Fully Dense UNet for 2D Sparse Photoacoustic Tomography Artifact Removal

TL;DR

This paper introduces a Fully Dense UNet architecture designed to effectively remove artifacts from 2D photoacoustic tomography images reconstructed from sparsely sampled data, improving image quality over standard methods.

Contribution

The paper presents a novel Fully Dense UNet architecture specifically tailored for artifact removal in sparse 2D PAT images, demonstrating its superiority over standard UNet.

Findings

FD-UNet outperforms standard UNet in artifact removal quality.

The proposed method significantly improves reconstructed image clarity.

The architecture effectively handles sparse data reconstruction challenges.

Abstract

Photoacoustic imaging is an emerging imaging modality that is based upon the photoacoustic effect. In photoacoustic tomography (PAT), the induced acoustic pressure waves are measured by an array of detectors and used to reconstruct an image of the initial pressure distribution. A common challenge faced in PAT is that the measured acoustic waves can only be sparsely sampled. Reconstructing sparsely sampled data using standard methods results in severe artifacts that obscure information within the image. We propose a modified convolutional neural network (CNN) architecture termed Fully Dense UNet (FD-UNet) for removing artifacts from 2D PAT images reconstructed from sparse data and compare the proposed CNN with the standard UNet in terms of reconstructed image quality.