Table 4 Prediction Performance in Real Life Data for LR

ADNI

Accuracy

0.842 (0.736)

0.831 (0.747)

0.842 (0.747)

0.907

0.852

0.907

0.907

0.902

0.918

0.923

0.907

Sensitivity

0.810

0.796

0.810

0.964

0.810

0.964

0.964

0.993

0.971

0.985

0.964

Specificity

0.935

0.935

0.935

0.739

0.978

0.739

0.739

0.630

0.761

0.739

0.739

AUC

0.798 (0.880)

0.789 (0.884)

0.798 (0.883)

0.851

0.894

0.894

0.894

0.928

0.866

0.862

0.851

Brier Score

0.107

0.109

0.110

0.093

0.113

0.093

0.093

0.098

0.082

0.077

0.093

OASIS

Accuracy

0.733 (0.741)

0.733 (0.688)

0.733 (0.741)

0.827

0.800

0.827

0.840

0.787

0.760

0.760

0.773

Sensitivity

0.818

0.818

0.818

1.000

0.977

1.000

1.000

1.000

0.841

0.841

0.864

Specificity

0.613

0.613

0.613

0.581

0.548

0.581

0.613

0.484

0.645

0.645

0.645

AUC

0.727(0.764)

0.727 (0.701)

0.727(0.764)

0.790

0.763

0.886

0.893

0.867

0.756

0.756

0.772

Brier Score

0.129

0.130

0.129

0.173

0.126

0.173

0.160

0.213

0.240

0.240

0.227

Pima Indian Diabetes

Accuracy

0.727

0.727

0.727

0.727

0.708

0.727

0.734

0.734

0.786

0.792

0.786

Sensitivity

0.705

0.705

0.705

0.867

0.667

0.867

0.876

0.895

0.864

0.893

0.874

Specificity

0.776

0.776

0.776

0.429

0.796

0.429

0.429

0.388

0.627

0.588

0.608

AUC

0.711

0.711

0.711

0.648

0.731

0.682

0.692

0.696

0.746

0.741

0.741

Brier Score

0.197

0.197

0.197

0.273

0.199

0.273

0.266

0.266

0.214

0.208

0.214

Colon

Accuracy

0.846

0.769

0.769

0.923

0.769

0.923

0.923

0.923

0.846

0.846

0.692

Sensitivity

1.000

0.750

0.750

1.000

1.000

1.000

1.000

1.000

0.667

0.667

0.333

Specificity

0.778

0.778

0.778

0.889

0.667

0.889

0.889

0.889

0.900

0.900

0.900

AUC

0.889

0.764

0.764

0.944

0.833

0.944

0.944

0.944

0.783

0.783

0.567

Brier Score

0.121

0.224

0.240

0.077

0.276

0.077

0.077

0.077

0.154

0.154

0.308