Annu Int Conf IEEE Eng Med Biol Soc. 2022 Jul;2022:2149-2152. doi: 10.1109/EMBC48229.2022.9871389.
Maximum intensity projection (MIP) is a standard volume-rendering technique for 3D volumetric data processing. For example, given a 3D CT data, it simply projects the voxel values with its maximum intensity on a specific view to output a 2D image. Recently, MIP is further combined with Btrfly Net for vertebrae labelling task. However, this simple reformations of 3D data leads to loss of rich context information in volumetric data. In this paper, we propose a learned orthographic pooling approach instead of image processing based MIP. Typically, a simple conv-simple and bottleneck pooling modules are introduced to learn the orthographic projection of 3D data and output 2D intermediate feature maps. To this end, the learned orthographic pooling helps preserve detail information of 3D context during projection. Furthermore, an unified Btrfly Net is provided for vertebrae labelling by integrating the orthographic pooling sub-network. The novel Btrfly Net with orthographic pooling sub-network is evaluated on the 2014 MICCAI vertebra localization challenge dataset. Compared to original Butfly Net with MIP, orthographic pooling, the learned MIP largely boosts the performance of vertebrae labelling.
最大强度投影（MIP）是一种用于处理 3D 体数据的标准体绘制技术。例如，给定一个 3D CT 数据，它只需将具有最大强度的体素值投影到特定视图上，以输出 2D 图像。最近，MIP 进一步与 Btrfly Net 结合用于脊椎标签任务。然而，这种简单的 3D 数据变换会导致体积数据中丰富的上下文信息丢失。在本文中，我们提出了一种基于学习的正射投影方法，而不是基于图像处理的 MIP。通常，引入一个简单的卷积-简单和瓶颈池化模块来学习 3D 数据的正射投影，并输出 2D 中间特征图。为此，学习到的正射投影有助于在投影过程中保留 3D 上下文的细节信息。此外，通过集成正射池化子网，提供了一个统一的 Btrfly Net 用于脊椎标签。在 2014 年 MICCAI 脊椎定位挑战赛数据集上对具有正射池化子网的新型 Btrfly Net 进行了评估。与具有 MIP 的原始 Btrfly Net 相比，学习到的 MIP 极大地提高了脊椎标签的性能。
