Table 3.

Predictive performance of artificial intelligence models for overall survival of patients who received definitive treatment for laryngeal squamous cell carcinoma (n = 1020).

Prediction model	Linear slope	Linearity-y-intercept	C-index
Deep neural network with regression	0.731 ± 0.048	9.659 ± 0.964	0.893 ± 0.017
Cox proportional hazard model	0.619 ± 0.058	24.483 ± 2.407	0.747 ± 0.009
Random survival forest	0.079 ± 0.057	53.250 ± 2.349	0.596 ± 0.015
Deep neural network with multi-classification	1.000 ± 0.047	0.126 ± 0.762	0.859 ± 0.018
Deep neural network with only T/N staging	Not available	Not available	0.504 ± 0.007