Skip to main content
. Author manuscript; available in PMC: 2018 May 1.
Published in final edited form as: Med Phys. 2017 Apr 13;44(5):1846–1856. doi: 10.1002/mp.12214

Table 6.

Diagnostic performance comparison in AUC between the classifier and the radiologists over sensitivity 90 or higher.

Performance Comparison (AUC) over sensitivity 90 or higher Difference in AUC
CADx Radiologists AUCL – AUCR [95% CI] p-value
AUCL [95% CI] Reconstructions AUCR [95% CI]
0.085[0.063, 0.094] IIR1 0.003 [0, 0.015] 0.061 [0.04, 0.086] <0.0001*
IIR2 0.006 [0, 0.026] 0.069 [0.041, 0.089] <0.0001*
IIR3 0.009 [0.003, 0.024] 0.085 [0.063, 0.094] <0.0001*
FDK 0.004 [0, 0.031] 0.034 [0.013, 0.063] <0.0001*
*

Statistically significant (p-value < 0.0125).