Page 52 - 南京医科大学自然版
P. 52
第45卷第2期
·192 · 南 京 医 科 大 学 学 报 2025年2月
0 10 20 30 40 50 60 70 80 90 100
Points
Intra & Peri⁃radscore
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
1
Lesion enhancement pattern
0
1
Peritumoral edema
0
Total points
0 10 20 30 40 50 60 70 80 90 100 110 120 130
N2⁃3 stage ALN rate
0.1 0.5 0.9
Lesion enhancement pattern:0=mass,1=non⁃mass enhancement;Peritumoral edema:0=negative,1=positive.
图3 联合模型列线图
Figure 3 Nomogram of the combined model
A Training datase B Validation dataset
1.0 Clinical AUC:0.770 1.0 Clinical AUC:0.688
(95%CI:0.707-0.833) (95%CI:0.588-0.788)
Intra & Peri⁃model:0.911
Intra & Peri⁃model:0.858
0.8 (95%CI:0.872-0.951) 0.8 (95%CI:0.770-0.947)
Sensitivity 0.6 (95%CI:0.885-0.962) Sensitivity 0.6 (95%CI:0.828-0.955)
Nocnogram AUC:0.923
Nocnogram AUC:0.892
0.4
0.4
0.2 0.2
0 0
00 0.2 0.4 0.6 0.8 1.0 00 0.2 0.4 0.6 0.8 1.0
1-Specifieity 1-Specifieity
C D
1.0 1.0
0.8 0.8
Fraction of positives 0.6 Fraction of positives 0.6
0.4
0.4
Clinical
0.2 Perfectiy calibrated 0.2 Perfectiy calibrated
Clinical
Intra & Peri⁃ Intra & Peri⁃
0 Nomogram 0 Nomogram
00 0.2 0.4 0.6 0.8 1.0 00 0.2 0.4 0.6 0.8 1.0
Mean predicted probability Mean predicted probability
E F
0.5 Clinical 0.5 Clinical
Intra & Peri⁃model Intra & Peri⁃model
0.4 Nomogram 0.4 Nomogram
Treal all 0.3 Treal all
0.3
Treat none
Net Benefit 0.2 Net Benefit 0.2
Treat none
0.1
0.1
0 0
-0.1 -0.1
00 0.2 0.4 0.6 0.8 1.0 00 0.2 0.4 0.6 0.8 1.0
Threshold probability Threshold probability
A,B:ROC curves of training(A)and validation(B)sets;C,D:Calibration curves of training(C)and validation(D)sets;E,F:Decision curves of
training(E)and validation(F)sets. The X⁃axis indicates the threshold probability. The Y⁃axis represents the net benefit.
图4 模型的评价和比较
Figure 4 Evaluation and comparison of the developed models

