Radiomics model integrating MRI and ECV enhances prediction accuracy for progression in high-grade glioma.
One block of code here has a permanent identifier (oscr:b:…), listed with the matches below: citable for good, with a page of its own. Select a match and press Cite to copy it. What a block identifier is.
The 2 matches
- [1] § Methods › Model building ↔ Model.py, lines 173 to 313 · score 0.88 · cls token, LayerNorm, Transformer encoders, dropout, positional, blocks · block
oscr:b:7A5QTDP3M5AY5-L· Report a mismatch - [2] § Methods › Model building ↔ Model.py, lines 173 to 313 · score 0.81 · MobileNetV2, ReLU, blocks, encoders, head, ECV images · block
oscr:b:7A5QTDP3M5AY5-L· Report a mismatch
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
Python · 313 lines · 10 KB · no license · 2 matches
- import nibabel as nib
- import numpy as np
- from skimage.transform import resize
- def resample_to_spacing(img_nii, target_spacing=(3.0, 3.0, 3.0)):
- """
- 将 NIfTI 图像重采样到指定体素间距 (3mm, 3mm, 3mm)。
- img_nii: nibabel Nifti1Image
- 返回: resampled_volume (numpy array, float32)
- """
- volume = img_nii.get_fdata().astype(np.float32)
- original_spacing = img_nii.header.get_zooms()[:3] # (sx, sy, sz)
- zoom = np.array(original_spacing) / np.array(target_spacing)
- # 目标体素数
- new_shape = np.round(np.array(volume.shape) * zoom).astype(int)
- # 使用 scikit-image resize 实现(三线性插值)
- resampled = resize(
- volume,
- new_shape,
- order=1, # 线性插值
- mode='edge',
- anti_aliasing=False
- ).astype(np.float32)
- return resampled
- def zscore_normalize(volume, eps=1e-8):
- """
- 对单模态 3D 体数据做 Z-Score 标准化: (x - mean) / std
- """
- mean = volume.mean()
- std = volume.std()
- volume_norm = (volume - mean) / (std + eps)
- return volume_norm
- import random
- from scipy.ndimage import rotate
- def random_flip_3d(volume):
- # 概率 0.5 沿三个维度随机翻转
- if random.random() < 0.5:
- axis = random.choice([0, 1, 2])
- volume = np.flip(volume, axis=axis).copy()
- return volume
- def random_rotate_3d(volume, max_angle=10):
- # 在 [-max_angle, max_angle] 范围随机旋转某一个轴平面
- angle = random.uniform(-max_angle, max_angle)
- axis = random.choice([(0, 1), (0, 2), (1, 2)]) # 选择旋转平面
- volume = rotate(volume, angle=angle, axes=axis, reshape=False, order=1, mode='nearest')
- return volume
- def random_crop_3d(volume, crop_size=(64, 64, 64)):
- """
- 简单随机裁剪,假定 volume 形状 >= crop_size
- """
- z, y, x = volume.shape
- cz, cy, cx = crop_size
- z1 = random.randint(0, max(0, z - cz))
- y1 = random.randint(0, max(0, y - cy))
- x1 = random.randint(0, max(0, x - cx))
- return volume[z1:z1+cz, y1:y1+cy, x1:x1+cx]
- import torch
- from torch.utils.data import Dataset
- class SingleModality3DDataset(Dataset):
- """
- 用于 ECV model 或 T2 model:
- - image_paths: list[str], 指向 NIfTI 文件路径
- - labels: list[int] 或 list[float],二分类/风险标签
- - transform: 可选的 3D 数据增强函数
- """
- def __init__(self, image_paths, labels, transform=None, target_spacing=(3,3,3), crop_size=(64,64,64)):
- self.image_paths = image_paths
- self.labels = labels
- self.transform = transform
- self.target_spacing = target_spacing
- self.crop_size = crop_size
- def __len__(self):
- return len(self.image_paths)
