OSCR

Contrast-free identification of glioma blood-brain barrier status via generative diffusion AI and non-contrast MRI.

Code ↔ Paper

6 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 6 matches
  1. [1] § Methods › Experimental setting ↔ main/train_CBSI_ide.py, lines 22–148 · score 0.81 · weight decay, identification model, Adam, cosine, setup, warm
  2. [2] § Methods › Image preprocessing and postprocessing ↔ preprocess/Preprocess_grayscale_norm.py, lines 8–40 · score 0.73 · binary brain mask, preprocessing grayscale, background, voxel, volumes, MRI
  3. [3] § Methods › Network architecture ↔ main/models/Networks_gen/Networks_simple_UNet_DDPM.py, lines 121–255 · score 0.53 · ReLU, upsampling, downsample, embeddings, activation, blocks
  4. [4] § Methods › Metrics ↔ main/models/Networks_gen/Validation_inference.py, lines 15–151 · score 0.52 · synthetic images, PSNR, SSIM, metrics, T1Gd, MAE
  5. [5] § Results › Performance on the identification of BBB status ↔ main/train_CBSI_ide.py, lines 22–148 · score 0.52 · precision recall, identification model, ROC, trained, CBSI
  6. [6] § Results › Performance on T1Gd synthesis ↔ main/Nii_utils.py, lines 151–168 · score 0.51 · absolute error, ground truth, MAE

