Table 1 All network architectures are listed which are used in this manuscript.
From: Lesion probability mapping in MS patients using a regression network on MR fingerprinting
Network | Loss | Inputs | Outputs | Naming |
|---|---|---|---|---|
1 | MSE | 35 (MRF baseline) | 5 (\(T_1\), \({T_2}^*\), WM-, GM-, lesion prob. maps) | MSE-5 |
2 | MAE | 35 (MRF baseline) | 5 (\(T_1\), \({T_2}^*\), WM-, GM-, lesion prob. maps) | MAE-5 |
3 | LCL | 35 (MRF baseline) | 5 (\(T_1\), \({T_2}^*\), WM-, GM-, lesion prob. maps) | LCL-5 |
4 | MSE | 35 (MRF baseline) | 1 (lesion prob. map) | MSE-1 |
5 | MAE | 35 (MRF baseline) | 1 (lesion prob. map) | MAE-1 |
6 | LCL | 35 (MRF baseline) | 1 (lesion prob. map) | LCL-1 |
7 | DICE | 35 (MRF baseline) | 1 (lesion prob. map) | DICE-1 |
8 | MSE | 2 (\(T_1\), \({T_2}^*\) map) | 1 (lesion prob. map) | MSE-2-1 |