A pytorch implementation of Vgg16FCN8s for Brain MRI segmentation.
Dataset used for development and evaluation was available on Kaggle: [kaggle.com/mateuszbuda/lgg-mri-segmentation].
Training
python train.py <--epochs 20> <--batch_size 16> <--lr 0.0005>
Testing
python test.py <--ckpt your_ckpt.pt>
Training Loss | Validation IoU |
---|---|
Model | Mean IoU |
---|---|
Vgg16FCN8s | 0.8043 |
The white area is model's prediction and the green line is the Ground Truth.
Testing image | TCGA_CS_4943_20000902_16 | TCGA_DU_8164_19970111_28 | TCGA_CS_4944_20010208_11 |
---|---|---|---|
MRI image | |||
Prediction and Ground Truth |