Table 5.

Results of the classification analysis for “old stroke” and “no-stroke” individuals. The AUC values computed by averaging the results of the validation data (mean ± SD) are shown for the two models (SVM with linear kernel and RF) and for all the MRI sequences and brain tissues/structures when using the texture features extracted from the 5 texture analysis methods. The presented values are obtained for the best tuning parameter in each case.

AUC: Mean (SD)		GLRLM			GLCM			LBP			WCF			WSF
		NAWM	SS	WMH	NAWM	SS	WMH	NAWM	SS	WMH	NAWM	SS	WMH	NAWM	SS	WMH
FLAIR	RF	<0.6	0.691 (0.109)	0.674 (0.108)	<0.6	0.612 (0.099)	0.608 (0.111)	<0.5	0.742 (0.100)	<0.6	0.761 (0.097)	0.647 (0.099)	0.702 (0.108)	0.669 (0.114)	0.635 (0.094)	<0.6
	SVM	<0.6	0.676 (0.097)	0.770 (0.089)	<0.5	0.666 (0.090)	0.614 (0.124)	<0.5	0.751 (0.103)	0.682 (0.136)	0.637 (0.121)	<0.6	<0.6	0.637 (0.137)	<0.6	<0.6
T2W	RF	<0.5	0.643 (0.099)	<0.6	<0.5	0.641 (0.107)	0.617 (0.102)	<0.5	0.680 (0.112)	0.608 (0.116)	<0.5	0.680 (0.097)	0.752 (0.097)	<0.5	0.705 (0.116)	0.665 (0.103)
	SVM	0.665 (0.084)	0.738 (0.121)	0.646 (0.128)	0.601 (0.138)	0.644 (0.111)	<0.5	<0.6	0.763 (0.116)	0.671 (0.122)	<0.5	0.608 (0.157)	<0.5	<0.6	0.737 (0.103)	0.677 (0.123)
T1W	RF	0.609 (0.104)	0.654 (0.112)	<0.6	<0.6	0.662 (0.091)	0.659 (0.113)	0.667 (0.125)	0.649 (0.120)	0.611 (0.140)	0.624 (0.105)	0.682 (0.109)	0.664 (0.115)	<0.6	<0.5	<0.6
	SVM	<0.6	0.662 (0.104)	<0.5	0.642 (0.126)	<0.6	<0.5	<0.6	0.676 (0.122)	0.630 (0.126)	0.628 (0.143)	<0.6	<0.6	<0.6	<0.5	<0.6

^*Values in bold indicate the best AUC results (AUC > 0.7).

AUC: area under the curve, RF: random forest classifier, SVM: support vector machine classifier, GLRM: grey–level run length matrix features, GLCM: grey-level co-occurrence matrix features, LBP: local binary patterns features, WCF: wavelet co-occurrence features, WSF: wavelet statistical features, NAWM: normal-appearing white matter, SS: subcortical structures, WMH: white matter hyperintensities.