光学学报, 2019, 39 (6): 0617001, 网络出版: 2019-06-17
基于自编码器的荧光分子断层成像快速重建 下载: 1062次
Fast Reconstruction Method for Fluorescence Molecular Tomography Based on Autoencoder
医用光学 荧光分子断层成像 数据降维 深度学习 自编码器 图像重建 medical optics fluorescence molecular tomography data dimensionality reduction deep learning autoencoder image reconstruction
摘要
多激发点荧光分子断层成像(FMT)重建过程中生成的系统矩阵规模较大,导致计算复杂度高,重建时间长。为了加快重建速度并保证其准确性,基于人工神经网络理论,通过降低系统矩阵规模,提出了一种快速FMT重建方法。具体来说,采用的降维方法是自编码器,即一种典型的人工神经网络,训练数据为由系统矩阵和表面荧光测量值组成的矩阵,然后使用自编码器网络的编码部分得到原始矩阵在低维空间上的表示。为了测试所提方法的性能,设计了一系列数值模拟实验,包括非匀质圆柱体实验和数字鼠实验。实验结果表明,该方法能有效缩短重建时间,得到较高的重建精度。
Abstract
The large-scale system matrix generated during the reconstruction procedure of multiple excitation points based on fluorescence molecular tomography (FMT) leads to the high computational complexity and long reconstruction time. In order to shorten the reconstruction time and ensure its accuracy, based on the theory of artificial neural network (ANN), we propose a fast reconstruction method for FMT by reducing the dimension of system matrix in this paper. Specifically, the dimension reduction tool is the autoencoder (AE), which is a famous ANN architecture, and during the training of AE, the input matrix data consists of system matrix and surface fluorescence measurement data, then the representation of the previous matrix in the lower dimensional space is obtained by utilizing encoder part of AE. To test the performance of our method, a series numerical simulation experiments are devised, including non-heterogeneous cylinder and digital mouse experiments. Experimental results demonstrate that our method can effectively shorten the time of FMT reconstruction as well as obtain a good reconstruction accuracy.
卢笛, 卫潇, 曹欣, 贺小伟, 侯榆青. 基于自编码器的荧光分子断层成像快速重建[J]. 光学学报, 2019, 39(6): 0617001. Di Lu, Xiao Wei, Xin Cao, Xiaowei He, Yuqing Hou. Fast Reconstruction Method for Fluorescence Molecular Tomography Based on Autoencoder[J]. Acta Optica Sinica, 2019, 39(6): 0617001.