Reconstructing Undersampled Photoacoustic Microscopy Images using Deep Learning.

Published

Journal Article

One primary technical challenge in photoacoustic microscopy (PAM) is the necessary compromise between spatial resolution and imaging speed. In this study, we propose a novel application of deep learning principles to reconstruct undersampled PAM images and transcend the trade-off between spatial resolution and imaging speed. We compared various convolutional neural network (CNN) architectures, and selected a fully dense U-net (FD U-net) model that produced the best results. To mimic various undersampling conditions in practice, we artificially downsampled fully-sampled PAM images of mouse brain vasculature at different ratios. This allowed us to not only definitively establish the ground truth, but also train and test our deep learning model at various imaging conditions. Our results and numerical analysis have collectively demonstrated the robust performance of our model to reconstruct PAM images with as few as 2% of the original pixels, which can effectively shorten the imaging time without substantially sacrificing the image quality.

Full Text

Duke Authors

Cited Authors

  • DiSpirito Iii, A; Li, D; Vu, T; Chen, M; Zhang, D; Luo, J; Horstmeyer, R; Yao, J

Published Date

  • October 16, 2020

Published In

Volume / Issue

  • PP /

PubMed ID

  • 33064648

Pubmed Central ID

  • 33064648

Electronic International Standard Serial Number (EISSN)

  • 1558-254X

International Standard Serial Number (ISSN)

  • 0278-0062

Digital Object Identifier (DOI)

  • 10.1109/tmi.2020.3031541

Language

  • eng