• Acta Photonica Sinica
  • Vol. 52, Issue 2, 0210003 (2023)
Junzhao GAO1,2, Dangfei HUANG1,2,*, Lechao ZHANG1,2, Dong SONG3..., Jinghui HONG3, Lili ZHANG1,2, Hongyu TANG1,2 and Yao ZHOU1,2|Show fewer author(s)
Author Affiliations
  • 1College of Optoelectronic Engineering,Changchun University of Technology,Changchun 130022,China
  • 2Zhongshan Research Institute of Changchun University of Technology,Zhongshan 528437,China
  • 3The First Hospital of Jilin University,Changchun 130021,China
  • show less
    DOI: 10.3788/gzxb20235202.0210003 Cite this Article
    Junzhao GAO, Dangfei HUANG, Lechao ZHANG, Dong SONG, Jinghui HONG, Lili ZHANG, Hongyu TANG, Yao ZHOU. Cancer Tissue Recognition by Muller Matrix Imaging Full Array Curve[J]. Acta Photonica Sinica, 2023, 52(2): 0210003 Copy Citation Text show less
    Experimental device and schematic diagram for measuring backward Mueller matrix
    Fig. 1. Experimental device and schematic diagram for measuring backward Mueller matrix
    Schematic diagram of Mueller matrix element information
    Fig. 2. Schematic diagram of Mueller matrix element information
    Unstained pathological section sample
    Fig. 3. Unstained pathological section sample
    Mueller matrix element information link diagram
    Fig. 4. Mueller matrix element information link diagram
    Lung cancer slice No. 22 normalized Mueller matrix
    Fig. 5. Lung cancer slice No. 22 normalized Mueller matrix
    Mueller matrix data cube
    Fig. 6. Mueller matrix data cube
    Selection and characteristic curve of 22 ROI of lung cancer
    Fig. 7. Selection and characteristic curve of 22 ROI of lung cancer
    Radar figure of lung cancer characteristic curve 22
    Fig. 8. Radar figure of lung cancer characteristic curve 22
    Comparison of internal characteristics of characteristic curves of cancerous and normal random samples from lung cancer sections
    Fig. 9. Comparison of internal characteristics of characteristic curves of cancerous and normal random samples from lung cancer sections
    Classification process of lung cancer slice characteristic curve sampling
    Fig. 10. Classification process of lung cancer slice characteristic curve sampling
    Accuracy discrimination of lung cancer slice measured map and template map
    Fig. 11. Accuracy discrimination of lung cancer slice measured map and template map
    Single element SVM classification process for M11 and M21 lung cancer slices
    Fig. 12. Single element SVM classification process for M11 and M21 lung cancer slices
    Mueller matrix characteristic curve classification of breast cancer No.44 high magnification imaging
    Fig. 13. Mueller matrix characteristic curve classification of breast cancer No.44 high magnification imaging
    Mueller matrix characteristic curve classification for low magnification imaging of breast cancer No.44
    Fig. 14. Mueller matrix characteristic curve classification for low magnification imaging of breast cancer No.44
    Actual/forecastReal cancerousReal normalReal backgroundTotal
    Cancerous tissue46 0703 539949 618
    Normal tissue5 12929 80313935 071
    Background14 97910 301811 631836 911
    Total66 17843 643811 779921 600
    Table 1. Confusion matrix of lung cancer slice classification results
    Category(Classification pixel)accuracy/%
    Cancerous tissue(46 070/49 618)92.85%
    Normal tissue(29 803/35 071)84.98%
    Background(811 631/811 779)96.98%
    Overall accuracy(887 504/921 600)96.300 3%
    Kappa0.809 7
    Table 2. Overall classification accuracy statistics of lung cancer sections
    Category(Classification pixel)accuracy/%
    Cancerous tissue(46 070/49 618)92.85%
    Normal tissue(29 803/35 071)84.98%
    Overall accuracy(75 873/84 689)89.590 1%
    Table 3. Accuracy statistics of normal and cancerous tissues in lung cancer sections
    Junzhao GAO, Dangfei HUANG, Lechao ZHANG, Dong SONG, Jinghui HONG, Lili ZHANG, Hongyu TANG, Yao ZHOU. Cancer Tissue Recognition by Muller Matrix Imaging Full Array Curve[J]. Acta Photonica Sinica, 2023, 52(2): 0210003
    Download Citation