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

Python · 231 lines · 11 KB · MIT · 2 matches

  1. import os
  2. import time
  3. import glob
  4. import math
  5. import argparse
  6. from tqdm import tqdm
  7. import torch.nn as nn
  8. import torch.optim as optim
  9. from datetime import datetime
  10. from torch.utils.data import DataLoader
  11. from torch.optim.lr_scheduler import MultiStepLR
  12. from torch.utils.tensorboard import SummaryWriter
  13. from sklearn.metrics import roc_auc_score, precision_recall_fscore_support
  15. from Nii_utils import *
  16. from dataset.Dataset_ide import Dataset_harmonize_2D_t1_t2f_t1c
  17. from models.Networks_ide.model import EfficientNet
  18. from models.Networks_ide.Validation_inference import Model_Validation, Model_Inference
  22. def main(opt):
  23. train_writer = SummaryWriter(join(opt.save_dir, 'log/train'), flush_secs=2)
  24. val_writer = SummaryWriter(join(opt.save_dir, 'log/val'), flush_secs=2)
  25. print(opt.save_dir)
  26. # -------------- Identification Model Setup ----------------
  27. net = EfficientNet.from_name(model_name=f'efficientnet-{opt.model_name[-2:]}', in_channels=opt.inchannel, num_classes=opt.classes, dropout_rate=opt.drop).to(opt.device)
  28. # Load pretrained weights for EfficientNet
  29. net_weights = net.state_dict()
  30. pre_weights = torch.load(f'./main/models/Networks_ide/pretrain/efficientnet-b0-355c32eb.pth') # Download from the official website
  31. pre_dict = {k: v for k, v in pre_weights.items() if net_weights[k].numel() == v.numel()}
  32. net.load_state_dict(pre_dict, strict=False)
  33. print(f'The model is 2D {opt.model_name}')
  35. # ----------------------loss & optimizer------------------------
  36. criterion = nn.BCEWithLogitsLoss().to(opt.device) # Sigmoid-BCELoss
  37. optimizer = optim.Adam(net.parameters(), lr=opt.lr_max, weight_decay=5e-4)
  39. # -------------- Learning Rate Scheduler ----------------
  40. if opt.warmup:
  41. # warm_up_with_cosine_lr
  42. int_decay = opt.lr_min / opt.lr_max
  43. zoom = 1 - int_decay
  45. warm_up_with_cosine_lr = lambda \
  46. epoch: int_decay + epoch / opt.warm_up_epochs * zoom if epoch <= opt.warm_up_epochs else int_decay + zoom * 0.5 * (
  47. math.cos((epoch - opt.warm_up_epochs) / (opt.max_epoch - opt.warm_up_epochs) * math.pi) + 1)
  48. scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=warm_up_with_cosine_lr)
  49. else:
  50. scheduler = MultiStepLR(optimizer, milestones=[int((1 / 3) * opt.max_epoch), int((2 / 3) * opt.max_epoch)], gamma=0.1, last_epoch=-1)
  52. # -------------- Dataset Setup ----------------
  53. ckpt_dir = join(opt.save_dir, 'train_model')
  54. os.makedirs(ckpt_dir, exist_ok=True)
  56. # ======== Load dataset and dataloader ========
  57. root_dir = './Glioma_DATA/Preprocessing_DATA/Train'
  58. val_dir = './Glioma_DATA/Preprocessing_DATA/Val'
  60. train_set = Dataset_harmonize_2D_t1_t2f_t1c(opt, root_dir=root_dir, dataset='Train')
  61. val_set = Dataset_harmonize_2D_t1_t2f_t1c(opt, root_dir=val_dir, dataset='Val')
  63. train_loader = DataLoader(train_set, batch_size=opt.bs, shuffle=True, num_workers=opt.num_threads, drop_last=False)
  64. val_loader = DataLoader(val_set, batch_size=opt.val_bs, shuffle=False, num_workers=opt.num_threads, drop_last=False)
  65. print(f'Size of train_dataset:{len(train_set)}.')
  66. # print(f'Size of val_dataset:{len(val_set)}.')
  67. print('Data prepared.')
  69. # -------------- Metrics Init ----------------
  70. threshold = 0.5
  71. best_AUC, best_epoch = 0, 0
  72. Save_Parameter(opt)
  73. # ==================== Training Loop ====================
  74. print('Start training.')
  75. for epoch in tqdm(range(opt.max_epoch)):
  76. train_start_time = time.time()
  77. net.train()
  78. for param_group in optimizer.param_groups:
  79. lr = param_group['lr']
  80. break
  81. train_acc_list = []
  82. train_loss_list = []
  83. y_true = torch.tensor([]).to(opt.device)
  84. y_pred = torch.tensor([]).to(opt.device)
  85. y_binary = torch.tensor([]).to(opt.device)
  86. # ------------------ Batch Training ------------------
  87. for i, DATA in enumerate(train_loader):
  88. image = DATA['image'].to(opt.device)
  89. label = DATA['label'].to(opt.device)
  90. net.zero_grad()
  91. y = net(image)
  92. loss = criterion(y, label)
  93. # Optional flood loss
  94. if opt.do_flood:
  95. flood_loss = (loss - opt.flood).abs() + opt.flood
  96. flood_loss.backward()
  97. else:
  98. loss.backward()
  99. optimizer.step()
  101. # Metrics logging
  102. y = torch.sigmoid(y)
  103. y_binary = torch.cat([y_binary, (y.detach() > threshold)])
  104. y_true = torch.cat([y_true, label.detach()])
  105. y_pred = torch.cat([y_pred, y.detach()])
  107. hit = ((y.detach() < threshold) ^ label.bool()).sum()
  108. train_acc_list.append(np.array(hit.cpu()))
  109. train_loss_list.append(np.array(loss.detach().cpu()))
  111. scheduler.step()
  112. # Calculate epoch-level metrics
  113. train_loss = np.array(train_loss_list).mean()
  114. train_acc = np.array(train_acc_list).sum() / len(train_set)
  115. train_auc = roc_auc_score(y_true.cpu(), y_pred.cpu())
  117. train_precision, train_recall, train_F1_score, _ = precision_recall_fscore_support(y_true.int().cpu(),
  118. y_binary.int().cpu(),
  119. average='binary')
  120. # TensorBoard Logging
  121. train_writer.add_scalar('ide_lr', lr, epoch)
  122. train_writer.add_scalar('ide_loss', train_loss, epoch)
  123. train_writer.add_scalar('ide_AUC', train_auc, epoch)
  124. train_writer.add_scalar('ide_ACC', train_acc, epoch)
  125. train_writer.add_scalar('ide_F1_score', train_F1_score, epoch)
