Table 3.

Differences in performance between the view selection protocols for each classification protocol.

	Classification Protocol
View Protocol Comparison	A	B	C
All images vs Random images	+0.01 [0.00; 1]†	+0.02 [0.00; 0.04]*	+0.00 [-0.02; 1]†
All images vs First image only	+0.00 [0.00; 1]†	+0.01 [0.00; 1]†	-0.04 [-0.08; -0.01]*
First image only vs Random images	+0.01 [0.00; 1]†	+0.00 [0.00; 1]†	+0.04 [0.00; 0.09]*

Note – Data are indicated as ΔAUC [confidence interval].

Protocol A: Feature averaging

Protocol B: Classifier output averaging

Protocol C: No averaging

^*Two-sided 95% confidence interval

^†Non-inferiority interval