GloResNet: A lightweight 3D CNN with global topological features for preterm brain injury prediction
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913180561702912 |
|---|---|
| author | Yuan, Boyu Lu, Jiamiao Zhang, Weichuan Wu, Benqing Wang, Tuo Wang, Changshan Sun, Changming Guo, Liang |
| author_facet | Yuan, Boyu Lu, Jiamiao Zhang, Weichuan Wu, Benqing Wang, Tuo Wang, Changshan Sun, Changming Guo, Liang |
| contents | This study introduces an automated deep learning framework for predicting brain injury (BI) in preterm infants from T2-weighted MRI (dHCP dataset). We propose GloResNet, a lightweight 3D CNN based on ResNet-10, pretrained on MedicalNet to address data scarcity. A global manifold mapping strategy first resamples each 3D volume to 128x128x128 and then applies subject-wise z-score intensity normalization, thereby preserving global topology while standardizing appearance. Training integrates mixup, class weighting, and test-time augmentation for robustness. In 5-fold cross-validation, GloResNet achieved 75.18% average accuracy (peak 81.82%), with specificity 0.81 and sensitivity 0.76. Results demonstrate that a topology-aware lightweight CNN has the capability to effectively predict neonatal BI, offering a non-invasive screening tool. The source code of this paper can be obtained from the GitHub repository: https://github.com/ICL-SUST/GloResNet-Preterm-Brain |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_02498 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GloResNet: A lightweight 3D CNN with global topological features for preterm brain injury prediction Yuan, Boyu Lu, Jiamiao Zhang, Weichuan Wu, Benqing Wang, Tuo Wang, Changshan Sun, Changming Guo, Liang Computer Vision and Pattern Recognition This study introduces an automated deep learning framework for predicting brain injury (BI) in preterm infants from T2-weighted MRI (dHCP dataset). We propose GloResNet, a lightweight 3D CNN based on ResNet-10, pretrained on MedicalNet to address data scarcity. A global manifold mapping strategy first resamples each 3D volume to 128x128x128 and then applies subject-wise z-score intensity normalization, thereby preserving global topology while standardizing appearance. Training integrates mixup, class weighting, and test-time augmentation for robustness. In 5-fold cross-validation, GloResNet achieved 75.18% average accuracy (peak 81.82%), with specificity 0.81 and sensitivity 0.76. Results demonstrate that a topology-aware lightweight CNN has the capability to effectively predict neonatal BI, offering a non-invasive screening tool. The source code of this paper can be obtained from the GitHub repository: https://github.com/ICL-SUST/GloResNet-Preterm-Brain |
| title | GloResNet: A lightweight 3D CNN with global topological features for preterm brain injury prediction |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2606.02498 |