  126. train_writer.add_scalar('ide_time', (time.time() - train_start_time) / 60, epoch)
  127. train_writer.close()
  129. # ------------------ Validation Phase ------------------
  130. AUC_val, _ = Model_Validation(opt, epoch, net, {'val':val_loader}, dataset='val', save_dir=opt.save_dir, writer={'val':val_writer}, train=True, criterion=criterion)
  132. if AUC_val > best_AUC:
  133. try:
  134. os.remove(join(opt.save_dir, 'train_model', f'best_AUC_epoch{best_epoch}.pth'))
  135. except:
  136. pass
  137. best_epoch = epoch
  138. if epoch > 10:
  139. best_AUC = AUC_val
  140. torch.save(net.state_dict(),join(opt.save_dir, 'train_model', f'best_AUC_epoch{best_epoch}.pth'))
  141. torch.save(net.state_dict(),join(opt.save_dir, 'train_model', f'latest_epoch{epoch}.pth'))
  142. try:
  143. os.remove(join(opt.save_dir, 'train_model', f'latest_epoch{epoch - 1}.pth'))
  144. except:
  145. pass
  146. torch.save(net.state_dict(),join(opt.save_dir, 'train_model', 'final' + '.pth'))
  148. return net
  151. def pred(opt, net=None):
  152. if not net:
  153. net = EfficientNet.from_name(model_name=f'efficientnet-{opt.model_name[-2:]}', in_channels=opt.inchannel, num_classes=opt.classes, dropout_rate=opt.drop).to(opt.device)
  154. net.load_state_dict(torch.load(sorted(glob.glob(join(opt.save_dir, 'train_model', 'best_AUC_epoch*.pth')))[-1]), strict=True)
  156. val_dir = './Glioma_DATA/Preprocessing_DATA/Val'
  157. test_dir = './Glioma_DATA/Preprocessing_DATA/Test'
  158. with torch.no_grad():
  159. Model_Inference(opt, val_dir, net, dataset='Val', label_known=True)
  160. Model_Inference(opt, test_dir, net, dataset='Test', label_known=False)
  163. if __name__ == '__main__':
  164. parser = argparse.ArgumentParser()
  165. # -------------------- Training settings
  166. parser.add_argument('--gpu', type=str, default='0', help='which gpu is used')
  167. parser.add_argument('--seed', type=int, default=42, help='random seed')
  168. parser.add_argument('--max_epoch', type=int, default=100, help='all_epochs')
  169. parser.add_argument('--lr_min', type=float, default=1e-5, help='initial learning rate')
  170. parser.add_argument('--lr_max', type=float, default=5e-4, help='max learning rate')
  171. parser.add_argument('--bs', type=int, default=2, help='training input batch size')
  172. parser.add_argument('--num_threads', type=int, default=1, help='# threads for loading dataset')
  174. # -------------------- Inference settings
  175. parser.add_argument('--val_bs', type=int, default=4, help='Val/Test batch size')
  176. parser.add_argument('--ref_timestep', type=int, default=100, help='<=1000, 1000 is time-consuming but of higher quality')
  177. parser.add_argument('--save_dir', type=str, default='', help="./main/trained_models/CBSI_ide/{bs*_ImageSize*_epoch*_seed*_time}/") # Path for saving model parameters
  179. # -------------------- Data settings
  180. parser.add_argument('--data_dim', type=str, default='2D')
  181. parser.add_argument('--ImageSize', type=int, default=424, help='Spatial dimension cropped to 424 * 424')
  182. parser.add_argument("--gen_save_dir", type=str, default='', help="./main/trained_models/CBSI_gen/{pred_*_...class_seg_time}/")
  183. parser.add_argument('--MR_max', type=int, default=255, help='max value of preprocessed MR image')
  184. parser.add_argument('--MR_min', type=int, default=0, help='min value of preprocessed MR image')
  185. parser.add_argument('--preloading', type=bool, default=True, help='preloading the image')
  187. # -------------------- Model settings
  188. parser.add_argument('--model_name', type=str, default='EfficientNet_b0')
  189. parser.add_argument('--inchannel', type=int, default=3, help='input channel (T1, T2-FLAIR, and synthetic T1Gd)')
  190. parser.add_argument('--classes', type=int, default=1, help='')
  191. parser.add_argument('--drop', type=float, default=0.2, help='dropout rate 0~1 ')
  193. # -------------------- Loss function
  194. parser.add_argument('--do_flood', type=bool, default=True, help='do flood loss')
  195. parser.add_argument('--flood', type=float, default=0.1, help='flood loss threshold')
  196. parser.add_argument('--warmup', action='store_false')
  197. parser.add_argument('--warm_up_epochs', type=int, default=5, help='warm_up_epochs')
  199. # -------------------- Quick test settings
  200. parser.add_argument('--quick_test', action='store_true')
  201. parser.add_argument('--inference_only', action='store_true')
  202. opt = parser.parse_args()
  203. # torch.cuda.is_available = lambda: False
  204. os.environ["CUDA_VISIBLE_DEVICES"] = opt.gpu
  205. setup_seed(opt.seed)
  206. opt.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
  207. print("CBSI_ide Start")
  208. if opt.quick_test:
  209. opt.max_epoch = 10
  210. opt.ref_timestep = 10
  212. # -------------- Experiment naming & directory setup --------------
  213. if not opt.save_dir or not opt.inference_only:
  214. current_time = datetime.now().strftime('%b%d_%H-%M-%S')
  215. opt.save_dir = './main/trained_models/CBSI_ide/bs{}_ImageSize{}_epoch{}_seed{}_{}'.format(opt.bs, opt.ImageSize, opt.max_epoch, opt.seed, current_time)
  216. os.makedirs(opt.save_dir, exist_ok=True)
  217. os.makedirs(opt.save_dir, exist_ok=True)
  219. if not opt.inference_only:
  220. net = main(opt)
  221. pred(opt, net)
  222. else:
  223. pred(opt)
  225. if not opt.inference_only:
  226. print("CBSI_ide Training Done")
  227. else:
  228. print("CBSI_ide Inference Done")
  229. print("-------------------------------------------")
  230. print(f"Attention !! Results can be viewed here :\n {opt.save_dir}")
  231. print("-------------------------------------------")

