Table 4 Case-level performance metrics of the metamodel with full post-processing. The performance metrics for scans with hemorrhage are also stratified according to delay from symptom onset to imaging. TP = true positive, tn = true negative, fp = false positive, fn = false negative, fpr = false positive rate, npv = negative predictive value, ppv = positive predictive value, n/a = not applicable, ci = confidence interval.

From: High sensitivity in spontaneous intracranial hemorrhage detection from emergency head CT scans using ensemble-learning approach

 

TP

TN

FP

FN

Sensitivity % [95% CI]

Specificity % [95% CI]

FPR % [95% CI]

NPV % [95% CI]

PPV % [95% CI]

Accuracy % [95% CI]

All Scans

106

6873

806

12

89.8% [84.4–95.3%]

89.5% [88.8–90.2%]

10.5% [9.8–11.2%]

99.8% [99.7–99.9%]

11.6% [9.5–13.7%]

89.5% [88.8–90.2%]

Symptom onset time confirmed < 12 h

59

  

0

100% [100–100%]

N/A

N/A

N/A

N/A

100% [100–100%]

Symptom onset time presumably < 12 h

19

  

0

100% [100–100%]

N/A

N/A

N/A

N/A

100% [100–100%]

Symptom onset time presumably 12–24 h or Confirmed 12–24 h

13

  

4

76.5% [56.3–96.6%]

N/A

N/A

N/A

N/A

76.5% [56.3–96.6%]

Symptom onset time 24 h − 7 days

10

  

4

71.4% [47.8–95.1%]

N/A

N/A

N/A

N/A

71.4% [47.8–95.1%]

Symptom onset time > 7 days or unclear

5

  

4

55.6% [23.1–88.0%]

N/A

N/A

N/A

N/A

55.6% [23.1–88.0%]

Negative scans

 

6873

806

 

N/A

89.5% [88.8–90.2%]

10.5% [9.8–11.2%]

N/A

N/A

89.5% [88.8–90.2%]