Skip to main content

Table 2 Model accuracy evaluation based on bootstrap resampling method

From: Improve individual treatment by comparing treatment benefits: cancer artificial intelligence survival analysis system for cervical carcinoma

Model Dataset Number C-index C-index C-index Brier-score Brier-score Brier-score
12-month 36-month 60-month 12-month 36-month 60-month
Cox Dataset1 14,946 0.822 0.822 0.822 0.077 0.124 0.133
Dataset2 14,946 0.818 0.818 0.818 0.078 0.124 0.137
Dataset3 14,946 0.819 0.819 0.819 0.075 0.123 0.135
Dataset4 14,946 0.819 0.819 0.819 0.076 0.125 0.138
Dataset5 14,946 0.823 0.823 0.823 0.075 0.122 0.130
AFT Dataset1 14,946 0.824 0.824 0.824 0.077 0.123 0.131
Dataset2 14,946 0.819 0.819 0.819 0.078 0.124 0.136
Dataset3 14,946 0.821 0.821 0.821 0.076 0.123 0.133
Dataset4 14,946 0.821 0.821 0.821 0.077 0.125 0.136
Dataset5 14,946 0.825 0.825 0.825 0.075 0.122 0.128
MTLR Dataset1 14,946 0.827 0.830 0.830 0.077 0.133 0.131
Dataset2 14,946 0.822 0.824 0.825 0.078 0.124 0.137
Dataset3 14,946 0.824 0.828 0.829 0.076 0.122 0.134
Dataset4 14,946 0.824 0.826 0.827 0.077 0.125 0.137
Dataset5 14,946 0.828 0.830 0.830 0.075 0.121 0.130
  1. MTLR multi-task logistic regression, AFT accelerated failure time