Photoacoustic microscopy with sparse data by convolutional neural networks

Photoacoustic microscopy with sparse data by convolutional neural networks

The point-by-point scanning mechanism of photoacoustic microscopy (PAM) results in low-speed imaging, limiting the application of PAM. In this work, we propose a method to improve the quality of sparse PAM images using convolutional neural networks (CNNs), thereby speeding up image acquisition while...

Full description

Bibliographic Details
Main Authors: Jiasheng Zhou, Da He, Xiaoyu Shang, Zhendong Guo, Sung-Liang Chen, Jiajia Luo
Format: Article
Language:English
Published: Elsevier 2021-06-01
Series:Photoacoustics
Subjects:
Photoacoustic microscopy
Convolutional neural network
Sparse image
Image enhancement
Online Access:http://www.sciencedirect.com/science/article/pii/S2213597921000045
id doaj-8cd61704ab1e4f768c6e1a3296f2a317
record_format Article
spelling doaj-8cd61704ab1e4f768c6e1a3296f2a3172021-05-26T04:26:33ZengElsevierPhotoacoustics2213-59792021-06-0122100242Photoacoustic microscopy with sparse data by convolutional neural networksJiasheng Zhou0Da He1Xiaoyu Shang2Zhendong Guo3Sung-Liang Chen4Jiajia Luo5University of Michigan-Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, ChinaUniversity of Michigan-Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, ChinaUniversity of Michigan-Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, ChinaUniversity of Michigan-Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, ChinaUniversity of Michigan-Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai 200240, China; Corresponding authors.Biomedical Engineering Department, Peking University, Beijing 100191, China; Corresponding authors.The point-by-point scanning mechanism of photoacoustic microscopy (PAM) results in low-speed imaging, limiting the application of PAM. In this work, we propose a method to improve the quality of sparse PAM images using convolutional neural networks (CNNs), thereby speeding up image acquisition while maintaining good image quality. The CNN model utilizes attention modules, residual blocks, and perceptual losses to reconstruct the sparse PAM image, which is a mapping from a 1/4 or 1/16 low-sampling sparse PAM image to a latent fully-sampled one. The model is trained and validated mainly on PAM images of leaf veins, showing effective improvements quantitatively and qualitatively. Our model is also tested using in vivo PAM images of blood vessels of mouse ears and eyes. The results suggest that the model can enhance the quality of the sparse PAM image of blood vessels in several aspects, which facilitates fast PAM and its clinical applications.http://www.sciencedirect.com/science/article/pii/S2213597921000045Photoacoustic microscopyConvolutional neural networkSparse imageImage enhancement
collection DOAJ
language English
format Article
sources DOAJ
author Jiasheng Zhou
Da He
Xiaoyu Shang
Zhendong Guo
Sung-Liang Chen
Jiajia Luo
spellingShingle Jiasheng Zhou
Da He
Xiaoyu Shang
Zhendong Guo
Sung-Liang Chen
Jiajia Luo
Photoacoustic microscopy with sparse data by convolutional neural networks
Photoacoustics
Photoacoustic microscopy
Convolutional neural network
Sparse image
Image enhancement
author_facet Jiasheng Zhou
Da He
Xiaoyu Shang
Zhendong Guo
Sung-Liang Chen
Jiajia Luo
author_sort Jiasheng Zhou
title Photoacoustic microscopy with sparse data by convolutional neural networks
title_short Photoacoustic microscopy with sparse data by convolutional neural networks
title_full Photoacoustic microscopy with sparse data by convolutional neural networks
title_fullStr Photoacoustic microscopy with sparse data by convolutional neural networks
title_full_unstemmed Photoacoustic microscopy with sparse data by convolutional neural networks
title_sort photoacoustic microscopy with sparse data by convolutional neural networks
publisher Elsevier
series Photoacoustics
issn 2213-5979
publishDate 2021-06-01
description The point-by-point scanning mechanism of photoacoustic microscopy (PAM) results in low-speed imaging, limiting the application of PAM. In this work, we propose a method to improve the quality of sparse PAM images using convolutional neural networks (CNNs), thereby speeding up image acquisition while maintaining good image quality. The CNN model utilizes attention modules, residual blocks, and perceptual losses to reconstruct the sparse PAM image, which is a mapping from a 1/4 or 1/16 low-sampling sparse PAM image to a latent fully-sampled one. The model is trained and validated mainly on PAM images of leaf veins, showing effective improvements quantitatively and qualitatively. Our model is also tested using in vivo PAM images of blood vessels of mouse ears and eyes. The results suggest that the model can enhance the quality of the sparse PAM image of blood vessels in several aspects, which facilitates fast PAM and its clinical applications.
topic Photoacoustic microscopy
Convolutional neural network
Sparse image
Image enhancement
url http://www.sciencedirect.com/science/article/pii/S2213597921000045
work_keys_str_mv AT jiashengzhou photoacousticmicroscopywithsparsedatabyconvolutionalneuralnetworks
AT dahe photoacousticmicroscopywithsparsedatabyconvolutionalneuralnetworks
AT xiaoyushang photoacousticmicroscopywithsparsedatabyconvolutionalneuralnetworks
AT zhendongguo photoacousticmicroscopywithsparsedatabyconvolutionalneuralnetworks
AT sungliangchen photoacousticmicroscopywithsparsedatabyconvolutionalneuralnetworks
AT jiajialuo photoacousticmicroscopywithsparsedatabyconvolutionalneuralnetworks
_version_ 1721426639875735552

Similar Items