft.ifft2(magnitude * np.exp(1j * phase_stretched)))
  22. return pst_image
  24. def apply_qhed(image):
  25. """Apply Quantile Histogram Equalization and Denoising (QHED) to the image."""
  26. image = image.convert("RGB")
  27. enhancer = ImageEnhance.Contrast(image)
  28. enhanced_image = enhancer.enhance(2.0)
  29. denoised_image = enhanced_image.filter(ImageFilter.MedianFilter(size=3))
  30. return denoised_image
  32. # --- Dataset Class ---
  33. class TripleInputDataset(datasets.ImageFolder):
  34. """Dataset that returns raw, PST, and QHED images."""
  35. def __getitem__(self, index):
  36. path, label = self.samples[index]
  37. image = Image.open(path).convert("RGB")
  39. # Raw Image
  40. raw_image = self.transform(image)
  42. # PST Preprocessed Image
  43. pst_image = apply_pst(image)
  44. pst_image = Image.fromarray((pst_image * 255).astype(np.uint8)).convert("RGB")
  45. pst_image = self.transform(pst_image)
  47. # QHED Preprocessed Image
  48. qhed_image = apply_qhed(image)
  49. qhed_image = self.transform(qhed_image)
  51. return raw_image, pst_image, qhed_image, label
  53. # --- Residual Block ---
  54. class ResidualBlock(nn.Module):
  55. """A residual block with convolutional layers and skip connections."""
  56. def __init__(self, in_channels, out_channels, stride=1):
  57. super(ResidualBlock, self).__init__()
  58. self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=stride, padding=1)
  59. self.bn1 = nn.BatchNorm2d(out_channels)
  60. self.relu = nn.ReLU(inplace=True)
  61. self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1)
  62. self.bn2 = nn.BatchNorm2d(out_channels)
  64. self.skip = nn.Sequential()
  65. if stride != 1 or in_channels != out_channels:
  66. self.skip = nn.Sequential(
  67. nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=stride, bias=False),
  68. nn.BatchNorm2d(out_channels)
  69. )
  71. def forward(self, x):
  72. identity = self.skip(x)
  73. out = self.conv1(x)
  74. out = self.bn1(out)
  75. out = self.relu(out)
  76. out = self.conv2(out)
  77. out = self.bn2(out)
  78. out += identity
  79. return self.relu(out)
  81. # --- ECNN High-Capacity Model ---
  82. class ECNNHighCapacity(nn.Module):
  83. """High-Capacity Ensemble CNN with Residual Blocks."""
  84. def __init__(self, num_classes):
  85. super(ECNNHighCapacity, self).__init__()
  86. self.raw_cnn = self._create_cnn_branch()
  87. self.pst_cnn = self._create_cnn_branch()
  88. self.qhed_cnn = self._create_cnn_branch()
  90. # Combine features from all three branches and reduce to 512
  91. self.fc = nn.Sequential(
  92. nn.Linear(256 * 3, 512), # Combine and reduce features to 512
  93. nn.ReLU(),
  94. nn.Dropout(0.5),
  95. )
  97. def _create_cnn_branch(self):
  98. return nn.Sequential(
  99. ResidualBlock(3, 64),
  100. ResidualBlock(64, 128, stride=2),
  101. ResidualBlock(128, 128),
  102. ResidualBlock(128, 256, stride=2),
  103. nn.AdaptiveAvgPool2d((1, 1))
  104. )
  106. def forward(self, raw_x, pst_x, qhed_x):
  107. # Process each input branch
  108. raw_features = self.raw_cnn(raw_x).view(raw_x.size(0), -1) # Shape: (batch_size, 256)
  109. pst_features = self.pst_cnn(pst_x).view(pst_x.size(0), -1) # Shape: (batch_size, 256)
  110. qhed_features = self.qhed_cnn(qhed_x).view(qhed_x.size(0), -1) # Shape: (batch_size, 256)
  112. # Concatenate features
  113. combined_features = torch.cat((raw_features, pst_features, qhed_features), dim=1) # Shape: (batch_size, 768)
  115. # Reduce to 512
  116. return self.fc(combined_features) # Shape: (batch_size, 512)
  119. # --- Combined Fine-Tuned Model ---
  120. class CombinedFineTunedModel(nn.Module):
  121. """Fine-tuned model combining ViT and ECNN outputs."""
  122. def __init__(self, num_classes):
  123. super(CombinedFineTunedModel, self).__init__()
  124. # Pretrained ViT with hidden states enabled
  125. self.vit = ViTForImageClassification.from_pretrained(
  126. "google/vit-base-patch16-224-in21k",
  127. output_hidden_states=True # Enable hidden states for feature extraction
  128. )
  129. self.vit_hidden_size = 768 # Default hidden size of ViT
  130. self.ecnn = ECNNHighCapacity(num_classes) # ECNN with raw, PST, and QHED inputs
  132. # Linear projection to align ViT features with ECNN features
  133. self.vit_projection = nn.Linear(self.vit_hidden_size, 512)
  135. # Fully connected layer to combine ViT and ECNN features
  136. self.fc = nn.Sequential(
  137. nn.Linear(512 + 512, 256), # Combine ViT (512) + ECNN (512)
