Skip to main content

Table 1 Performance results of the threefold cross-validation for tumor and kidney segmentation with a standard 3D U-Net model on 120 CT scans from the KiTS19 challenge

From: MIScnn: a framework for medical image segmentation with convolutional neural networks and deep learning

Metric Training Validation
Tversky loss 0.3672 0.4609
Soft Dice similarity coefficient 0.8776 0.8235
Categorical cross-entropy − 0.8584 − 0.7899
Dice similarity coefficient: background 0.9994
Dice similarity coefficient: kidney 0.9319
Dice similarity coefficient: tumor 0.6750
  1. Each metric is computed between the provided ground truth and our model prediction and then averaged between the three folds