Acta Optica Sinica, Volume. 41, Issue 11, 1117001(2021)

Parameters Inversion Algorithm of Biological Tissues Based on a Neural Network Model

Ge Xu, Liquan Dong*, Lingqin Kong, Yuejin Zhao, Ming Liu, Mei Hui, Xiaohua Liu, Falong Wang, and Jing Yuan
Author Affiliations
  • Beijing Key Laboratory for Precision Optoelectronic Measurement Instrument and Technology, School of Optics and Photonics, Beijing Institute of Technology, Beijing 100081, China
  • show less
    Figures & Tables(6)
    Topological structure of BP neural network model
    Neural network inversion algorithm
    Inversion of absorption and scattering coefficients in 2-point detection distance (r=0.1 cm and r=0.3 cm). (a) Inversion of absorption coefficient; (b) inversion of scattering coefficient
    Comparison of absorption coefficient obtained by neural network inversion algorithm and MCML
    Comparison of scattering coefficient obtained by neural network inversion algorithm and MCML
    • Table 1. MAE and R2 of absorption and scattering coefficients in multi-point detection distances retrieved with neural network algorithm

      View table

      Table 1. MAE and R2 of absorption and scattering coefficients in multi-point detection distances retrieved with neural network algorithm

      Detectiondistance pointMAER2
      μa /cm-1μs /cm-1μa /cm-1μs /cm-1
      2-point0.0031.5740.99970.9915
      4-point0.0022.1050.99990.9833
      6-point0.0011.5520.99990.9844
      8-point0.0011.8480.99990.9847
      10-point0.0021.4260.99980.9869
      12-point0.0033.0080.99950.9498
      14-point0.0053.3890.99900.9142
    Tools

    Get Citation

    Copy Citation Text

    Ge Xu, Liquan Dong, Lingqin Kong, Yuejin Zhao, Ming Liu, Mei Hui, Xiaohua Liu, Falong Wang, Jing Yuan. Parameters Inversion Algorithm of Biological Tissues Based on a Neural Network Model[J]. Acta Optica Sinica, 2021, 41(11): 1117001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Medical Optics and Biotechnology

    Received: Sep. 27, 2020

    Accepted: Jan. 8, 2021

    Published Online: Jun. 7, 2021

    The Author Email: Dong Liquan (kylind@bit.edu.cn)

    DOI:10.3788/AOS202141.1117001

    Topics