Figure 2.

U-Net architecture for voxel-level brain-age prediction. Raw T1-weighted MRI scans were pre-processed using SPM12, obtaining modulated grey and white matter volume maps registered to the MNI152 template. 52³ blocks of both grey and white matter are passed through U-Net architecture to obtain 12³ blocks of voxel-level brain-age prediction. Additional auxiliary block-level brain-age loss functions were added at each level of the U-Net to facilitate training.