Skip to main content
Egyptian Journal of Radiology and Nuclear Medicine

Table 5 Discriminating thresholds and diagnostic performance of Insp

From: Assessment of artificial intelligence-aided chest computed tomography in diagnosis of chronic obstructive airway disease: an observational study

Coordinates of ROC curve

Threshold for achieving optimal sensitivity and specificity

Threshold for achieving near-90% sensitivity

Threshold for achieving near-90% specificity

AUC

0.860

Standard error

0.048

95%CI

0.766–0.955

p value

 < 0.0001**

Cut-off

 ≥ 19.1

 ≥ 11.0

 ≥ 26.1

Sensitivity

 TP/(TP + FN)

78.0%

88.9%

66.7%

Specificity

 TN/(TN + FP)

82.1%

64.1%

89.7%

  1. LAA-950 (%) for distinguishing severe from moderate COPD using forced expiratory volume in the first second (FEV1) spirometric parameter as a reference (gold) standard
  2. ROC curve  receiver operating characteristics curve, AUC area under ROC curve, CI  confidence interval, TP true positive, FN false negative, TN true negative, FP false positive
  3. * Significant p value < 0.05
  4. ** Highly significant p value < 0.01
Back to article page