Table 1.

Performance of each machine learning model when using the whole dataset.

Model	Data type	MAE
SVR	WM data	5.589
SVR	GM data	5.004
SVR	GM+WM data	4.571
SVR	vol	7.187
LR	PC from GM data	13.609
LR	PC from WM data	13.613
GPR	curv	7.200
GPR	thk+vol	6.385
GPR	thk+vol+curv	6.132

The results are presented as the mean MAE for a 5-fold cross validation. WM, white matter volumetric map; GM, gray matter volumetric map; vol, regional volume; curv, regional mean curvature; thk, regional thickness; PC, principal components.