- def __getitem__(self, idx):
- path = self.image_paths[idx]
- label = self.labels[idx]
- nii = nib.load(path)
- vol = resample_to_spacing(nii, target_spacing=self.target_spacing)
- vol = zscore_normalize(vol)
- # 数据增强
- if self.transform:
- vol = self.transform(vol)
- # 若需要裁剪
- vol = random_crop_3d(vol, crop_size=self.crop_size)
- # [D, H, W] -> [1, D, H, W]
- vol = np.expand_dims(vol, axis=0)
- vol_t = torch.from_numpy(vol).float()
- label_t = torch.tensor(label).float()
- return vol_t, label_t
- from PIL import Image
- import torchvision.transforms as T
- class Combined2DClinicalDataset(Dataset):
- """
- 用于 MobHy-Net:
- - 2D T2-FLAIR + 2D 虚拟 ECV + 临床特征
- """
- def __init__(self, t2_paths, ecv_paths, clinical_features, labels, img_size=224):
- self.t2_paths = t2_paths
- self.ecv_paths = ecv_paths
- self.clinical_features = clinical_features
- self.labels = labels
- self.img_transform = T.Compose([
- T.Resize((img_size, img_size)),
- T.RandomHorizontalFlip(),
- T.RandomRotation(10),
- T.ToTensor(),
- T.Normalize(mean=[0.5]*3, std=[0.5]*3),
- ])
- def __len__(self):
- return len(self.labels)
- def __getitem__(self, idx):
- t2_img = Image.open(self.t2_paths[idx]).convert('RGB')
- ecv_img = Image.open(self.ecv_paths[idx]).convert('RGB')
- t2_tensor = self.img_transform(t2_img) # [3, H, W]
- ecv_tensor = self.img_transform(ecv_img) # [3, H, W]
- clin = torch.tensor(self.clinical_features[idx], dtype=torch.float32)
- label = torch.tensor(self.labels[idx], dtype=torch.float32)
- return t2_tensor, ecv_tensor, clin, label
- import torch.nn as nn
- import math
- class PatchEmbedding3D(nn.Module):
- """
- 将 3D 体数据切分为 3D patches,并映射到 D 维嵌入空间。
- 输入: (B, C, D, H, W)
- 输出: (B, N_patches, embed_dim)
- """
- def __init__(self, in_channels=1, patch_size=(4,4,4), embed_dim=128):
- super().__init__()
- self.patch_size = patch_size
- self.proj = nn.Conv3d(
- in_channels,
- embed_dim,
- kernel_size=patch_size,
- stride=patch_size
- )
- def forward(self, x):
- # x: (B, C, D, H, W)
- x = self.proj(x) # (B, embed_dim, D', H', W')
- x = x.flatten(2) # (B, embed_dim, N_patches)
- x = x.transpose(1, 2) # (B, N_patches, embed_dim)
- return x
- class ViT3D(nn.Module):
- def __init__(
- self,
- in_channels=1,
- patch_size=(4,4,4),
- embed_dim=128,
- depth=6,
- num_heads=4,
- mlp_ratio=4.0,
- num_classes=1,
- use_cls_token=True,
- dropout=0.1
- ):
- super().__init__()
- self.use_cls_token = use_cls_token
- self.patch_embed = PatchEmbedding3D(in_channels, patch_size, embed_dim)
- # 这里先假设输入体裁剪为固定大小,例如 [1, 64, 64, 64]
- # 需要根据 patch_size 推出 N_patches
- dummy = torch.zeros(1, in_channels, 64, 64, 64)
- with torch.no_grad():
- n_patches = self.patch_embed(dummy).shape[1]
- # cls token
- if use_cls_token:
- self.cls_token = nn.Parameter(torch.zeros(1, 1, embed_dim))
- n_positions = n_patches + 1
- else:
- self.cls_token = None
- n_positions = n_patches
- # 位置编码
- self.pos_embed = nn.Parameter(torch.zeros(1, n_positions, embed_dim))
- nn.init.trunc_normal_(self.pos_embed, std=0.02)
- self.pos_drop = nn.Dropout(dropout)
- self.blocks = nn.ModuleList([
- TransformerEncoderBlock(embed_dim, num_heads, mlp_ratio, dropout)
