mlsanityData to 7 Oct 2026

GloResNet: A lightweight 3D CNN with global topological features for preterm brain injury prediction

arXiv 2606.02498 ↗First posted 1 Jun 2026cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

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

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 18 of its 116 references that are in the map. First posted on arXiv 1 Jun 2026.

Builds on

18 of 116 references.

References · 18

  1. 2018Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networkscited by 278
  2. 20203D Deep Learning on Medical Images: A Reviewcited by 76
  3. 2021Transferable Visual Words: Exploiting the Semantics of Anatomical Patterns for Self-supervised Learningcited by 68
  4. 20243D MedDiffusion: A 3D Medical Latent Diffusion Model for Controllable and High-quality Medical Image Generationcited by 37
  5. 2020Models Genesiscited by 26
  6. 2021A persistent homology-based topological loss for CNN-based multi-class segmentation of CMRcited by 21
  7. 2021NDPNet: A novel non-linear data projection network for few-shot fine-grained image classificationcited by 13
  8. 2025Dynamic Accumulated Attention Map for Interpreting Evolution of Decision-Making in Vision Transformercited by 12
  9. 2023MiAMix: Enhancing Image Classification through a Multi-stage Augmented Mixed Sample Data Augmentation Methodcited by 8
  10. 2025Feature Complementation Architecture for Visual Place Recognitioncited by 4
  11. 2026Meningioma Analysis and Diagnosis using Limited Labeled Samplescited by 4
  12. 2024Few-Shot Learning for Annotation-Efficient Nucleus Instance Segmentationcited by 3
6 more
  1. 2024Neuron Abandoning Attention Flow: Visual Explanation of Dynamics inside CNN Modelscited by 3
  2. 2026Second-order Gaussian directional derivative representations for image high-resolution corner detectioncited by 3
  3. 2026Deep learning-based neurodevelopmental assessment in preterm infantscited by 3
  4. 2023Feature Activation Map: Visual Explanation of Deep Learning Models for Image Classificationcited by 2
  5. 2024The Developing Human Connectome Project: A Fast Deep Learning-based Pipeline for Neonatal Cortical Surface Reconstructioncited by 2
  6. 2026Adaptive receptive field-based spatial-frequency feature reconstruction network for fine-grained few-shot image classificationcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Phase-Aware Spatial-Frequency Fusion for Few-Shot Fine-Grained Image Classificationcited by 0