Single Frame Laser Diode Photoacoustic Imaging: Denoising and Reconstruction

TL;DR

This paper introduces a deep learning approach for denoising and reconstructing photoacoustic images obtained from low-cost laser diodes, enabling high-quality imaging with minimal frames and low signal-to-noise ratios.

Contribution

It presents novel deep learning methods for denoising and reconstructing PA images from very few frames, improving image quality without increasing laser exposure.

Findings

Effective denoising of PA images with single or few frames

Automatic reconstruction of point sources from noisy data

Enhanced image quality at low signal-to-noise ratios

Abstract

A new development in photoacoustic (PA) imaging has been the use of compact, portable and low-cost laser diodes (LDs), but LD-based PA imaging suffers from low signal intensity recorded by the conventional transducers. A common method to improve signal strength is temporal averaging, which reduces frame rate and increases laser exposure to patients. To tackle this problem, we propose a deep learning method that will denoise the PA images before beamforming with a very few frames, even one. We also present a deep learning method to automatically reconstruct point sources from noisy pre-beamformed data. Finally, we employ a strategy of combined denoising and reconstruction, which can supplement the reconstruction algorithm for very low signal-to-noise ratio inputs.