Supervised Uncertainty Quantification for Segmentation with Multiple Annotations
This work addresses uncertainty quantification for medical segmentation to inform doctors about confidence, but it is incremental as it builds on existing probabilistic models.
The paper tackled the problem of uncalibrated predictive uncertainty in medical segmentation by using multi-annotator variability as groundtruth aleatoric uncertainty in a supervised learning approach, resulting in improved uncertainty estimates, sample accuracy, and diversity on lung nodule and prostate MRI datasets.
The accurate estimation of predictive uncertainty carries importance in medical scenarios such as lung node segmentation. Unfortunately, most existing works on predictive uncertainty do not return calibrated uncertainty estimates, which could be used in practice. In this work we exploit multi-grader annotation variability as a source of 'groundtruth' aleatoric uncertainty, which can be treated as a target in a supervised learning problem. We combine this groundtruth uncertainty with a Probabilistic U-Net and test on the LIDC-IDRI lung nodule CT dataset and MICCAI2012 prostate MRI dataset. We find that we are able to improve predictive uncertainty estimates. We also find that we can improve sample accuracy and sample diversity. In real-world applications, our method could inform doctors about the confidence of the segmentation results.