TABLE 3.
The accuracy of the age prediction measured by Pearson's r (predicted vs. chronological age), R2, root mean square error (RMSE), and mean absolute error (MAE) for each of the cluster‐specific models. 95% confidence intervals are indicated in square brackets. RMSE and MAE are reported in years. N feat represents the number of features contained in the cluster. p‐Values were <.001 for all models
| Model | N feat | r | R2 | RMSE | MAE |
|---|---|---|---|---|---|
| Cluster 1 | 7 | 0.32 [0.31, 0.33] | 0.10 [0.10, 0.11] | 7.00 | 5.79 |
| Cluster 2 | 16 | 0.38 [0.36, 0.39] | 0.14 [0.13, 0.15] | 6.86 | 5.64 |
| Cluster 3 | 11 | 0.48 [0.47, 0.49] | 0.23 [0.23, 0.24] | 6.48 | 5.29 |
| Cluster 4 | 5 | 0.13 [0.11, 0.14] | 0.02 [0.01, 0.02] | 7.36 | 6.13 |
| Cluster 5 | 4 | 0.21 [0.19, 0.22] | 0.04 [0.04, 0.05] | 7.25 | 6.04 |