Table 9.

Performance results obtained with 2 classes (crackle vs. other)—training: fixed duration; testing: fixed duration.

Classifiers	Accuracy	AUCCrack	F1Crack	MCCCrack	F1Other	MCCOther
LDA_10MRMR	88.9 ± 0.6	93.9 ± 0.7	92.0 ± 0.3	85.1 ± 1.2	81.9 ± 1.9	78.3 ± 2.7
LDA_100MRMR	88.9 ± 0.0	92.8 ± 0.6	91.8 ± 0.0	85.5 ± 0.0	82.8 ± 0.0	79.9 ± 0.0
LDA_Full	88.1 ± 0.2	93.0 ± 0.2	91.5 ± 0.2	83.6 ± 0.3	79.9 ± 0.4	75.3 ± 0.4
SVMrbf_10MRMR	91.2 ± 0.2	95.2 ± 0.9	94.0 ± 0.1	87.7 ± 0.3	83.5 ± 0.5	79.7 ± 0.6
SVMrbf_100MRMR	93.3 ± 0.2	97.6 ± 0.4	95.4 ± 0.1	90.6 ± 0.2	87.4 ± 0.3	84.5 ± 0.4
SVMrbf_Full	93.5 ± 0.6	97.7 ± 0.3	95.6 ± 0.4	91.0 ± 0.9	88.0 ± 1.2	85.3 ± 1.5
RUSBoost_10MRMR	93.1 ± 0.3	97.5 ± 0.1	95.3 ± 0.2	90.4 ± 0.4	87.4 ± 0.7	84.5 ± 0.8
RUSBoost_100MRMR	95.2 ± 0.7	98.8 ± 0.2	96.7 ± 0.5	93.2 ± 1.0	91.0 ± 1.2	88.9 ± 1.4
RUSBoost_Full	94.7 ± 0.9	98.8 ± 0.4	96.3 ± 0.7	92.6 ± 1.2	90.4 ± 1.4	88.3 ± 1.6
CNN_dualInput	99.6 ± 0.1	99.6 ± 0.2	99.8 ± 0.1	99.1 ± 0.3	99.3 ± 0.2	99.1 ± 0.3
CNN_Spectrogram	98.5 ± 0.5	97.8 ± 1.1	99.0 ± 0.3	96.2 ± 1.2	97.2 ± 0.9	96.2 ± 1.2
CNN_melSpectrogram	99.4 ± 0.2	99.2 ± 0.5	99.6 ± 0.2	98.5 ± 0.6	98.9 ± 0.5	98.5 ± 0.6