Please click here to download .
Tong Tong, Wenhui Huang, Kun Wang, Zicong He, Lin Yin, Xin Yang, Shuixing Zhang, Jie Tian
Medical image reconstruction methods based on deep learning have recently demonstrated powerful performance in photoacoustic tomography (PAT) from limited-view and sparse data. However, because most of these methods must utilize conventional linear reconstruction methods to implement signal-to-image transformations, their performance is restricted. In this paper, we propose a novel deep learning reconstruction approach that integrates appropriate data pre-processing and training strategies. The Feature Projection Network (FPnet) presented herein is designed to learn this signal-to-image transformation through data-driven learning rather than through direct use of linear reconstruction. To further improve reconstruction results, our method integrates an image post-processing network (U-net). Experiments show that the proposed method can achieve high reconstruction quality from limited-view data with sparse measurements. When employing GPU acceleration, this method can achieve a reconstruction speed of 15 frames per second.
All data used in our numerical simulations and in vivo experiments are open access. In our study, there are four datasets for numerical simulations and two datasets for in vivo experiments (see Table I for more detailed information). We only provide the ground truth images in Brain, Abdomen, Vessel and LiverCancer datasets. The user can use their own forward model to produce simulated signals. For in vivo dataset, both ground truth images and signals extracted from the MSOT inVision 128 system are provided. The images and signals are named by numbers.
The code of FPnet is open access here and you can download from here.
We welcome researchers to reproduce our results using our data and model code. If you use our dataset or code, please, cite the following paper:
Tong Tong, Wenhui Huang, Kun Wang, Zicong He, Lin Yin, Xin Yang, Shuixing Zhang, Jie Tian. Domain Transform Network for Photoacoustic Tomography from Limited-view and Sparsely Sampled Data. (accepted by Photoacoustics)