Recalibration of Aleatoric and Epistemic Regression Uncertainty in Medical Imaging
This work addresses the need for well-calibrated uncertainty in medical imaging to enable robust rejection of unreliable predictions, though it is incremental as it adapts an existing calibration technique to a specific domain.
The paper tackles the systematic underestimation of predictive uncertainty in medical imaging regression tasks by applying a simple σ scaling method to calibrate both aleatoric and epistemic uncertainty, showing it reliably recalibrates uncertainty while maintaining accuracy across various datasets and architectures.
The consideration of predictive uncertainty in medical imaging with deep learning is of utmost importance. We apply estimation of both aleatoric and epistemic uncertainty by variational Bayesian inference with Monte Carlo dropout to regression tasks and show that predictive uncertainty is systematically underestimated. We apply $ σ$ scaling with a single scalar value; a simple, yet effective calibration method for both types of uncertainty. The performance of our approach is evaluated on a variety of common medical regression data sets using different state-of-the-art convolutional network architectures. In our experiments, $ σ$ scaling is able to reliably recalibrate predictive uncertainty. It is easy to implement and maintains the accuracy. Well-calibrated uncertainty in regression allows robust rejection of unreliable predictions or detection of out-of-distribution samples. Our source code is available at https://github.com/mlaves/well-calibrated-regression-uncertainty