Effect of the output activation function on the probabilities and errors in medical image segmentation

Abstract

The sigmoid activation is the standard output activation function in binary classification and segmentation with neural networks. Still, there exist a variety of other potential output activation functions, which may lead to improved results in medical image segmentation. In this work, we consider how the asymptotic behavior of different output activation and loss functions affects the prediction probabilities and the corresponding segmentation errors. For cross entropy, we show that a faster rate of change of the activation function correlates with better predictions, while a slower rate of change can improve the calibration of probabilities. For dice loss, we found that the arctangent activation function is superior to the sigmoid function. Furthermore, we provide a test space for arbitrary output activation functions in the area of medical image segmentation. We tested seven activation functions in combination with three loss functions on four different medical image segmentation tasks to provide a classification of which function is best suited in this application scenario.

References in corpus (8)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• The Medical Segmentation Decathlon
• Activation Functions: Comparison of trends in Practice and Research for Deep Learning
• TernausNet: U-Net with VGG11 Encoder Pre-Trained on ImageNet for Image Segmentation
• A survey on modern trainable activation functions
• What makes ImageNet good for transfer learning?
• Pitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning
• Padé Activation Units: End-to-end Learning of Flexible Activation Functions in Deep Networks