Skip to main content

Table 3 Effectiveness of our R-Net and comparison between different cellular localization algorithms and pathologists on big patches

From: Automated tumor proportion scoring for PD-L1 expression based on multistage ensemble strategy in non-small cell lung cancer

  R-Net MAE RMSE PCCs SRCC ICC
Mi 12.34 17.82 0.834 0.921 0.917
9.64 15.71 0.882 0.927 0.919
U-Net 10.22 15.45 0.865 0.936 0.935
8.83 14.48 0.904 0.937 0.935
S3-Net 8.18 13.46 0.905 0.945 0.943
7.56 12.73 0.928 0.956 0.953
C-Net 8.00 12.19 0.906 0.951 0.951
7.55 11.82 0.933 0.965 0.963
  1. ICC intraclass correlation coefficient, MAE mean absolute error, MAPE mean absolute percent error, PCC Pearson product-moment correlation coefficient, RMSE root mean squared error, SRCC Spearman’s rank correlation coefficient