Table 1 Experimental settings.
Configuration category | Key parameter | Set the value |
|---|---|---|
Hardware environment | Processor | Intel Xeon Gold 6248R @ 3.0 GHz (2) |
Memory capacity | 256GB DDR4 ECC | |
Display card configuration | NVIDIA RTX A6000 (48GB video memory)x4 | |
Inventory system | 2 TB NVMe SSD + 16 TB HDD | |
Software environment | Operating system | Ubuntu 20.04 LTS |
Deep learning framework | PyTorch 1.12.1 + CUDA 11.6 | |
Python edition | 3.8.12 | |
Model parameter | Input dimensions | 512 × 512 |
Batch size | 16 | |
Optimizer | Adam | |
Learning rate strategy | Cosine damping | |
Number of training rounds | 200 | |
Loss function | Dice + CE |