Table 2.

Average sensitivity of cross-validation results, feature vector size per columns. Rows correspond to scalar feature mappings.

	Measure	Feat.Map.	HG	500	1000	5000	10000
SZAH vs. SZNAH	FC	OS-LAAM	L	100	100	100	100
			R	100	100	100	100
		BF-LAAM	L	100	100	100	100
			R	100	100	100	100
	LA	ReHo	-	100	100	100	100
		ALFF	-	96.7	96.7	100	100
		fALFF	-	100	100	100	100
SZAH vs. HC	FC	OS-LAAM	L	28.3	21.7	11.7	8.3
			R	38.3	30	18.3	10
		BF-LAAM	L	96.7	96.7	93.3	85
			R	96.7	93.3	91.7	93.3
	LA	ReHo	-	100	100	96.6	96.7
		ALFF	-	45	41.7	21.7	18.3
		fALFF	-	100	100	100	100
SZNAH vs. HC	FC	OS-LAAM	L	40	40	20	15
			R	50	40	25	20
		BF-LAAM	L	100	100	90	80
			R	100	95	90	85
	LA	ReHo	-	100	100	95	95
		ALFF	-	60	50	30	20
		fALFF	-	100	100	100	100
SZ vs. HC	FC	OS-LAAM	L	30	27.5	17.5	20
			R	40	35	25	22.5
		BF-LAAM	L	97.5	100	95	92.5
			R	95	95	90	92.5
	LA	ReHo	-	100	100	96.6	96.7
		ALFF	-	47.5	45	37.5	35
		fALFF	-	97.5	100	97.5	97.5

Column HG indicates the left (L) or right (R) Heschl’s Gyrus ROI. Results above 90% are highlighted in bold. Key to abbreviations: FC = Functional connectivity. LA = Local Activity. OS-LAAM = one-sided lattice auto-associative memories. BF-LAAM = background/foreground lattice auto-associative memories. ReHo = regional homogeneity. ALFF = amplitude of low frequency fluctuations. fALFF = fractional amplitude of low frequency fluctuations.