Table 1.

Average accuracy 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	97.5	97.5	97.5	92.5
			R	92.5	92.5	95	95.2
		BF-LAAM	L	100	97.5	95	90
			R	100	100	100	100
	LA	ReHo	-	100	100	100	100
		ALFF	-	85	87.5	92.5	92.5
		fALFF	-	97.5	100	100	97.5
SZAH vs. HC	FC	OS-LAAM	L	43.7	42.7	30.7	28.3
			R	52.3	50	33	28
		BF-LAAM	L	96.7	98	96.3	93
			R	98.3	96	92.7	93
	LA	ReHo	-	98	98.3	96.7	96.6
		ALFF	-	48.7	49	30.3	31.7
		fALFF	-	100	100	98.3	98.3
SZnAH vs. HC	FC	OS-LAAM	L	65	63	59.5	55
			R	74	70	60	55
		BF-LAAM	L	100	100	95.5	93
			R	100	98	95.5	93.5
	LA	ReHo	-	97.5	98	95.5	96
		ALFF	-	78	76	62	53
		fALFF	-	100	100	100	100
SZ vs. HC	FC	OS-LAAM	L	32.4	31	26	25
			R	41.4	36.7	26.9	25.2
		BF-LAAM	L	95.5	97.1	88.6	85.7
			R	94.3	91.2	86.7	87.1
	LA	ReHo	-	95.7	95.7	97	95.5
		ALFF	-	48.8	42.9	35.2	35.2
		fALFF	-	98.5	100	97.1	97.1

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.