train_CBSI_ide.py at commit be829a1, under MIT · at the source

Overview

Authors: Kaiyi Zheng1,2,3, Yiwen Zhang1,2,3, Hai Shu4, Ruolin Xiao1,2,3, Xinming Li5, Jianhua Ma6, Qianjin Feng1,2,3, Yuankui Wu7, Wei Yang1,2,3, Liming Zhong1,2,3
ORCID iDs: Hai Shu, Qianjin Feng, Wei Yang, Liming Zhong
  1. School of Biomedical Engineering, Southern Medical University, Guangzhou, China
  2. Guangdong Provincial Key Laboratory of Medical Image Processing, Guangzhou, China
  3. Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Guangzhou, China
  4. Department of Biostatistics, School of Global Public Health, New York University, New York, USA
  5. Department of Radiology, Zhujiang Hospital, Southern Medical University, Guangzhou, China
  6. School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, China
  7. Department of Medical Imaging Center, Nanfang Hospital, Southern Medical University, Guangzhou, China
Institutions: Key Laboratory of Guangdong Province (China); Southern Medical University (China); New York University (United States); Zhujiang Hospital (China); Xi'an Jiaotong University (China); Nanfang Hospital (China)
Journal: Nature communications, volume 17, issue 1, article 2162
Dates: received 26 October 2024; accepted 28 January 2026; published online 3 March 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s41467-026-69578-8 · PMID 41776178 · PMCID PMC12960725 · OpenAlex W7133330152
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: structural MRI / diffusion (modality), human (organism), other condition (population)
Methods: Connectivity, Statistics, Machine learning
Keywords: Biomedical engineering, Magnetic resonance imaging, Computational science
MeSH: Blood-Brain Barrier*, Brain Neoplasms*, Glioma*, Magnetic Resonance Imaging*, Artificial Intelligence, Contrast Media, Gadolinium, Generative Artificial Intelligence, Humans (* major topic)
Topic: Glioma Diagnosis and Treatment (Genetics, Medicine), according to OpenAlex
Funding: National Natural Science Foundation of China (National Science Foundation of China) (62101239)
Citations: cited by 1 paper (Europe PMC); 53 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 6 matches between paragraphs and lines of code.

