论文《特定病人的单个投影试图的三维重建》相关知识笔记

what are coronal and sagittal views of the images？

Encoder-Decoder框架
所谓编码，就是将输入序列转化成一个固定长度的向量；解码，就是将之前生成的固定向量再转化成输出序列。
编码-存储-解码

---

Input :
Given a sequence of 2D projections denoted as { X 1 , X 2 , ⋯   , X N } \left\{X_{1}, X_{2}, \cdots, X_{N}\right\} {X1​,X2​,⋯,XN​} where $X_i\in {\mathbb{R}}^{H_{2\mathrm{D}}\times W_{2\mathrm{D}}}$ for all $1\le i\le N$ ,N is the number of given 2D projections. Each entry is a pixel-wise intensity value.

Output :
volumetric 3D image Y describing the corresponding 3D physical scene.
$Y_{\text{pred }}\in {\mathbb{R}}^{C_{3\mathrm{D}}\times H_{3\mathrm{D}}\times W_{3\mathrm{D}}}$, while $Y_{\text{truth }}\in {\mathbb{R}}^{C_{3\mathrm{D}}\times H_{3\mathrm{D}}\times W_{3\mathrm{D}}}$ is the ground- truth 3D imaIge as the reconstruction targIet. Each entry is a voxel-wise intensity value.

(目标是生成描述相应3D物理场景的体积3D图像Y。)

In order to use a sequence of $2\mathrm{D}$ projections as model input, we stack all the $2\mathrm{D}$ projections together as a single $3\mathrm{D}$ tensor. In other words, a set of $2\mathrm{D}$ projections { X 1 , X 2 , ⋯   , X N } ( X i ∈ R H 2 D × W 2 D ) \left\{X_{1}, X_{2}, \cdots, X_{N}\right\}\left(X_{i} \in \mathbb{R}^{H_{2 D} \times W_{2 \mathrm{D}}}\right) {X1​,X2​,⋯,XN​}(Xi​∈RH2D​×W2D​) are stacked as a 3D volume $\mathrm{Z}\in {\mathbb{R}}^{N\times H_{2D}\times W_{2\mathrm{D}}}$
where $N$ is the number of $2\mathrm{D}$ projections. In what follows, we introduce the model