Cargando…

A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements

Photoacoustic tomography (PAT) is a propitious imaging modality, which is helpful for biomedical study. However, fast PAT imaging and denoising is an exigent task in medical research. To address the problem, recently, methods based on compressed sensing (CS) have been proposed, which accede the low...

Descripción completa

Detalles Bibliográficos
Autores principales: Shahid, Husnain, Khalid, Adnan, Liu, Xin, Irfan, Muhammad, Ta, Dean
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7943731/
https://www.ncbi.nlm.nih.gov/pubmed/33716643
http://dx.doi.org/10.3389/fnins.2021.598693
_version_ 1783662552572493824
author Shahid, Husnain
Khalid, Adnan
Liu, Xin
Irfan, Muhammad
Ta, Dean
author_facet Shahid, Husnain
Khalid, Adnan
Liu, Xin
Irfan, Muhammad
Ta, Dean
author_sort Shahid, Husnain
collection PubMed
description Photoacoustic tomography (PAT) is a propitious imaging modality, which is helpful for biomedical study. However, fast PAT imaging and denoising is an exigent task in medical research. To address the problem, recently, methods based on compressed sensing (CS) have been proposed, which accede the low computational cost and high resolution for implementing PAT. Nevertheless, the imaging results of the sparsity-based methods strictly rely on sparsity and incoherence conditions. Furthermore, it is onerous to ensure that the experimentally acquired photoacoustic data meets CS’s prerequisite conditions. In this work, a deep learning–based PAT (Deep-PAT)method is instigated to overcome these limitations. By using a neural network, Deep-PAT is not only able to reconstruct PAT from a fewer number of measurements without considering the prerequisite conditions of CS, but also can eliminate undersampled artifacts effectively. The experimental results demonstrate that Deep-PAT is proficient at recovering high-quality photoacoustic images using just 5% of the original measurement data. Besides this, compared with the sparsity-based method, it can be seen through statistical analysis that the quality is significantly improved by 30% (approximately), having average SSIM = 0.974 and PSNR = 29.88 dB with standard deviation ±0.007 and ±0.089, respectively, by the proposed Deep-PAT method. Also, a comparsion of multiple neural networks provides insights into choosing the best one for further study and practical implementation.
format Online
Article
Text
id pubmed-7943731
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-79437312021-03-11 A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements Shahid, Husnain Khalid, Adnan Liu, Xin Irfan, Muhammad Ta, Dean Front Neurosci Neuroscience Photoacoustic tomography (PAT) is a propitious imaging modality, which is helpful for biomedical study. However, fast PAT imaging and denoising is an exigent task in medical research. To address the problem, recently, methods based on compressed sensing (CS) have been proposed, which accede the low computational cost and high resolution for implementing PAT. Nevertheless, the imaging results of the sparsity-based methods strictly rely on sparsity and incoherence conditions. Furthermore, it is onerous to ensure that the experimentally acquired photoacoustic data meets CS’s prerequisite conditions. In this work, a deep learning–based PAT (Deep-PAT)method is instigated to overcome these limitations. By using a neural network, Deep-PAT is not only able to reconstruct PAT from a fewer number of measurements without considering the prerequisite conditions of CS, but also can eliminate undersampled artifacts effectively. The experimental results demonstrate that Deep-PAT is proficient at recovering high-quality photoacoustic images using just 5% of the original measurement data. Besides this, compared with the sparsity-based method, it can be seen through statistical analysis that the quality is significantly improved by 30% (approximately), having average SSIM = 0.974 and PSNR = 29.88 dB with standard deviation ±0.007 and ±0.089, respectively, by the proposed Deep-PAT method. Also, a comparsion of multiple neural networks provides insights into choosing the best one for further study and practical implementation. Frontiers Media S.A. 2021-02-24 /pmc/articles/PMC7943731/ /pubmed/33716643 http://dx.doi.org/10.3389/fnins.2021.598693 Text en Copyright © 2021 Shahid, Khalid, Liu, Irfan and Ta. http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Shahid, Husnain
Khalid, Adnan
Liu, Xin
Irfan, Muhammad
Ta, Dean
A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements
title A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements
title_full A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements
title_fullStr A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements
title_full_unstemmed A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements
title_short A Deep Learning Approach for the Photoacoustic Tomography Recovery From Undersampled Measurements
title_sort deep learning approach for the photoacoustic tomography recovery from undersampled measurements
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7943731/
https://www.ncbi.nlm.nih.gov/pubmed/33716643
http://dx.doi.org/10.3389/fnins.2021.598693
work_keys_str_mv AT shahidhusnain adeeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT khalidadnan adeeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT liuxin adeeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT irfanmuhammad adeeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT tadean adeeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT shahidhusnain deeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT khalidadnan deeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT liuxin deeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT irfanmuhammad deeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements
AT tadean deeplearningapproachforthephotoacoustictomographyrecoveryfromundersampledmeasurements