Table 1 An overview of all methods we compare and benchmark.
From: Jumping over baselines with new methods to predict activation maps from resting-state fMRI
Model name | Proposed here | Parcellation—feature extraction | Type of fitting | # of features |
|---|---|---|---|---|
MMP-RR-PCR | ✓ | MMP—partial correlations | SV ridge regression | 379 |
Rest-task GICA RR | ✓ | ICA on task data | SV ridge regression | 80 |
Rest-rest GICA RR | ✓ | ICA on rest data | SV ridge regression | 80 |
MMP-RR-DR | ✓ | MMP w/ dual regression | SV ridge regression | 379 |
MMP-RR | ✓ | MMP | SV ridge regression | 379 |
GPR-RR | ✓ | Random projection | SV ridge regression | 379 |
AF-Mod | ✓ | Mean activation maps | SV linear regression | 1 |
GICA-DR-OLS12 | ✗ | ICA w/ Dual regression | parcel-wise linear regression | 50 |
MMP-ParcelRR23 | ✗ | MMP | Parcel-wise ridge regression | 360 |
MMP-OLS | ✓ | MMP | SV linear regression | 379 |
AF13 | ✗ | Mean activation maps | None | \(\emptyset\) |
