Lan Hengrong, Jiang Daohuai, Yang Changchun, Gao Feng, Gao Fei
Hybrid Imaging System Laboratory, Shanghai Engineering Research Center of Intelligent Vision and Imaging, School of Information Science and Technology, ShanghaiTech University, Shanghai 201210, China.
Chinese Academy of Sciences, Shanghai Institute of Microsystem and Information Technology, Shanghai 200050, China.
Photoacoustics. 2020 Jun 20;20:100197. doi: 10.1016/j.pacs.2020.100197. eCollection 2020 Dec.
Conventional reconstruction algorithms (e.g., delay-and-sum) used in photoacoustic imaging (PAI) provide a fast solution while many artifacts remain, especially for limited-view with ill-posed problem. In this paper, we propose a new convolutional neural network (CNN) framework Y-Net: a CNN architecture to reconstruct the initial PA pressure distribution by optimizing both raw data and beamformed images once. The network combines two encoders with one decoder path, which optimally utilizes more information from raw data and beamformed image. We compared our result with some ablation studies, and the results of the test set show better performance compared with conventional reconstruction algorithms and other deep learning method (U-Net). Both and experiments are used to validated our method, which still performs better than other existing methods. The proposed Y-Net architecture also has high potential in medical image reconstruction for other imaging modalities beyond PAI.
光声成像（PAI）中使用的传统重建算法（如延迟求和法）能快速给出解决方案，但仍存在许多伪影，特别是对于存在不适定问题的有限视角情况。在本文中，我们提出了一种新的卷积神经网络（CNN）框架Y-Net：一种通过一次性优化原始数据和波束形成图像来重建初始光声压力分布的CNN架构。该网络将两个编码器与一条解码器路径相结合，从而最佳地利用来自原始数据和波束形成图像的更多信息。我们将我们的结果与一些消融研究进行了比较，测试集的结果表明，与传统重建算法和其他深度学习方法（U-Net）相比，性能更佳。同时使用了 和 实验来验证我们的方法，该方法仍然比其他现有方法表现更好。所提出的Y-Net架构在PAI之外的其他成像模态的医学图像重建中也具有很高的潜力。