  138. nn.ReLU(),
  139. nn.Dropout(0.5),
  140. nn.Linear(256, num_classes) # Final classification layer
  141. )
  143. def forward(self, raw_x, pst_x, qhed_x):
  144. # Extract ViT features from the CLS token
  145. vit_outputs = self.vit(raw_x, output_hidden_states=True)
  146. vit_hidden_states = vit_outputs.hidden_states[-1] # Last hidden state
  147. vit_cls_token = vit_hidden_states[:, 0, :] # CLS token (batch_size, 768)
  148. vit_features = self.vit_projection(vit_cls_token) # Project to (batch_size, 512)
  150. # Get ECNN features
  151. ecnn_features = self.ecnn(raw_x, pst_x, qhed_x) # Output: (batch_size, 512)
  153. # Concatenate features from ViT and ECNN
  154. combined_features = torch.cat((vit_features, ecnn_features), dim=1) # Shape: (batch_size, 1024)
  156. # Pass through the final fully connected layer
  157. return self.fc(combined_features)
  160. # --- Training Setup ---
  161. train_dir = "/kaggle/input/binary-augmented-split-adhd/ADHD - Augmented Split Binary/train"
  162. val_dir = "/kaggle/input/binary-augmented-split-adhd/ADHD - Augmented Split Binary/validation"
  164. # Dataset Transform
  165. transform = transforms.Compose([
  166. transforms.Resize((224, 224)),
  167. transforms.ToTensor(),
  168. transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
  169. ])
  171. # Datasets and Loaders
  172. batch_size = 8
  173. train_dataset = TripleInputDataset(root=train_dir, transform=transform)
  174. val_dataset = TripleInputDataset(root=val_dir, transform=transform)
  176. train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
  177. val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
  179. # Instantiate Model
  180. num_classes = 2
  181. model = CombinedFineTunedModel(num_classes)
  182. device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
  183. model.to(device)
  185. # Optimizer, Scheduler, and Loss
  186. optimizer = AdamW(model.parameters(), lr=1e-4, weight_decay=1e-4)
  187. scheduler = CosineAnnealingLR(optimizer, T_max=10)
  188. criterion = nn.CrossEntropyLoss()
  190. # --- Save Checkpoint ---
  191. def save_checkpoint(epoch, model, optimizer, scheduler, train_loss, val_loss, train_acc, val_acc, path):
  192. checkpoint = {
  193. "epoch": epoch,
  194. "model_state_dict": model.state_dict(),
  195. "optimizer_state_dict": optimizer.state_dict(),
  196. "scheduler_state_dict": scheduler.state_dict(),
  197. "train_loss": train_loss,
  198. "val_loss": val_loss,
  199. "train_acc": train_acc,
  200. "val_acc": val_acc,
  201. }
  202. torch.save(checkpoint, path)
  203. print(f"Checkpoint saved: {path}")
  205. # --- Training Loop ---
  206. epochs = 10
  207. for epoch in range(epochs):
  208. model.train()
  209. running_loss, correct, total = 0.0, 0, 0
  211. for raw_x, pst_x, qhed_x, labels in train_loader:
  212. raw_x, pst_x, qhed_x, labels = raw_x.to(device), pst_x.to(device), qhed_x.to(device), labels.to(device)
  214. optimizer.zero_grad()
  215. outputs = model(raw_x, pst_x, qhed_x)
  216. loss = criterion(outputs, labels)
  217. loss.backward()
  218. optimizer.step()
  220. running_loss += loss.item()
  221. _, preds = torch.max(outputs, 1)
  222. correct += (preds == labels).sum().item()
  223. total += labels.size(0)
  225. scheduler.step()
  226. train_loss = running_loss / len(train_loader)
  227. train_acc = correct / total
  228. print(f"Epoch {epoch+1}, Train Loss: {train_loss:.4f}, Train Acc: {train_acc:.4f}")
  230. # Validation Loop
  231. model.eval()
  232. val_loss, val_correct, val_total = 0.0, 0, 0
  233. with torch.no_grad():
  234. for raw_x, pst_x, qhed_x, labels in val_loader:
  235. raw_x, pst_x, qhed_x, labels = raw_x.to(device), pst_x.to(device), qhed_x.to(device), labels.to(device)
  236. outputs = model(raw_x, pst_x, qhed_x)
  237. loss = criterion(outputs, labels)
  238. val_loss += loss.item()
  239. _, preds = torch.max(outputs, 1)
  240. val_correct += (preds == labels).sum().item()
  241. val_total += labels.size(0)
  243. val_loss /= len(val_loader)
  244. val_acc = val_correct / val_total
  245. print(f"Epoch {epoch+1}, Val Loss: {val_loss:.4f}, Val Acc: {val_acc:.4f}")
  247. # Save Model
  248. checkpoint_path = f"fine_tuned_combined_model_epoch_{epoch+1}.pth"
  249. save_checkpoint(
  250. epoch=epoch + 1,
  251. model=model,
  252. optimizer=optimizer,
  253. scheduler=scheduler,
  254. train_loss=train_loss,
  255. val_loss=val_loss,
  256. train_acc=train_acc,
  257. val_acc=val_acc,
  258. path=checkpoint_path,
  259. )

