Table 3.

List of Features for the SVM, RF, and xGB Models Used to Train the Final Multilayer and Multisector Models for Classifying Control and Glaucoma Subjects

SVM				RF				xGB
Layer	Sector	$lo{g}_{10}\left(p\right)$	Folds	Layer	Sector	$lo{g}_{10}\left(p\right)$	Folds	Layer	Sector	$lo{g}_{10}\left(p\right)$	Folds
Superficial	IT	−12.6	121	Superficial	IT	−12.6	121	Superficial	IT	−12.6	121
Deep	IT	−12.0	121	Deep	IT	−12.0	121	Deep	IT	−12.0	121
Retina	IT	−11.9	121	Retina	IT	−11.9	121	Retina	IT	−11.9	121
Superficial	CP	−11.8	121	Superficial	CP	−11.8	121	Superficial	CP	−11.8	121
Superficial	ST	−10.3	121	Superficial	ST	−10.3	121	Superficial	ST	−10.3	121
Retina	CP	−10.3	121	Retina	CP	−10.3	121	Retina	CP	−10.3	121
Superficial	IN	−9.5	121	Superficial	IN	−9.5	121	Superficial	IN	−9.5	121
Retina	ST	−9.4	121	Retina	ST	−9.4	121	Retina	ST	−9.4	121
Deep	CP	−9.2	121	Deep	CP	−9.2	121	Deep	CP	−9.2	121
Deep	SN	−8.7	121	Deep	SN	−8.7	121	Deep	SN	−8.7	121
Superficial	SN	−8.4	121	Superficial	SN	−8.4	121	Superficial	SN	−8.4	121
Deep	ST	−8.2	121	Deep	ST	−8.2	121	Deep	ST	−8.2	121
Retina	IN	−8.1	121	Retina	IN	−8.1	121	Retina	IN	−8.1	121
Superficial	T	−7.7	121	Superficial	T	−7.7	121	Superficial	T	−7.7	121
Superficial	N	−7.6	121	Superficial	N	−7.6	121	Superficial	N	−7.6	121
Retina	SN	−7.5	121	Retina	SN	−7.5	121	Retina	SN	−7.5	121
Deep	IN	−7.2	121	Deep	IN	−7.2	121	Deep	IN	−7.2	121
Choroid	CP	−7.2	121	Choroid	CP	−7.2	121	Choroid	CP	−7.2	121
Retina	T	−7.1	121	Retina	T	−7.1	121	Retina	T	−7.1	121
Choroid	IT	−7.0	121	Choroid	IT	−7.0	121	Choroid	IT	−7.0	121
Retina	N	−6.3	121	Retina	N	−6.3	120	Retina	N	−6.3	121
Choroid	ST	−6.2	121	Choroid	ST	−6.3	93	Choroid	ST	−6.2	121
Deep	N	−6.1	121	Deep	N	−6.3	29	Deep	N	−6.1	121
Choroid	T	−5.0	121					Choroid	T	−5.0	121
Deep	T	−4.7	116					Deep	T	−4.7	121
Choriocapillaris	IT	−4.6	110					Choriocapillaris	IT	−4.6	121
Choroid	IN	−4.6	15					Choroid	IN	−4.4	121
Choroid	SN	−4.3	1					Choroid	N	−4.1	121
								Choroid	SN	−3.9	121

The selected features are ranked from the top to the bottom with the $lo{g}_{10}\left(\mathit{p}\right)$ as the averaged logarithm value over the folds in which the feature was selected.