Learning to Predict Error for MRI Reconstruction
arXiv:2002.05582
Abstract
In healthcare applications, predictive uncertainty has been used to assess predictive accuracy. In this paper, we demonstrate that predictive uncertainty estimated by the current methods does not highly correlate with prediction error by decomposing the latter into random and systematic errors, and showing that the former is equivalent to the variance of the random error. In addition, we observe that current methods unnecessarily compromise performance by modifying the model and training loss to estimate the target and uncertainty jointly. We show that estimating them separately without modifications improves performance. Following this, we propose a novel method that estimates the target labels and magnitude of the prediction error in two steps. We demonstrate this method on a large-scale MRI reconstruction task, and achieve significantly better results than the state-of-the-art uncertainty estimation methods.
arXiv commentAccepted to MICCAI 2021
References in corpus (8)
- What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
- On Calibration of Modern Neural Networks
- Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
- fastMRI: An Open Dataset and Benchmarks for Accelerated MRI
- Accurate Uncertainties for Deep Learning Using Calibrated Regression
- Decomposition of Uncertainty in Bayesian Deep Learning for Efficient and Risk-sensitive Learning
- Deep Bayesian Inversion
- i-RIM applied to the fastMRI challenge