Technical Specifications

- Method: Deep Learning
- Models: UNet, SegNet, V-Net
- Datasets: LIDC-IDRI, NSCLC Radiomics, Kaggle Data Science Bowl
- Accuracy: 85-95% depending on dataset and preprocessing
- Training Time: 12-48 hours
- Hardware Requirements: NVIDIA Tesla V100 GPU, 32GB RAM
- Programming Language: Python
- Libraries: TensorFlow, Keras, PyTorch
- Operating System: Ubuntu 18.04
- Developer: Neural* AI Research Lab & TsinghuaAIR Lab