- for _ in range(depth)
- ])
- self.norm = nn.LayerNorm(embed_dim)
- self.head = nn.Sequential(
- nn.LayerNorm(embed_dim),
- nn.Linear(embed_dim, num_classes)
- )
- def forward(self, x):
- # x: (B, C, D, H, W)
- B = x.shape[0]
- x = self.patch_embed(x) # (B, N, C)
- if self.use_cls_token:
- cls_tokens = self.cls_token.expand(B, -1, -1) # (B, 1, C)
- x = torch.cat((cls_tokens, x), dim=1) # (B, 1+N, C)
- x = x + self.pos_embed
- x = self.pos_drop(x)
- for blk in self.blocks:
- x = blk(x)
- x = self.norm(x)
- if self.use_cls_token:
- feat = x[:, 0] # (B, C)
- else:
- feat = x.mean(dim=1) # mean pooling
- logits = self.head(feat).squeeze(-1) # (B,) for num_classes=1
- return logits
- class MobHyNet(nn.Module):
- def __init__(self, clinical_in_dim, num_classes=1, img_feat_dim=512,
- attn_heads=4, dropout=0.1):
- super().__init__()
- # 两个 MobileNetV2 提取器
- self.t2_extractor = MobileNetFeatureExtractor(pretrained=True)
- self.ecv_extractor = MobileNetFeatureExtractor(pretrained=True)
- # 将通道从 2560 -> 512
- self.conv_reduce = nn.Conv2d(
- in_channels=self.t2_extractor.out_channels * 2,
- out_channels=img_feat_dim,
- kernel_size=1
- )
- self.bn_reduce = nn.BatchNorm2d(img_feat_dim)
- self.act = nn.ReLU(inplace=True)
- # 临床特征 -> 100 维
- self.clin_fc = nn.Sequential(
- nn.Linear(clinical_in_dim, 100),
- nn.ReLU(inplace=True),
- nn.Dropout(dropout)
- )
- # 将图像特征展平为 [B, N, C] 形式
- # 注意 N = Hf * Wf, C = img_feat_dim
- self.attn_block = SimpleSelfAttnBlock(embed_dim=img_feat_dim + 100,
- num_heads=attn_heads,
- dropout=dropout)
- self.classifier = nn.Sequential(
- nn.LayerNorm(img_feat_dim + 100),
- nn.Linear(img_feat_dim + 100, 128),
- nn.ReLU(inplace=True),
- nn.Dropout(dropout),
- nn.Linear(128, num_classes)
- )
- def forward(self, t2_img, ecv_img, clin_feat):
- # 图像特征
- t2_f = self.t2_extractor(t2_img) # (B, C, Hf, Wf)
- ecv_f = self.ecv_extractor(ecv_img) # (B, C, Hf, Wf)
- x = torch.cat([t2_f, ecv_f], dim=1) # (B, 2C, Hf, Wf)
- x = self.conv_reduce(x)
- x = self.bn_reduce(x)
- x = self.act(x) # (B, img_feat_dim, Hf, Wf)
- B, C, Hf, Wf = x.shape
- x = x.view(B, C, -1).transpose(1, 2) # (B, N=Hf*Wf, C)
- # 临床特征
- clin = self.clin_fc(clin_feat) # (B, 100)
- # 扩展到与 N 对齐(简单广播,每个 patch 拼同一临床向量)
- clin_expand = clin.unsqueeze(1).expand(B, x.size(1), 100) # (B, N, 100)
- x_comb = torch.cat([x, clin_expand], dim=-1) # (B, N, C+100)
- # 多头自注意力
- x_attn = self.attn_block(x_comb) # (B, N, C+100)
- # 这里使用 mean pooling 汇聚
- feat = x_attn.mean(dim=1) # (B, C+100)
- logits = self.classifier(feat).squeeze(-1) # (B,)
- return logits
Model.py at commit 1718af6, no license · at the source
Overview
- Department of Radiation Oncology, The First Affiliated Hospital with Nanjing Medical University,Nanjing, China
- The First School of Clinical Medicine, Nanjing Medical University,Nanjing, China
- Department of Oncology, Sir Run Run Hospital, Nanjing Medical University,Nanjing, China
- Department of Radiation Oncology, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University,Guangzhou, China
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 2 matches between paragraphs and lines of code.
