Skip to main content

Table 1 The proportions of patient classes in the test set

From: Prediction of inappropriate pre-hospital transfer of patients with suspected cardiovascular emergency diseases using machine learning: a retrospective observational study

Prediction model

Transferred to incapable hospital (N = 1770)

Transferred to

capable hospital

(N = 26,507)

Total

(N = 28,277)

P-value

Predicted cardiopulmonary resuscitation (Class 1)

364 (20.56)

161 (0.61)

525 (1.86)

< 0.001

Predicted intubation (Class 2)

431 (24.35)

1089 (4.11)

1520 (5.36)

< 0.001

Predicted central catheterization

(Class 3)

179 (10.11)

846 (3.19)

1025 (3.62)

< 0.001

Predicted massive transfusion

(Class 4)

14 (0.79)

160 (0.60)

174 (0.62)

0.413

Predicted emergency percutaneous coronary intervention (Class 5)

158 (8.93)

2356 (8.89)

2514 (8.89)

0.991

Predicted intensive care unit admission after ED process (Class 6)

133 (7.51)

4304 (16.24)

4437 (15.69)

< 0.001

Predicted emergency operation (Class 7)

123 (6.95)

191 (0.72)

314 (1.11)

< 0.001

Predicted performed magnetic resonance imaging in the ED (Class 8)

44 (2.49)

607 (2.29)

651 (2.30)

0.653

Predicted performed echocardiography in the ED (Class 9)

21 (1.19)

830 (3.13)

851 (3.01)

< 0.001

Predicted performed computed tomography angiography in the ED (Class 10)

104 (5.88)

1432 (5.40)

1536 (5.43)

0.426

Predicted psychiatric management in the ED (Class 11)

5 (0.28)

143 (0.54)

148 (0.52)

0.200

Predicted admission after ED process (Class 12)

724 (40.90)

5050 (19.05)

5774 (20.42)

< 0.001

Predicted discharge after ED process (Class 13)

364 (20.56)

161 (0.61)

11,664 (41.25)

< 0.001

  1. Variables are expressed as counts (%); ED, emergency department