Table 2 Response prediction performance comparison between baseline, delta and longitudinal models in the independent test set for endpoint PFS6 by evaluating AUC, ACC, SENS, SPEC, PREC and bACC, respectively

CNN-baseline

Image data

43

0.518

[0.329,0.696]

0.535

[0.372,0.674]

0.750

[0.565,0.909]

0.263

[0.067,0.478]

0.562

[0.387,0.737]

0.507

[0.377,0.643]

RF-baseline

Clinical data

43

0.667

[0.485,0.833]

0.651

[0.512,0.791]

0.833

[0.667,0.962]

0.421

[0.200,0.650]

0.645

[0.480,0.812]

0.627

[0.488,0.774]

RF-baseline

Radiomics

43

0.448

[0.291,0.607]

0.442

[0.302,0.605]

0.333

[0.150,0.526]

0.579

[0.350,0.800]

0.500

[0.250,0.750]

0.456

[0.306,0.601]

RF-baseline

DF-imm

43

0.588

[0.409,0.767]

0.558

[0.419,0.698]

0.833

[0.679,0.960]

0.211

[0.050,0.417]

0.571

[0.406,0.735]

0.522

[0.403,0.638]

RF-delta

Clinical data

21

0.435

[0.173,0.714]

0.333

[0.143,0.571]

0.167

[0.000,0.417]

0.556

[0.200,0.875]

0.333

[0.000,0.750]

0.361

[0.163,0.559]

RF-delta

Radiomics

36

0.489

[0.276,0.706]

0.528

[0.361,0.694]

0.524

[0.304,0.737]

0.533

[0.273,0.786]

0.611

[0.389,0.833]

0.529

[0.357,0.688]

RF-delta

DF-imm

36

0.660

[0.451,0.846]

0.611

[0.444,0.778]

0.714

[0.500,0.900]

0.467

[0.200,0.733]

0.652

[0.455,0.842]

0.590

[0.433,0.750]

RF-longitudinal

Clinical data

33

0.700

[0.508,0.877]

0.576

[0.394,0.727]

0.467

[0.200,0.733]

0.667

[0.438,0.875]

0.530

[0.250,0.800]

0.567

[0.405,0.733]

RF-longitudinal

Radiomics

40

0.581

[0.407,0.749]

0.628

[0.488,0.767]

0.667

[0.464,0.850]

0.579

[0.348,0.800]

0.667

[0.474,0.852]

0.623

[0.466,0.763]

RF-longitudinal

DF-imm

40

0.740

[0.563,0.883]

0.700

[0.550,0.825]

0.818

[0.647,0.958]

0.556

[0.312,0.783]

0.692

[0.500,0.864]

0.687

[0.550,0.827]