Abstract
Conventional photoacoustic imaging may suffer from the limited view and bandwidth of ultrasound transducers. A deep learning approach is proposed to handle these problems and is demonstrated both in simulations and in experiments on a multi-scale model of leaf skeleton. We employed an experimental approach to build the training and the test sets using photographs of the samples as ground truth images. Reconstructions produced by the neural network show a greatly improved image quality as compared to conventional approaches. In addition, this work aimed at quantifying the reliability of the neural network predictions. To achieve this, the dropout Monte-Carlo procedure is applied to estimate a pixel-wise degree of confidence on each predicted picture. Last, we address the possibility to use transfer learning with simulated data in order to drastically limit the size of the experimental dataset.
Highlights
Photoacoustic (PA) imaging is an emerging biomedical modality based on the generation of acoustic waves by light absorption
The deep learning reconstruction yields an almost artefact-free reconstruction with errors located only on the smallest appendages resulting from the manual cutting, and on few vertical structures which are not completely recovered
The displayed shifted structured similarity index (sSSIM) values are an average over all the test sets from the 30 different realizations. We repeated this procedure with weights initialized by those obtained at the end of a training session on a simulation dataset composed of 1400 pairs
Summary
Photoacoustic (PA) imaging is an emerging biomedical modality based on the generation of acoustic waves by light absorption. At the US propagation time scale, the object illumination is quasi instantaneous as the speed of light is several orders of magnitude higher than the speed of sound, resulting in the emission of strongly coherent acoustics waves [2]. These waves interfere constructively or destructively depending on the structure of the object, often leading to two well-known artefacts on the reconstructed image: the limited bandwidth and the limited view artefacts [3]. Both type of artefacts will further be referred to as the visibility problem in this paper
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
