Uncertainty for Safe Utilization of Machine Learning in Medical Imaging

Chapter 1 Propagation and Attribution of Uncertainty in Medical Imaging Pipelines

Chapter 2 RR-CP: Reliable-Region-Based Conformal Prediction for Trustworthy Medical Image Classification

Chapter 3 Bayesian Uncertainty Estimation in Landmark Localization Using Convolutional Gaussian Processes

Chapter 4 TriadNet: Sampling-Free Predictive Intervals for Lesional Volume in 3D Brain MR Images

Chapter 5 Examining the Effects of Slice Thickness on the Reproducibility of CT Radiomics for Patients with Colorectal Liver Metastases

Chapter 6 Benchmarking Scalable Epistemic Uncertainty Quantification in Organ Segmentation

Chapter 7 Numerical Uncertainty of Convolutional Neural Networks Inference for Structural Brain MRI Analysis

Chapter 8 How Inter-rater Variability Relates to Aleatoric and Epistemic Uncertainty: A Case Study with Deep Learning-Based Paraspinal Muscle Segmentation

Chapter 9 Uncertainty Estimation and Propagation in Accelerated MRI Reconstruction

Chapter 10 Uncertainty-Based Quality Assurance of Carotid Artery Wall Segmentation in Black-Blood MRI

Chapter 11 Multi-layer Aggregation as a Key to Feature-Based OOD Detection

Chapter 12 Feature-Based Pipeline for Improving Unsupervised Anomaly Segmentation on Medical Images

Chapter 13 Redesigning Out-of-Distribution Detection on 3D Medical Images

Chapter 14 On the Use of Mahalanobis Distance for Out-of-distribution Detection with Neural Networks for Medical Imaging

Chapter 15 Dimensionality Reduction for Improving Out-of-Distribution Detection in Medical Image Segmentation

Chapter 16 Breaking Down Covariate Shift on Pneumothorax Chest X-Ray Classification

Chapter 17 Robustness Stress Testing in Medical Image Classification

Chapter 18 Confidence-Aware and Self-supervised Image Anomaly Localisation

Chapter 19 Uncertainty Estimation in Liver Tumor Segmentation Using the Posterior Bootstrap

Chapter 20 Pitfalls of Conformal Predictions for Medical Image Classification

Chapter 21 Proper Scoring Loss Functions Are Simple and Effective for Uncertainty Quantification of White Matter Hyperintensities