Table 1.
Network Hyperparameters, Tuning, and Computing for Image Translation using CycleGAN and DC2Anet.
| Implementation (Hyperparameter, Tuning, and Computing) | Value |
|---|---|
| Resolution (FLAIR) | 256 × 256 |
| Resolution (T2W) | 512 × 512 |
| Preprocessing | Rigid registration to ensure 256 × 256 resolution for FLAIR images |
| Input Image Size | 256 × 256 pixels |
| Training Epochs | 400 |
| Batch Size | 2 |
| Optimizer | Adam |
| Learning Rate | 2·10−4 (fixed for the first 200 epochs, linear decay to 0 for the next 200 epochs) |
| Generator Iterations | 3 |
| Discriminator Iterations | 1 |
| Cycle Consistency Loss Weight (λcyc) | 10 |
| Adversarial Loss Weight (λGAN) | 1 |
| Computing Platform | Google Colab |
| Programming Language | Python 3.7 |
| Deep Learning Framework | TensorFlow 2.4.1 |