Dense Dilated UNet: Deep Learning for 3D Photoacoustic Tomography Image Reconstruction
This work addresses image reconstruction challenges in medical imaging for photoacoustic tomography, but it is incremental as it modifies an existing CNN architecture.
The paper tackled the problem of reconstructing 3D photoacoustic tomography images from incomplete data, which causes artifacts and blurring, by proposing a Dense Dilation UNet (DD-UNet) that outperformed the Fully Dense UNet in image quality metrics and feature reconstruction.
In photoacoustic tomography (PAT), the acoustic pressure waves produced by optical excitation are measured by an array of detectors and used to reconstruct an image. Sparse spatial sampling and limited-view detection are two common challenges faced in PAT. Reconstructing from incomplete data using standard methods results in severe streaking artifacts and blurring. We propose a modified convolutional neural network (CNN) architecture termed Dense Dilation UNet (DD-UNet) for correcting artifacts in 3D PAT. The DD-Net leverages the benefits of dense connectivity and dilated convolutions to improve CNN performance. We compare the proposed CNN in terms of image quality as measured by the multiscale structural similarity index metric to the Fully Dense UNet (FD-UNet). Results demonstrate that the DD-Net consistently outperforms the FD-UNet and is able to more reliably reconstruct smaller image features.