In the field of medical imaging, 3D deep learning models play a crucial role
in building powerful predictive models of disease progression. However, the
size of these models presents significant challenges, both in terms of
computational resources and data requirements. Moreover, achieving high-quality
pretraining of 3D models proves to be even more challenging. To address these
issues, hybrid 2.5D approaches provide an effective solution for utilizing 3D
volumetric data efficiently using 2D models. Combining 2D and 3D techniques
offers a promising avenue for optimizing performance while minimizing memory
requirements. In this paper, we explore 2.5D architectures based on a
combination of convolutional neural networks (CNNs), long short-term memory
(LSTM), and Transformers. In addition, leveraging the benefits of recent
non-contrastive pretraining approaches in 2D, we enhanced the performance and
data efficiency of 2.5D techniques even further. We demonstrate the
effectiveness of architectures and associated pretraining on a task of
predicting progression to wet age-related macular degeneration (AMD) within a
six-month period on two large longitudinal OCT datasets.

在医学影像领域，3D 深度学习模型在构建疾病进展预测模型方面起着重要作用。然而，这些模型的规模在计算资源和数据需求方面带来了重大挑战。为了解决这些问题，2.5D 方法提供了一种有效的利用 3D 体积数据的方式。结合 2D 和 3D 技术为优化性能并降低内存需求提供了一条有前景的途径。在本文中，我们探索基于卷积神经网络（CNNs），长短期记忆（LSTM）和 Transformer 的 2.5D 架构。此外，利用近期的 2D 非对比性预训练方法，我们进一步提高了 2.5D 技术的性能和数据效率。我们在两个大型纵向光学相干断层扫描（OCT）数据集上展示了架构和相关预训练在预测湿性年龄相关性黄斑变性（AMD）的六个月内进展方面的有效性。

视网膜 OCT 高效预测模型的预训练 2.5D 深度模型

Pretrained Deep 2.5D Models for Efficient Predictive Modeling from  Retinal OCT

3D deep learning models are shown to be as vulnerable to adversarial examples
as 2D models. However, existing attack methods are still far from stealthy and
suffer from severe performance degradation in the physical world. Although 3D
data is highly structured, it is difficult to bound the perturbations with
simple metrics in the Euclidean space. In this paper, we propose a novel
$\epsilon$-isometric ($\epsilon$-ISO) attack to generate natural and robust 3D
adversarial examples in the physical world by considering the geometric
properties of 3D objects and the invariance to physical transformations. For
naturalness, we constrain the adversarial example to be $\epsilon$-isometric to
the original one by adopting the Gaussian curvature as a surrogate metric
guaranteed by a theoretical analysis. For invariance to physical
transformations, we propose a maxima over transformation (MaxOT) method that
actively searches for the most harmful transformations rather than random ones
to make the generated adversarial example more robust in the physical world.
Experiments on typical point cloud recognition models validate that our
approach can significantly improve the attack success rate and naturalness of
the generated 3D adversarial examples than the state-of-the-art attack methods.

本文提出了一种新颖的 epsilon - 同构攻击方法，通过考虑 3D 物体的几何特性和物理变换的不变性，在物理世界中生成自然和强健的 3D 对抗性样本，并借助 MaxOT 方法来提高攻击成功率和自然性。实验证明，相较于现有攻击方法，我们的方法能够显著提高 3D 对抗性样本的攻击成功率和自然性。