Combined.ipynb at commit 3a399cf, no license · at the source

Overview

Authors: Shaymaa E Sorour1, Lamia Hassan1, Osman Elwasila1, Tsunenori Mine2, Mohamed Ali Nagy Elmaadaway3
  1. Department of Management Information Systems, School of Business, King Faisal University, 31982 Al-Ahsa, Saudi Arabia
  2. Department of Advanced Information Technology, Faculty of Information Science and Electrical Engineering, Kyushu University, Fukuoka, 819-0395 Japan
  3. Research Center of Excellence in Science and Mathematics Education Development, DSR, King Saud University, 2458, 11451 Riyadh, Saudi Arabia
Institutions: King Faisal University (Saudi Arabia); Kyushu University (Japan); King Saud University (Saudi Arabia)
Journal: Scientific reports, volume 16, issue 1, article 19017
Dates: received 12 July 2025; accepted 23 April 2026; published online 2 May 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s41598-026-50791-w · PMID 42069836 · PMCID PMC13280160 · OpenAlex W7159982485
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: structural MRI / diffusion (modality), human (organism), ADHD (population), clinical / translational (subfield)
Methods: Machine learning, Statistics, Connectivity, Preprocessing
Keywords: Attention-deficit/hyperactivity disorder (ADHD), Deep learning, Vision transformer (ViT), Enhanced convolutional neural network (ECNN), Medical image classification, Pediatric MRI, Hybrid architectures, Computational biology and bioinformatics, Diseases, Engineering, Health care, Mathematics and computing, Medical research, Neuroscience
MeSH: Attention Deficit Disorder with Hyperactivity*, Brain*, Deep Learning*, Image Processing, Computer-Assisted*, Magnetic Resonance Imaging*, Child, Convolutional Neural Networks, Humans (* major topic)
Topic: Attention Deficit Hyperactivity Disorder (Psychiatry and Mental health, Medicine), according to OpenAlex
Funding: Deanship of Scientific Research, King Faisal University (KFU254121)
Citations: not cited yet (Europe PMC); 66 references in the paper

Abstract

The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.

Repository

Its files are read in the Code ↔ Paper reader above, with 10 matches between paragraphs and lines of code.

shaymaasorour/Hybrid-ViT-ECNN-Framework-for-Advanced-ADHD-Diagnostic-Accuracy-in-Medical-Imaging

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 3a399cf849f18af7132ca1882adc4933baa6b856, 22 April 2026
Languages: Jupyter (5)
Size: 6 files, 5 scripts
Software Heritage: not archived
Found in: “Data availability”
Holds: README, 5 notebooks
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: PyTorch (5 files), NumPy (4 files), Pillow (4 files), Matplotlib (2 files), OpenCV (2 files), pandas (2 files), scikit-learn (2 files), seaborn (2 files), Hugging Face Transformers (2 files)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
6 files
At the source: github.com/shaymaasorour/Hybrid-ViT-ECNN-Framework-for-Advanced-ADHD-Diagnostic-Accuracy-in-Medical-Imaging

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;
  • 5 scripts, each with its path and the digest of its content;
  • 10 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

Code and data availability statement

The paper has a code and data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

Read it in the paper: doi.org/10.1038/s41598-026-50791-w.

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, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 5 authors, 14 keywords, 8 MeSH terms, 1 funder, 61 references.

