Abstract
We propose a model-driven projected algebraic reconstruction technique (PART)-network (PART-Net) that leverages the advantages of the traditional model-based method and the neural network to improve the imaging quality of diffuse fluorescence tomography. In this algorithm, nonnegative prior information is incorporated into the ART iteration process to better guide the optimization process, and thereby improve imaging quality. On this basis, PART in conjunction with a residual convolutional neural network is further proposed to obtain high-fidelity image reconstruction. The numerical simulation results demonstrate that the PART-Net algorithm effectively improves noise robustness and reconstruction accuracy by at least 1–2 times and exhibits superiority in spatial resolution and quantification, especially for a small-sized target ($r = 2\;{\rm mm}$), compared with the traditional ART algorithm. Furthermore, the phantom and in vivo experiments verify the effectiveness of the PART-Net, suggesting strong generalization capability and a great potential for practical applications.
© 2024 Optica Publishing Group
Full Article | PDF ArticleMore Like This
Lingxiu Xing, Limin Zhang, Wenjing Sun, Zhuanxia He, Yanqi Zhang, and Feng Gao
Biomed. Opt. Express 15(4) 2078-2093 (2024)
Xia Cheng, Siyu Sun, Yinglong Xiao, Wenjing Li, Jintao Li, Jingjing Yu, and Hongbo Guo
J. Opt. Soc. Am. A 41(5) 844-851 (2024)
Xuanxuan Zhang, Yunfei Jia, Jiapei Cui, Jiulou Zhang, Xu Cao, Lin Zhang, and Guanglei Zhang
J. Opt. Soc. Am. A 40(7) 1359-1371 (2023)