融合2.5D与3D模型，提升新冠肺CT诊断的跨源鲁棒性。
Halfway to 3D: Ensembling 2.5D and 3D Models for Robust COVID-19 CT Diagnosis
- 用多视角切片+体积建模双路并行，捕获局部与整体信息。
- 二分类准确率达94.48%，多分类最佳模型达79.35%准确率。
- 适合需跨设备、跨数据源的医疗影像分析场景。
我们提出一种深度学习框架，用于从胸部CT扫描中检测和分类新冠感染，通过融合2.5D与3D表征，捕捉切片级与体积分层信息。2.5D分支利用DINOv3视觉变换器处理轴向、冠状、矢状多视角切片，提取鲁棒特征；3D分支采用ResNet-18建模体积上下文，并通过方差风险外推（VREx）预训练结合监督对比学习，提升跨源鲁棒性。两分支预测在对数空间进行集成。在PHAROS-AIF-MIH基准测试中，二分类任务中集成模型达94.48%准确率与0.9426宏F1，优于单模型；多分类任务中，2.5D DINOv3模型表现最佳，准确率79.35%，宏F1为0.7497。结果表明，结合预训练切片表征与体积建模能有效提升多源医学影像分析的鲁棒性。代码已开源。
原文摘要 · Abstract (English)
We propose a deep learning framework for COVID-19 detection and disease classification from chest CT scans that integrates both 2.5D and 3D representations to capture complementary slice-level and volumetric information. The 2.5D branch processes multi-view CT slices (axial, coronal, sagittal) using a DINOv3 vision transformer to extract robust visual features, while the 3D branch employs a ResNet-18 architecture to model volumetric context and is pretrained with Variance Risk Extrapolation (VREx) followed by supervised contrastive learning to improve cross-source robustness. Predictions from both branches are combined through logit-level ensemble inference. Experiments on the PHAROS-AIF-MIH benchmark demonstrate the effectiveness of the proposed approach: for binary COVID-19 detection, the ensemble achieves 94.48% accuracy and a 0.9426 Macro F1-score, outperforming both individual models, while for multi-class disease classification the 2.5D DINOv3 model achieves the best performance with 79.35% accuracy and a 0.7497 Macro F1-score. These results highlight the benefit of combining pretrained slice-based representations with volumetric modeling for robust multi-source medical imaging analysis. Code is available at https://github.com/HySonLab/PHAROS-AIF-MIH
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。