arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.02498cs.CV
收藏

GloResNet：一种用于早产儿脑损伤预测的轻量级3D CNN与全局拓扑特征

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

Image Computing Laboratory, Shaanxi University of Science and Technology（陕西科技大学图像计算实验室）；Department of Neonatology, Shenzhen University of Advanced Technology General Hospital（深圳先进技术医院新生儿科）；Department of Neurosurgery, The First Affiliated Hospital of Xi’an Jiaotong University（西安交通大学第一附属医院神经外科）；CSIRO Technology（澳大利亚CSIRO技术）

Boyu Yuan, Jiamiao Lu, Weichuan Zhang, Benqing Wu, Tuo Wang, Changshan Wang, Changming Sun, Liang Guo

阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出基于ResNet-10的轻量级3D CNN GloResNet，结合全局流形映射和预处理策略，在dHCP数据集上实现早产儿脑损伤预测，平均准确率75.18%。

摘要：

本研究引入了一个自动化深度学习框架，用于从T2加权MRI（dHCP数据集）预测早产儿脑损伤（BI）。我们提出了GloResNet，一种基于ResNet-10的轻量级3D CNN，并在MedicalNet上预训练以应对数据稀缺。一种全局流形映射策略首先将每个3D体积重采样为128x128x128，然后应用逐样本z分数强度归一化，从而在标准化外观的同时保留全局拓扑。训练集成了mixup、类别加权和测试时增强以提高鲁棒性。在5折交叉验证中，GloResNet达到了75.18%的平均准确率（峰值81.82%），特异性0.81，敏感性0.76。结果表明，拓扑感知的轻量级CNN能够有效预测新生儿脑损伤，提供了一种非侵入性筛查工具。本文源代码可从GitHub仓库获取：https://github.com/ICL-SUST/GloResNet-Preterm-Brain

英文摘要：

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

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