Table 5.
Brain age prediction results reported in literatures.
| Articles | Modalties | Input data | Methods | Subjects (Age range) | r | MAE | R2 |
|---|---|---|---|---|---|---|---|
| Cole et al. (21) | sMRI | GM+WM volumes | GPR | 2001 NC (18–90) | 0.94 | 5.02 | 0.88 |
| Cole et al. (25) | sMRI | GM volume map Raw MRI map |
CNN | 2001 NC (18–90) | 0.96 0.94 | 4.16 4.65 | 0.92 0.88 |
| Franke et al. (17) | sMRI | GM+WM volumes | RVR | 394 NC (5–18) | 0.93 | 1.10 | - |
| Franke et al. (16) | sMRI | GM volume | RVM | 655 NC (19–86) | 0.92 | 5.00 | - |
| Li et al. (40) | rs-fMRI | Functional connectivity | CNN | 983 NC (8–22) | 0.61 | 2.15 | |
| Liem et al. (19) | rs-fMRI + sMRI | Functional connectivity;Structural measures | SVR+RF | 2354 NC (19–82) | - | 4.29 | - |
| Lin et al. (23) | DTI | Topological network properties | ANN | 112 NC (50–79) | 0.80 | 4.29 | - |
| Valizadeh et al. (20) | sMRI | Anatomical feature sets | SVM NN | 3144 NC (7–96) | - | - | 0.84 0.84 |
sMRI, Structural Magnetic Resonance Imaging; rs-fMRI, resting-state Functional Magnetic Resonance Imaging; DTI, Diffusion Tensor Imaging; GM, Gray Matter; WM, White matter; Raw MRI, rigid only registrated structural MRI; Structural measures: cortical thickness, cortical surface area, and subcortical volumes; Anatomical feature sets: cortical volume, thickness, area, subcortical volume, cerebellar volume, etc; GPR, Gaussian Process Regression; CNN, Convolutional Neural Network; RVR, Relevance Vector Regression; RVM, Relevance Vector Machine; SVR, Support Vector Regression; RF, Random Forest; ANN, Artificial Neural Network; SVM, Support Vector Machine; NN, Neural Network; NC, normal controls.