View in EDS

Bayesian neural networks enabled snapshot polarized light scattering spectroscopy with uncertainty quantification.

Saved in:
Bibliographic Details
Title: Bayesian neural networks enabled snapshot polarized light scattering spectroscopy with uncertainty quantification.
Authors: Tuniyazi, Abudusalamu1 (AUTHOR), Mu, Tingkui1 (AUTHOR) tkmu@mail.xjtu.edu.cn, Li, Qiuxia1 (AUTHOR), Li, Haoyang1 (AUTHOR), Jiang, Xiaosa2 (AUTHOR), Han, Feng1 (AUTHOR), Gong, Hang1 (AUTHOR), Wang, Wenjing1 (AUTHOR), Lang, Xuechan1 (AUTHOR), Wang, Bin1 (AUTHOR), Li, Zhiyuan1 (AUTHOR), Wu, Jie3 (AUTHOR), Qin, Bin1,2 (AUTHOR) qinbin16@163.com
Source: Optics & Lasers in Engineering. Jan2023, Vol. 160, pN.PAG-N.PAG. 1p.
Subjects: Light scattering, Bayesian analysis, Early detection of cancer, Convolutional neural networks, Spectrometry, Adenomatous polyps
Abstract: • Snapshot polarized light scattering spectroscopy (PLSS) achieves real time inversion of nuclear size with Bayesian deep learning. • The lack of labeled data for network training is solved by using synthetic spectra. • The nuclear size inversion task considered as classification problem. • The Bayesian CNN classification network outperforms the Bayesian FNN regression network in terms of uncertainty quantification. Polarized light-scattering spectroscopy (PLSS) is a promising noninvasive early cancer detection technique via inversing the nuclear size of epithelial cells from collected single scattered light. Conventionally, the nuclear size is inferred by using model-driven least square fitting (LSF) methods, which is time-consuming and challenging for real-time cancer diagnosis. Although data-driven deep-learning-based techniques can be employed to speed up the inverse process, the labeled data are usually limited and the reliability quantification of prediction is lacking. Herein, by synthesizing a large quantity of single scattered spectra with physical priori, a Bayesian-deep-learning (BDL)-based PLSS framework is presented for the first time as far as we know, and is expected to provide model uncertainty of network prediction for the reliability quantification. Further, to explore the applicability of the BDL uncertain estimation, the neural networks for early cancer diagnosis are designed as a convolutional neural network (CNN) classification problem and a fully connected neural network (FNN) regression problem, respectively. They are verified using the single scattered spectra of polystyrene microsphere tissue models and human colorectal lesion tissue obtained by our proposed snapshot PLSS system. The results show that both BDL networks obtain similar nuclear sizes to the ground truth and their inferring speeds are far faster than LSF methods. Further, it is interesting to note that only the CNN classification network can provide effective uncertainty for reliability quantification. The work presents a new paradigm for the real-time diagnosis of early cancer and provides a reference for the design of snapshot PLSS endoscopy. [ABSTRACT FROM AUTHOR]
Copyright of Optics & Lasers in Engineering is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 159979419
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Bayesian neural networks enabled snapshot polarized light scattering spectroscopy with uncertainty quantification.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Tuniyazi%2C+Abudusalamu%22">Tuniyazi, Abudusalamu</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Mu%2C+Tingkui%22">Mu, Tingkui</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> tkmu@mail.xjtu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Li%2C+Qiuxia%22">Li, Qiuxia</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Haoyang%22">Li, Haoyang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Jiang%2C+Xiaosa%22">Jiang, Xiaosa</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Han%2C+Feng%22">Han, Feng</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Gong%2C+Hang%22">Gong, Hang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Wenjing%22">Wang, Wenjing</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Lang%2C+Xuechan%22">Lang, Xuechan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Bin%22">Wang, Bin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Zhiyuan%22">Li, Zhiyuan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wu%2C+Jie%22">Wu, Jie</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Qin%2C+Bin%22">Qin, Bin</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> qinbin16@163.com</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Optics+%26+Lasers+in+Engineering%22">Optics & Lasers in Engineering</searchLink>. Jan2023, Vol. 160, pN.PAG-N.PAG. 1p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Light+scattering%22">Light scattering</searchLink><br /><searchLink fieldCode="DE" term="%22Bayesian+analysis%22">Bayesian analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Early+detection+of+cancer%22">Early detection of cancer</searchLink><br /><searchLink fieldCode="DE" term="%22Convolutional+neural+networks%22">Convolutional neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Spectrometry%22">Spectrometry</searchLink><br /><searchLink fieldCode="DE" term="%22Adenomatous+polyps%22">Adenomatous polyps</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: • Snapshot polarized light scattering spectroscopy (PLSS) achieves real time inversion of nuclear size with Bayesian deep learning. • The lack of labeled data for network training is solved by using synthetic spectra. • The nuclear size inversion task considered as classification problem. • The Bayesian CNN classification network outperforms the Bayesian FNN regression network in terms of uncertainty quantification. Polarized light-scattering spectroscopy (PLSS) is a promising noninvasive early cancer detection technique via inversing the nuclear size of epithelial cells from collected single scattered light. Conventionally, the nuclear size is inferred by using model-driven least square fitting (LSF) methods, which is time-consuming and challenging for real-time cancer diagnosis. Although data-driven deep-learning-based techniques can be employed to speed up the inverse process, the labeled data are usually limited and the reliability quantification of prediction is lacking. Herein, by synthesizing a large quantity of single scattered spectra with physical priori, a Bayesian-deep-learning (BDL)-based PLSS framework is presented for the first time as far as we know, and is expected to provide model uncertainty of network prediction for the reliability quantification. Further, to explore the applicability of the BDL uncertain estimation, the neural networks for early cancer diagnosis are designed as a convolutional neural network (CNN) classification problem and a fully connected neural network (FNN) regression problem, respectively. They are verified using the single scattered spectra of polystyrene microsphere tissue models and human colorectal lesion tissue obtained by our proposed snapshot PLSS system. The results show that both BDL networks obtain similar nuclear sizes to the ground truth and their inferring speeds are far faster than LSF methods. Further, it is interesting to note that only the CNN classification network can provide effective uncertainty for reliability quantification. The work presents a new paradigm for the real-time diagnosis of early cancer and provides a reference for the design of snapshot PLSS endoscopy. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Optics & Lasers in Engineering is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=159979419
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.optlaseng.2022.107298
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 1
        StartPage: N.PAG
    Subjects:
      – SubjectFull: Light scattering
        Type: general
      – SubjectFull: Bayesian analysis
        Type: general
      – SubjectFull: Early detection of cancer
        Type: general
      – SubjectFull: Convolutional neural networks
        Type: general
      – SubjectFull: Spectrometry
        Type: general
      – SubjectFull: Adenomatous polyps
        Type: general
    Titles:
      – TitleFull: Bayesian neural networks enabled snapshot polarized light scattering spectroscopy with uncertainty quantification.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Tuniyazi, Abudusalamu
      – PersonEntity:
          Name:
            NameFull: Mu, Tingkui
      – PersonEntity:
          Name:
            NameFull: Li, Qiuxia
      – PersonEntity:
          Name:
            NameFull: Li, Haoyang
      – PersonEntity:
          Name:
            NameFull: Jiang, Xiaosa
      – PersonEntity:
          Name:
            NameFull: Han, Feng
      – PersonEntity:
          Name:
            NameFull: Gong, Hang
      – PersonEntity:
          Name:
            NameFull: Wang, Wenjing
      – PersonEntity:
          Name:
            NameFull: Lang, Xuechan
      – PersonEntity:
          Name:
            NameFull: Wang, Bin
      – PersonEntity:
          Name:
            NameFull: Li, Zhiyuan
      – PersonEntity:
          Name:
            NameFull: Wu, Jie
      – PersonEntity:
          Name:
            NameFull: Qin, Bin
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Text: Jan2023
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-print
              Value: 01438166
          Numbering:
            – Type: volume
              Value: 160
          Titles:
            – TitleFull: Optics & Lasers in Engineering
              Type: main
ResultId 1