Table 1.

Diagnostic accuracy of the proposed CAD system. Feature selection had a significant impact on classifier performance with Friedman test ${\chi }^2=32.7$, 3 d.f., $p=3\times {10}^{-7}$. V: Wilcoxon signed rank statistic of performance compared to complete system; p: Associated Bonferroni-corrected p-value.

	Sensitivity	Specificity	DSC	Accuracy	V	p
Performance of the proposed whole CAD system
LOSO	100% ± 0.00	100% ± 0.00	100% ± 0.00	100% ± 0.00	–	–
Tenfold	100% ± 0.00	99% ± 1.00	99.50% ± 0.50	99.50% ± 0.50
Fourfold	100% ± 0.00	98% ± 2.00	99% ± 1.00	99% ± 1.00
Twofold	100% ± 0.00	97% ± 3.00	98% ± 2.00	98% ± 2.00
Performance of the proposed CAD system when using only the estimated energy at $\nu$1
LOSO	76% ±4.29	96% ± 1.97	75% ± 4.27	86% ± 2.37	118	0.0033
Tenfold	74% ± 1.26	98% ± 4.21	83.61% ± 9.15	86% ± 7.37
Fourfold	71% ± 8.28	98% ± 2.31	81.87% ± 5.05	84.50% ± 3.41
Twofold	71% ±4.24	99% ± 1.41	80.87% ± 2.14	80.30% ± 1.41
Performance of the proposed CAD system when using only the estimated energy at $\nu$2
LOSO	81% ± 3.94	94% ± 2.39	79.33% ± 3.93	87.5% ± 2.28	136	0.0014
Tenfold	79% ± 1.29	94% ± 5.07	85.45% ± 6.23	87% ± 4.83
Fourfold	80% ± 8.86	93% ± 6.31	83.83% ± 6.15	86.50% ± 3.41
Twofold	77% ± 1.41	91% ± 2.82	82.16% ± 2.20	85.50% ± 2.12
Performance of the proposed CAD system when using only the estimated energy at $\nu$3
LOSO	97% ±1.71	94% ± 2.39	95% ± 1.85	95.5% ± 1.43	136	0.0014
Tenfold	93% ± 8.23	97% ± 4.83	94.78% ± 5.56	95% ± 5.27
Fourfold	92% ± 1.13	95% ± 3.83	93.16% ± 7.82	93.50% ± 7.19
Twofold	91% ±7.07	95% ± 1.41	92.79% ± 3.20	93% ± 2.82