Skip to main content

Table 3 Response prediction performance comparison between baseline, delta and longitudinal models in the independent test set for endpoint PFS9 by evaluating AUC, ACC, SENS, SPEC, PREC and bACC, respectively

From: Integration of longitudinal deep-radiomics and clinical data improves the prediction of durable benefits to anti-PD-1/PD-L1 immunotherapy in advanced NSCLC patients

Model

Features

N test

AUC

[95% CI]

ACC

[95% CI]

SENS

[95% CI]

SPES

[95% CI]

PREC

[95% CI]

bACC

[95% CI]

CNN-baseline

Image data

43

0.429

[0.249,0.616]

0.674

[0.535,0.814]

1.000

[1.000,1.000]

0.000

[0.000,0.000]

0.674

[0.535,0.814]

0.500

[0.500, 0.500]

RF-baseline

Clinical data

43

0.563

[0.392,0.735]

0.581

[0.442,0.721]

0.793

[0.636,0.929]

0.143

[0.000,0.357]

0.657

[0.500,0.811]

0.468

[0.352,0.591]

RF-baseline

Radiomics

43

0.286

[0.112,0.494]

0.512

[0.372,0.651]

0.655

[0.480,0.815]

0.214

[0.000,0.455]

0.633

[0.464,0.800]

0.435

[0.303,0.576]

RF-baseline

DF-imm

43

0.541

[0.359,0.724]

0.628

[0.488,0.767]

0.759

[0.600,0.903]

0.357

[0.118,0.600]

0.710

[0.533,0.867]

0.558

[0.405,0.711]

RF-delta

Clinical data

21

0.550

[0.301,0.795]

0.524

[0.333,0.762]

0.636

[0.333,0.900]

0.400

[0.100,0.714]

0.538

[0.250,0.818]

0.518

[0.306,0.750]

RF-delta

Radiomics

36

0.598

[0.353,0.848]

0.639

[0.472,0.778]

0.680

[0.500,0.857]

0.545

[0.231,0.857]

0.773

[0.588,0.947]

0.613

[0.429,0.788]

RF-delta

DF-imm

36

0.525

[0.315,0.743]

0.556

[0.389,0.722]

0.760

[0.571,0.920]

0.091

[0.000,0.300]

0.655

[0.481,0.824]

0.425

[0.315,0.554]

RF-longitudinal

Clinical data

33

0.585

[0.367,0.783]

0.545

[0.364,0.697]

0.600

[0.381,0.812]

0.462

[0.182,0.727]

0.632

[0.412,0.850]

0.531

[0.360,0.698]

RF-longitudinal

Radiomics

40

0.528

[0.341,0.701]

0.558

[0.395,0.698]

0.724

[0.562,0.88]

0.214

[0.000,0.455]

0.656

[0.484,0.818]

0.469

[0.338,0.612]

RF-longitudinal

DF-imm

40

0.702

[0.515,0.867]

0.750

[0.625,0.875]

0.885

[0.750,1.000]

0.500

[0.214,0.769]

0.767

[0.606,0.914]

0.692

[0.540,0.840]

  1. For each metric, the 95% confidence interval is shown and the highest value is highlighted in bold