. 2022 Dec 15;11:e77945. doi: 10.7554/eLife.77945

Table 1. MRI datasets used in training, evaluation, and comparison to extant methods.

Methods employed include those proposed here (HippUnfold), the same processing but with manual segmentation (similar to previous work; DeKraker et al., 2020) (manual unfold), Freesurfer v7.2.0 (FS7) (Iglesias et al., 2015), and an atlas of manual segmentations (Berron et al., 2017) used in ASHS (Yushkevich et al., 2015b).

Name	Modalities	Resolution	Sample size (L+R)	Methods employed
HCP-YA	T1w, T2w	0.7 × 0.7 × 0.7 mm³	n=590 (training)	HippUnfold Manual unfold
HCP-YA	T1w, T2w	0.7 × 0.7 × 0.7 mm³	n=148 (testing)	HippUnfold Manual unfold FS7
HCP-A	T1w T2w SPACE T2w TSE	0.8 × 0.8 × 0.8 mm³ 0.8 × 0.8 × 0.8 mm³ 0.4 × 0.4 × 2.0 mm³	n=1312 for T1w, T2w SPACE n=200 for T2w TSE (FS7, ASHS) n=200 for T1w (HippUnfold)	HippUnfold FS7 ASHS
7T-TSE (from ASHS atlas)	T2w	0.4 × 0.4 × 1.0 mm³	n=70	HippUnfold Manual subfields