Table 4 The discrimination performance of RSF and CHR model for predicting LC and OS in the training and test set. tAUCs of RSF and CHR model were calculated and compared in the two sets. CHR Cox proportional Hazards regression, tAUC time-dependent area under the curve, RSF random survival forests, CI confidence interval. * p < 0.05, the difference reach statistically significance between RSF and CHR models in the validation set.

From: Machine learning prognosis model for locally recurrent rectal cancer patients after radioactive 125I seed implantation

Prediction

RSF model

CHR model

Train (95% CI)

Validation (95% CI)

Train (95% CI)

Validation (95% CI)

LC at 1 year

0.978 (0.958–0.998)

0.840 (0.758–0.928) *

0.877 (0.815–0.940)

0.692 (0.567–0.852)

LC at 2 year

0.999 (0.997–0.999)

0.888 (0.860–0.997) *

0.992 (0.978–0.998)

0.874 (0.769–0.979)

OS at 1 year

0.913 (0.708–0.881)

0.835 (0.801–0.952) *

0.893 (0.736–0.934)

0.800 (0.778–0.921)

OS at 2 year

0.907 (0.728–0.970)

0.761 (0.667–0.918) *

0.863 (0.727–0.918)

0.685 (0.639–0.723)

Back to article page