About this document
Aleatoric Uncertainty in Medical Segmentation by brecheisen is a document available to read on EtoBox.
1) The document proposes a method to estimate aleatoric uncertainty for medical image segmentation using convolutional neural networks. Aleatoric uncertainty accounts for noise inherent in the data. 2) The method uses test-time augmentation, where a distribution of predictions is estimated by Monte Carlo simulation with transformations and noise added to the input images. 3) The paper evaluates the proposed test-time augmentation-based aleatoric uncertainty estimation on fetal brain and brain tumor segme
- Author
- brecheisen
- Language
- EN