Table 2.
BNB-based models of different combinations of steps.
| Model | Performance | Steps includeda | BNBb classification | ||||||||||
|
|
SENSc (%) | SPECd (%) | PPVe (%) | NPVf (%) | ACCg (%) | Youden index | 1 | 2 | 4.1 | 4.2 |
|
||
| Model A | 54.7 | 82.1 | 56.8 | 80.9 | 73.9 | 0.368 | ✓ |
|
✓ | ✓ |
|
||
| Model B | 53.0 | 86.7 | 67.0 | 81.6 | 76.6 | 0.397 | ✓ | ✓ |
|
|
✓ | ||
| Model C | 54.0 | 87.3 | 67.8 | 82.1 | 77.3 | 0.413 | ✓ | ✓ | ✓ |
|
✓ | ||
| PAMTh model | 68.0 | 78.0 | 60.6 | 85.8 | 75.0 | 0.460 | ✓ | ✓ | ✓ | ✓ | ✓ | ||
aStep 1, text preprocessing; step 2, term frequency–inverse document frequency feature extraction and selection; step 4.1, manual feature addition; step 4.2, rule-based judgment.
bBNB: Bernoulli naïve Bayes.
cSENS: sensitivity.
dSPEC: specificity.
ePPV: positive predictive value.
fNPV: negative predictive value.
gACC: accuracy.
hPAMT: prehospital-activated major trauma.