Cargando…

High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning

Significance: Reducing the bit depth is an effective approach to lower the cost of an optical coherence tomography (OCT) imaging device and increase the transmission efficiency in data acquisition and telemedicine. However, a low bit depth will lead to the degradation of the detection sensitivity, t...

Descripción completa

Detalles Bibliográficos
Autores principales: Hao, Qiangjiang, Zhou, Kang, Yang, Jianlong, Hu, Yan, Chai, Zhengjie, Ma, Yuhui, Liu, Gangjun, Zhao, Yitian, Gao, Shenghua, Liu, Jiang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Society of Photo-Optical Instrumentation Engineers 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7666869/
https://www.ncbi.nlm.nih.gov/pubmed/33191687
http://dx.doi.org/10.1117/1.JBO.25.12.123702
_version_ 1783610217650454528
author Hao, Qiangjiang
Zhou, Kang
Yang, Jianlong
Hu, Yan
Chai, Zhengjie
Ma, Yuhui
Liu, Gangjun
Zhao, Yitian
Gao, Shenghua
Liu, Jiang
author_facet Hao, Qiangjiang
Zhou, Kang
Yang, Jianlong
Hu, Yan
Chai, Zhengjie
Ma, Yuhui
Liu, Gangjun
Zhao, Yitian
Gao, Shenghua
Liu, Jiang
author_sort Hao, Qiangjiang
collection PubMed
description Significance: Reducing the bit depth is an effective approach to lower the cost of an optical coherence tomography (OCT) imaging device and increase the transmission efficiency in data acquisition and telemedicine. However, a low bit depth will lead to the degradation of the detection sensitivity, thus reducing the signal-to-noise ratio (SNR) of OCT images. Aim: We propose using deep learning to reconstruct high SNR OCT images from low bit-depth acquisition. Approach: The feasibility of our approach is evaluated by applying this approach to the quantized 3- to 8-bit data from native 12-bit interference fringes. We employ a pixel-to-pixel generative adversarial network (pix2pixGAN) architecture in the low-to-high bit-depth OCT image transition. Results: Extensively, qualitative and quantitative results show our method could significantly improve the SNR of the low bit-depth OCT images. The adopted pix2pixGAN is superior to other possible deep learning and compressed sensing solutions. Conclusions: Our work demonstrates that the proper integration of OCT and deep learning could benefit the development of healthcare in low-resource settings.
format Online
Article
Text
id pubmed-7666869
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher Society of Photo-Optical Instrumentation Engineers
record_format MEDLINE/PubMed
spelling pubmed-76668692020-11-23 High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning Hao, Qiangjiang Zhou, Kang Yang, Jianlong Hu, Yan Chai, Zhengjie Ma, Yuhui Liu, Gangjun Zhao, Yitian Gao, Shenghua Liu, Jiang J Biomed Opt Special Series on Biomedical Imaging and Sensing Significance: Reducing the bit depth is an effective approach to lower the cost of an optical coherence tomography (OCT) imaging device and increase the transmission efficiency in data acquisition and telemedicine. However, a low bit depth will lead to the degradation of the detection sensitivity, thus reducing the signal-to-noise ratio (SNR) of OCT images. Aim: We propose using deep learning to reconstruct high SNR OCT images from low bit-depth acquisition. Approach: The feasibility of our approach is evaluated by applying this approach to the quantized 3- to 8-bit data from native 12-bit interference fringes. We employ a pixel-to-pixel generative adversarial network (pix2pixGAN) architecture in the low-to-high bit-depth OCT image transition. Results: Extensively, qualitative and quantitative results show our method could significantly improve the SNR of the low bit-depth OCT images. The adopted pix2pixGAN is superior to other possible deep learning and compressed sensing solutions. Conclusions: Our work demonstrates that the proper integration of OCT and deep learning could benefit the development of healthcare in low-resource settings. Society of Photo-Optical Instrumentation Engineers 2020-11-15 2020-12 /pmc/articles/PMC7666869/ /pubmed/33191687 http://dx.doi.org/10.1117/1.JBO.25.12.123702 Text en © 2020 The Authors https://creativecommons.org/licenses/by/4.0/ Published by SPIE under a Creative Commons Attribution 4.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.
spellingShingle Special Series on Biomedical Imaging and Sensing
Hao, Qiangjiang
Zhou, Kang
Yang, Jianlong
Hu, Yan
Chai, Zhengjie
Ma, Yuhui
Liu, Gangjun
Zhao, Yitian
Gao, Shenghua
Liu, Jiang
High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning
title High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning
title_full High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning
title_fullStr High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning
title_full_unstemmed High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning
title_short High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning
title_sort high signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning
topic Special Series on Biomedical Imaging and Sensing
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7666869/
https://www.ncbi.nlm.nih.gov/pubmed/33191687
http://dx.doi.org/10.1117/1.JBO.25.12.123702
work_keys_str_mv AT haoqiangjiang highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT zhoukang highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT yangjianlong highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT huyan highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT chaizhengjie highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT mayuhui highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT liugangjun highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT zhaoyitian highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT gaoshenghua highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning
AT liujiang highsignaltonoiseratioreconstructionoflowbitdepthopticalcoherencetomographyusingdeeplearning