Cite

This paper

Sorour, S. E., Hassan, L., Elwasila, O., Mine, T., & Elmaadaway, M. A. N. (2026). Hybrid Vi+ECNN framework for advanced ADHD diagnostic accuracy in medical imaging. Scientific reports, 16(1), 19017. https://doi.org/10.1038/s41598-026-50791-w

BibTeX

@article{sorour2026hybrid,
author = {Sorour, Shaymaa E and Hassan, Lamia and Elwasila, Osman and Mine, Tsunenori and Elmaadaway, Mohamed Ali Nagy},
title = {{Hybrid Vi+ECNN framework for advanced ADHD diagnostic accuracy in medical imaging}},
journal = {Scientific reports},
year = {2026},
month = may,
volume = {16},
number = {1},
pages = {19017},
publisher = {Nature Publishing Group},
issn = {2045-2322},
doi = {10.1038/s41598-026-50791-w},
url = {https://doi.org/10.1038/s41598-026-50791-w},
pmid = {42069836},
pmcid = {PMC13280160}
}

RIS

TY - JOUR
AU - Sorour, Shaymaa E
AU - Hassan, Lamia
AU - Elwasila, Osman
AU - Mine, Tsunenori
AU - Elmaadaway, Mohamed Ali Nagy
TI - Hybrid Vi+ECNN framework for advanced ADHD diagnostic accuracy in medical imaging
T2 - Scientific reports
J2 - Sci Rep
PY - 2026
DA - 2026/05/02
VL - 16
IS - 1
SP - 19017
SN - 2045-2322
PB - Nature Publishing Group
DO - 10.1038/s41598-026-50791-w
UR - https://doi.org/10.1038/s41598-026-50791-w
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41598-026-50791-w",
"type": "article-journal",
"title": "Hybrid Vi+ECNN framework for advanced ADHD diagnostic accuracy in medical imaging",
"container-title": "Scientific reports",
"author": [
{
"family": "Sorour",
"given": "Shaymaa E"
},
{
"family": "Hassan",
"given": "Lamia"
},
{
"family": "Elwasila",
"given": "Osman"
},
{
"family": "Mine",
"given": "Tsunenori"
},
{
"family": "Elmaadaway",
"given": "Mohamed Ali Nagy"
}
],
"container-title-short": "Sci Rep",
"volume": "16",
"issue": "1",
"page": "19017",
"DOI": "10.1038/s41598-026-50791-w",
"PMID": "42069836",
"PMCID": "PMC13280160",
"ISSN": "2045-2322",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41598-026-50791-w",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
2
]
]
}
}

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:10.1093/braincomms/fcag253 [code]
Disease detection and classification in temporal lobe epilepsy: step-wise versus simultaneous AI decision models in a multisite neuroimaging study.
Journal: Brain communications
In common: Hugging Face Transformers, OpenCV, Pillow, 6 other tools, structural MRI / diffusion, clinical / translational
[2] doi:10.1038/s43856-026-01817-x [code]
Visual prompt engineering for multimodal and irregularly sampled medical data.
Journal: Communications medicine
In common: Hugging Face Transformers, OpenCV, Pillow, 6 other tools, clinical / translational
[3] doi:10.1186/s12880-026-02481-2 [code]
Deep learning-based neuroanatomical profiling reveals population-specific brain changes in multiple sclerosis: a large-scale Middle Eastern study.
Journal: BMC medical imaging
In common: OpenCV, Pillow, seaborn, 4 other tools, structural MRI / diffusion, clinical / translational, 2 references
[4] doi:10.1017/s003329172610405x [code]
Cortical morphometric inverse divergence in attention-deficit/hyperactivity disorder correlates with cell-type-specific, laminar-specific and developmental transcriptomic signatures.
Journal: Psychological medicine
In common: pandas, Matplotlib, NumPy, ADHD, structural MRI / diffusion, 4 references
[5] doi:10.1038/s42003-026-10957-8 [code]
Brain defence by the extracellular matrix protein Cochlin.
Journal: Communications biology
In common: Hugging Face Transformers, OpenCV, Pillow, 6 other tools
[6] doi:10.1038/s41467-026-76837-1 [code]
Drug screen and machine learning predict neuroprotective agents in a preclinical human model of childhood dementia.
Journal: Nature communications
In common: Hugging Face Transformers, OpenCV, Pillow, 6 other tools
[7] doi:10.1038/s41467-026-74358-5 [code]
Brain-inspired spatial intelligence for embodied agents.
Journal: Nature communications
In common: Hugging Face Transformers, OpenCV, Pillow, 6 other tools
[8] doi:10.7554/elife.107933 [code]
Modality-agnostic decoding of vision and language from fMRI.
Journal: eLife
In common: Hugging Face Transformers, OpenCV, Pillow, 6 other tools
[9] doi:10.1002/hbm.70469 [code]
VarCoNet: A Variability-Aware Self-Supervised Framework for Functional Connectome Extraction From Resting-State fMRI.
Journal: Human brain mapping
In common: Hugging Face Transformers, OpenCV, Pillow, 6 other tools
[10] doi:10.1371/journal.pone.0354511 [code]
TokenUNet: A new case for transformers integration in efficient and interpretable 3D UNets for brain imaging segmentation.
Journal: PloS one
In common: Hugging Face Transformers, PyTorch, seaborn, 4 other tools, 2 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.

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.