Skip to main content

Table 3 Detailed results on the testing cohort

From: A process mining- deep learning approach to predict survival in a cohort of hospitalized COVID‐19 patients

Hour

Confusion matrix

AUROC

Specificity

Sensitivity

Accuracy

PM

RF

LR

PM

RF

LR

PM

RF

LR

PM

RF

LR

PM

RF

LR

6

84;8

4;7

54;38

5;6

77;15

8;3

0.776

0.628

0.611

0.913

0.587

0.837

0.636

0.545

0.273

0.883

0.583

0.776

12

81;10

5;6

58;33

5;6

75;16

8;3

0.782

0.635

0.608

0.890

0.637

0.824

0.545

0.545

0.273

0.853

0.627

0.765

18

80;10

4;7

57;33

5;6

76;14

7;4

0.806

0.658

0.640

0.889

0.633

0.844

0.636

0.545

0.364

0.861

0.624

0.792

24

67;17

4;7

51;33

5;6

70;14

6;5

0.799

0.640

0.644

0.798

0.607

0.833

0.636

0.545

0.455

0.779

0.600

0.789

30

71;11

3;8

50;32

5;6

68;14

6;5

0.814

0.656

0.646

0.866

0.610

0.829

0.727

0.545

0.455

0.849

0.602

0.785

36

56;25

3;8

51;30

5;6

66;15

6;5

0.814

0.654

0.641

0.691

0.630

0.815

0.727

0.545

0.455

0.696

0.619

0.771

42

68;10

4;7

48;30

5;6

62;16

6;5

0.817

0.657

0.631

0.872

0.615

0.795

0.636

0.545

0.455

0.843

0.606

0.752

48

52;18

4;7

44;26

5;6

55;15

6;5

0.806

0.680

0.657

0.743

0.629

0.786

0.636

0.545

0.455

0.728

0.617

0.740

54

55;11

4;7

44;22

5;6

52;14

6;5

0.853

0.692

0.659

0.833

0.667

0.788

0.636

0.545

0.455

0.805

0.649

0.740

60

62;2

5;6

44;20

5;6

51;13

6;5

0.843

0.713

0.662

0.969

0.688

0.797

0.545

0.545

0.455

0.907

0.667

0.746

66

52;9

4;7

42;19

5;6

47;14

6;5

0.875

0.718

0.641

0.852

0.689

0.770

0.636

0.545

0.455

0.819

0.667

0.722

72

44;11

3;8

39;16

5;6

43;12

6;5

0.9

0.709

0.625

0.800

0.709

0.782

0.727

0.545

0.455

0.788

0.681

0.727