Page 134 - 南京医科大学自然版

第46卷第7期
·1080 · 南 京 医 科 大 学 学 报 2026年7月

A B
1.00 1.00
Training AUC=0.811 Training AUC=0.737
Validation AUC=0.817 Validation AUC=0.720
0.75 0.75
Sensitivity 0.50 Sensitivity 0.50

0.25 0.25
Training Training
Validation Validation
0 0
0.0 0.25 0.50 0.75 1.00 0 0.25 0.50 0.75 1.00
1-Specificity 1-Specificity

C
1.00
Training AUC=0.755
Validation AUC=0.788
0.75
Sensitivity 0.50

0.25
Training
Validation
0
0 0.25 0.50 0.75 1.00
1-Specificity
图3 训练集和内部验证集在12（A）、24（B）、36（C）个月时生存预测ROC曲线
Figure 3 ROC curves of survival prediction for the training set and internal validation set at 12（A），24（B），and 36 months（C）

A B C
1.0 1.0 1.0
Survival probability 0.6 Survival probability 0.6 Survival probability 0.6
0.8
0.8
0.8
0.4
0.4
0.4
0.2
0 0.2 0 0.2 0
00 0.2 0.4 0.6 0.8 1.0 00 0.2 0.4 0.6 0.8 1.0 00 0.2 0.4 0.6 0.8 1.0
Nomogram⁃predicted probability Nomogram⁃predicted probability Nomogram⁃predicted probability
D E F
1.0 1.0 1.0
Survival probability 0.6 Survival probability 0.6 Survival probability 0.6
0.8
0.8
0.8
0.4
0.4
0.4
0.2
0 0.2 0 0.2 0
00 0.2 0.4 0.6 0.8 1.0 00 0.2 0.4 0.6 0.8 1.0 00 0.2 0.4 0.6 0.8 1.0
Nomogram⁃predicted probability Nomogram⁃predicted probability Nomogram⁃predicted probability
A：Calibration curve of the OS prediction model at 12 months for the training set. B：Calibration curve of the OS prediction model at 12 months for
the validation set. C：Calibration curve of the OS prediction model at 24 months for the training set. D：Calibration curve of the OS prediction model at
24 months for the validation set. E：Calibration curve of the OS prediction model at 36 months for the training set. F：Calibration curve of the OS predic⁃
tion model at 36 months for the validation set.
图4 训练集和内部验证集在12、24、36个月时OS预测模型的校准曲线
Figure 4 Calibration curves of OS prediction model for training set and internal validation set at 12，24，and 36 months

关 。此外，TyG 指数还被用于预测代谢综合征的发 TyG 指数引入 NSCLC 的联合免疫治疗和化疗中，发
生，显示出优于传统指标的预测能力 。本研究将 现较低的 TyG 指数与接受此类联合治疗的 NSCLC