410312774/MobHyNet
1718af6e8285a1080bcaa8c87e0779ee3eb380c3, 28 November 2025Availability: 1 check, the latest on 28 September 2026: the link answers
- 28 September 2026: the link answers
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:
- it points to the authors' code: 410312774/
MobHyNet
Read it in the paper: doi.org/10.1038/s41698-026-01475-1.
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;
- 1 script, each with its path and the digest of its content;
- 2 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.
Data availability statement
The paper has a data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:
- no repository, dataset or request procedure was recognized in it
Read it in the paper: doi.org/10.1038/s41698-026-01475-1.
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, 10 authors, 5 keywords, 49 references.
Cite
This paper
Jiang, G., Sun, X., Zhu, Y., Fei, Y., Xu, W., Jiang, Z., Shao, T., Cao, Y., Li, L., & Zhou, S. (2026). Radiomics model integrating MRI and ECV enhances prediction accuracy for progression in high-grade glioma. NPJ precision oncology, 10(1), 289. https://
BibTeX
@article{jiang2026radiom
author = {Jiang, Gefei and Sun, Xingjian and Zhu, Yuchen and Fei, Yinjiao and Xu, Weilin and Jiang, Zhichao and Shao, Tianchi and Cao, Yuandong and Li, Liting and Zhou, Shu},
title = {{Radiomics model integrating MRI and ECV enhances prediction accuracy for progression in high-grade glioma}},
journal = {NPJ precision oncology},
year = {2026},
month = may,
volume = {10},
number = {1},
pages = {289},
publisher = {Nature Publishing Group},
issn = {2397-768X},
doi = {10.1038/
url = {https://
pmid = {42129328},
pmcid = {PMC13402302}
}
RIS
TY - JOUR
AU - Jiang, Gefei
AU - Sun, Xingjian
AU - Zhu, Yuchen
AU - Fei, Yinjiao
AU - Xu, Weilin
AU - Jiang, Zhichao
AU - Shao, Tianchi
AU - Cao, Yuandong
AU - Li, Liting
AU - Zhou, Shu
TI - Radiomics model integrating MRI and ECV enhances prediction accuracy for progression in high-grade glioma
T2 - NPJ precision oncology
J2 - NPJ Precis Oncol
PY - 2026
DA - 2026/
VL - 10
IS - 1
SP - 289
SN - 2397-768X
PB - Nature Publishing Group
DO - 10.1038/
UR - https://
LA - en
ER -
CSL-JSON
{
"id": "10.1038/
"type": "article-journal",
"title": "Radiomics model integrating MRI and ECV enhances prediction accuracy for progression in high-grade glioma",
"container-title": "NPJ precision oncology",
"author": [
{
"family": "Jiang",
"given": "Gefei"
},
{
"family": "Sun",
"given": "Xingjian"
},
{
"family": "Zhu",
"given": "Yuchen"
},
{
"family": "Fei",
"given": "Yinjiao"
},
{
"family": "Xu",
"given": "Weilin"
},
{
"family": "Jiang",
"given": "Zhichao"
},
{
"family": "Shao",
"given": "Tianchi"
},
{
"family": "Cao",
"given": "Yuandong"
},
{
"family": "Li",
"given": "Liting"
},
{
"family": "Zhou",
"given": "Shu"
}
],
"container-title-short":
"volume": "10",
"issue": "1",
"page": "289",
"DOI": "10.1038/
"PMID": "42129328",
"PMCID": "PMC13402302",
"ISSN": "2397-768X",
"publisher": "Nature Publishing Group",
"URL": "https://
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
13
]
]
}
}
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:10.3389/fonc.2023.1143688 [code]
- Deep-learning and conventional radiomics to predict <i>IDH</
i> genotyping status based on magnetic resonance imaging data in adult diffuse glioma. Journal: Frontiers in oncologyIn common: scikit-image, Pillow, NiBabel, 3 other tools, structural MRI / diffusion, other condition, 2 references - [2] doi:10.1038/s41698-024-00632-8 [code]
- Combination of MRI-based prediction and CRISPR/
