Uncertainty for Safe Utilization of Machine Learning in Medical Imaging

This book constitutes the refereed proceedings of the Fourth Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, UNSURE 2022, held in conjunction with MICCAI 2022. The conference was hybrid event held from Singapore. For this workshop, 13 papers from 22 submissions were accepted for publication. They focus on developing awareness and encouraging research in the field of uncertainty modelling to enable safe implementation of machine learning tools in the clinical world.

Inhalt

Uncertainty Modelling.- MOrphologically-aware Jaccard-based ITerative Optimization (MOJITO) for Consensus Segmentation.- Quantification of Predictive Uncertainty via Inference-Time Sampling.- Uncertainty categories in medical image segmentation: a study of source-related diversity..- On the pitfalls of entropy-based uncertainty for multi-class semi-supervised segmentation.- What Do Untargeted Adversarial Examples Reveal In Medical Image Segmentation?..- Uncertainty calibration.- Improved post-hoc probability calibration for out-of-domain MRI segmentation..- Improving error detection in deep learning-based radiotherapy autocontouring using Bayesian uncertainty.- A Plug-and-Play Method to Compute Uncertainty.- Calibration of Deep Medical Image Classifiers: An Empirical Comparison using Dermatology and Histopathology Datasets.- Annotation uncertainty and out of distribution management.- nnOOD: A Framework for Benchmarking Self-supervised Anomaly Localisation Methods.- Generalized Probabilistic U-Net for medical image segmentation.- Joint paraspinal muscle segmentation and inter-rater labeling variability prediction with multi-task TransUNet.- Information Gain Sampling for Active Learning in Medical Image Classification.