Validating uncertainty in medical image translation
This work addresses uncertainty validation for researchers and clinicians in medical imaging, but it is incremental as it applies an existing method to a new domain.
The paper tackled the problem of unreliable uncertainty measures in deep neural networks for medical image translation, specifically in CT-to-MR tasks, by using dropout to estimate epistemic and aleatoric uncertainty, showing that both types are captured as defined.
Medical images are increasingly used as input to deep neural networks to produce quantitative values that aid researchers and clinicians. However, standard deep neural networks do not provide a reliable measure of uncertainty in those quantitative values. Recent work has shown that using dropout during training and testing can provide estimates of uncertainty. In this work, we investigate using dropout to estimate epistemic and aleatoric uncertainty in a CT-to-MR image translation task. We show that both types of uncertainty are captured, as defined, providing confidence in the output uncertainty estimates.