Cas12a-based detection for IDH genotyping in glioma. Journal: NPJ precision oncologyIn common: Pillow, PyTorch, SciPy, 1 other tool, structural MRI / diffusion, other condition, 3 references - [3] doi:10.3389/fnimg.2025.1630245 [code]
- A Bayesian deep segmentation framework for glioblastoma tumor segmentation using follow-up MRIs.Journal: Frontiers in neuroimagingIn common: scikit-image, Pillow, NiBabel, 3 other tools, structural MRI / diffusion, other condition, 1 reference
- [4] doi:10.1093/noajnl/vdae205 [code]
- Pediatric brain tumor classification using deep learning on MR images with age fusion.Journal: Neuro-oncology advancesIn common: Pillow, NiBabel, PyTorch, 2 other tools, clinical / translational, other condition, 2 references
- [5] doi:10.3389/fonc.2025.1497195 [code]
- Early characterization and prediction of glioblastoma and brain metastasis treatment efficacy using medical imaging-based radiomics and artificial intelligence algorithms.Journal: Frontiers in oncologyIn common: scikit-image, NiBabel, PyTorch, 1 other tool, clinical / translational, other condition, 2 references
- [6] doi:10.1038/s41598-022-16609-1 [code]
- AutoComBat: a generic method for harmonizing MRI-based radiomic features.Journal: Scientific reportsIn common: scikit-image, Pillow, PyTorch, 2 other tools, structural MRI / diffusion, other condition, 1 reference
- [7] doi:10.1038/s41467-024-52414-2 [code]
- Longitudinal deep neural networks for assessing metastatic brain cancer on a large open benchmark.Journal: Nature communicationsIn common: scikit-image, NiBabel, PyTorch, 2 other tools, other condition, 1 reference
- [8] doi:10.1038/s41698-025-00884-y [code]
- MRI transformer deep learning and radiomics for predicting IDH wild type TERT promoter mutant gliomas.Journal: NPJ precision oncologyIn common: Pillow, PyTorch, NumPy, structural MRI / diffusion, clinical / translational, other condition, 2 references
- [9] doi:10.1093/neuonc/noaf058 [code]
- Applications of artificial intelligence and advanced imaging in pediatric diffuse midline glioma.Journal: Neuro-oncologyIn common: NiBabel, PyTorch, SciPy, 1 other tool, structural MRI / diffusion, clinical / translational, other condition, 1 reference
- [10] doi:10.1093/nsr/nwae109 [code]
- NEATmap: a high-efficiency deep learning approach for whole mouse brain neuronal activity trace mapping.Journal: National science reviewIn common: scikit-image, Pillow, PyTorch, 2 other tools, 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.
Claim this paper
Correct its record
Say what each link of this record is, remove the ones that are not the paper's, add the ones that are missing. The correction becomes a new version of the record, in its Versions section.
Validate its tracing map
You validate the map as this page shows it: 1 repository of the authors' code, each at its verified commit and with its license, 1 script, and 2 matches between paragraphs and code (see the Code and Map sections). It then receives a DOI on Zenodo, with you (your ORCID iD) and OSCR as its creators; the code itself is not deposited.
The map's fingerprint: sha256:264ce655080eda9c…
Add the badge to the code's README
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 -->
[](https://openscicode.org/badge/doi/10.1038/s41698-026-01475-1)
<!-- oscr-badge:end -->reStructuredText
.. oscr-badge:start
.. image:: https://openscicode.org/badge/doi/10.1038/s41698-026-01475-1.svg
:target: https://openscicode.org/badge/doi/10.1038/s41698-026-01475-1
:alt: OSCR
.. oscr-badge:endHTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1038/s41698-026-01475-1"><img src="https://openscicode.org/badge/doi/10.1038/s41698-026-01475-1.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.