SMU-MedicalVision/CBSI-master

License: MIT
State: the link answers, verified on 30 September 2026
Evidence: files inventoried
Commit: be829a11c5a320e43c9211b2aa31b878bcd7c6c3, 23 July 2025
Languages: Python (20)
Size: 62 files, 20 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, license file, environment (requirements.txt)
Not found: CITATION.cff, tests, continuous integration, documentation
Tools: PyTorch (13 files), NumPy (7 files), pandas (3 files), scikit-learn (3 files), NiBabel (1 file), OpenCV (1 file), scikit-image (1 file), SimpleITK (1 file)
Availability: 1 check, the latest on 30 September 2026: the link answers
  • 30 September 2026: the link answers
22 files
At the source: github.com/SMU-MedicalVision/CBSI-master

Code availability statement

The paper has a code 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/s41467-026-69578-8.

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;
  • 20 scripts, each with its path and the digest of its content;
  • 6 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

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

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/s41467-026-69578-8.

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, issue, pages, dates, 10 authors, 3 keywords, 9 MeSH terms, 1 funder, 33 references.

Cite

This paper

Zheng, K., Zhang, Y., Shu, H., Xiao, R., Li, X., Ma, J., Feng, Q., Wu, Y., Yang, W., & Zhong, L. (2026). Contrast-free identification of glioma blood-brain barrier status via generative diffusion AI and non-contrast MRI. Nature communications, 17(1), 2162. https://doi.org/10.1038/s41467-026-69578-8

BibTeX

@article{zheng2026contrast,
author = {Zheng, Kaiyi and Zhang, Yiwen and Shu, Hai and Xiao, Ruolin and Li, Xinming and Ma, Jianhua and Feng, Qianjin and Wu, Yuankui and Yang, Wei and Zhong, Liming},
title = {{Contrast-free identification of glioma blood-brain barrier status via generative diffusion AI and non-contrast MRI}},
journal = {Nature communications},
year = {2026},
month = mar,
volume = {17},
number = {1},
pages = {2162},
publisher = {Nature Publishing Group},
issn = {2041-1723},
doi = {10.1038/s41467-026-69578-8},
url = {https://doi.org/10.1038/s41467-026-69578-8},
pmid = {41776178},
pmcid = {PMC12960725}
}

