**Accelerated Correction of Reflection Artifacts by Deep Neural Networks in Photo-Acoustic Tomography**

#### **Hongming Shan 1, Ge Wang 1 and Yang Yang 2,\***


Received: 20 May 2019; Accepted: 18 June 2019; Published: 28 June 2019

**Abstract:** Photo-Acoustic Tomography (PAT) is an emerging non-invasive hybrid modality driven by a constant yearning for superior imaging performance. The image quality, however, hinges on the acoustic reflection, which may compromise the diagnostic performance. To address this challenge, we propose to incorporate a deep neural network into conventional iterative algorithms to accelerate and improve the correction of reflection artifacts. Based on the simulated PAT dataset from computed tomography (CT) scans, this network-accelerated reconstruction approach is shown to outperform two state-of-the-art iterative algorithms in terms of the peak signal-to-noise ratio (PSNR) and the structural similarity (SSIM) in the presence of noise. The proposed network also demonstrates considerably higher computational efficiency than conventional iterative algorithms, which are time-consuming and cumbersome.

**Keywords:** photo-acoustic tomography; reflection artifacts; deep learning; convolutional neural network; time reversal; Landweber algorithm; U-net