RIS

TY - JOUR
AU - Zheng, Kaiyi
AU - Zhang, Yiwen
AU - Shu, Hai
AU - Xiao, Ruolin
AU - Li, Xinming
AU - Ma, Jianhua
AU - Feng, Qianjin
AU - Wu, Yuankui
AU - Yang, Wei
AU - Zhong, Liming
TI - Contrast-free identification of glioma blood-brain barrier status via generative diffusion AI and non-contrast MRI
T2 - Nature communications
J2 - Nat Commun
PY - 2026
DA - 2026/03/03
VL - 17
IS - 1
SP - 2162
SN - 2041-1723
PB - Nature Publishing Group
DO - 10.1038/s41467-026-69578-8
UR - https://doi.org/10.1038/s41467-026-69578-8
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41467-026-69578-8",
"type": "article-journal",
"title": "Contrast-free identification of glioma blood-brain barrier status via generative diffusion AI and non-contrast MRI",
"container-title": "Nature communications",
"author": [
{
"family": "Zheng",
"given": "Kaiyi"
},
{
"family": "Zhang",
"given": "Yiwen"
},
{
"family": "Shu",
"given": "Hai"
},
{
"family": "Xiao",
"given": "Ruolin"
},
{
"family": "Li",
"given": "Xinming"
},
{
"family": "Ma",
"given": "Jianhua"
},
{
"family": "Feng",
"given": "Qianjin"
},
{
"family": "Wu",
"given": "Yuankui"
},
{
"family": "Yang",
"given": "Wei"
},
{
"family": "Zhong",
"given": "Liming"
}
],
"container-title-short": "Nat Commun",
"volume": "17",
"issue": "1",
"page": "2162",
"DOI": "10.1038/s41467-026-69578-8",
"PMID": "41776178",
"PMCID": "PMC12960725",
"ISSN": "2041-1723",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41467-026-69578-8",
"language": "en",
"issued": {
"date-parts": [
[
2026,
3,
3
]
]
}
}

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:10.21037/qims-2026-0792 [code]
An nnU-Net-based framework with adaptive feature representation for 3D brain tumor segmentation.
Journal: Quantitative imaging in medicine and surgery
In common: SimpleITK, OpenCV, scikit-image, 5 other tools, structural MRI / diffusion, other condition, 1 reference
[2] doi:10.3389/fnins.2026.1870124 [code]
An end-to-end pipeline for automated fetal brain segmentation and biometry from 3D SSFP MRI.
Journal: Frontiers in neuroscience
In common: SimpleITK, OpenCV, scikit-image, 5 other tools, structural MRI / diffusion, 1 reference
[3] 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: SimpleITK, OpenCV, scikit-image, 5 other tools, structural MRI / diffusion, other condition
[4] doi:10.3389/fmed.2026.1875760 [code]
Adaptive multi-stage domain unlearning for white-matter lesion segmentation.
Journal: Frontiers in medicine
In common: SimpleITK, scikit-image, NiBabel, 4 other tools, structural MRI / diffusion, 2 references
[5] doi:10.3389/frai.2026.1771088 [code]
Few-shot deployment of pretrained MRI transformers in brain imaging tasks.
Journal: Frontiers in artificial intelligence
In common: SimpleITK, OpenCV, scikit-image, 5 other tools, structural MRI / diffusion
[6] doi:10.1371/journal.pone.0344600 [code]
Robust disease prognosis via diagnostic knowledge preservation: A sequential learning approach.
Journal: PloS one
In common: OpenCV, scikit-image, NiBabel, 4 other tools, other condition, 1 reference
[7] doi:10.1002/hipo.70124 [code]
Association Between Anterior Hippocampal Gyrification and Episodic Memory Performance in Neurotypical Young Adults.
Journal: Hippocampus
In common: SimpleITK, scikit-image, NiBabel, 4 other tools, structural MRI / diffusion, 1 reference
[8] doi:10.1186/s12880-026-02335-x [code]
Automatic lateral ventricle and choroid plexus segmentation method in infant brain MR images.
Journal: BMC medical imaging
In common: SimpleITK, scikit-image, NiBabel, 4 other tools, structural MRI / diffusion, 1 reference
[9] doi:10.1038/s41598-026-55397-w [code]
Fast surface reconstruction of human brain MRI: benchmarking deep-learning based morphometry tools.
Journal: Scientific reports
In common: SimpleITK, scikit-image, NiBabel, 4 other tools, structural MRI / diffusion, 1 reference
[10] doi:10.1016/j.adro.2026.102092 [code]
Effect of Anatomic Contextual Information on the Performance of a Convolutional Neural Network Tasked With Brain Metastasis Detection.
Journal: Advances in radiation oncology
In common: SimpleITK, scikit-image, NiBabel, 4 other tools, 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.