Uncertainty-Aware Deep Ensembles for Reliable and Explainable Predictions of Clinical Time Series.
Kristoffer WickstrømKarl Øyvind MikalsenMichael KampffmeyerArthur RevhaugRobert JenssenPublished in: IEEE J. Biomed. Health Informatics (2021)
Keyphrases
- chronic hepatitis
- temporal abstractions
- medical databases
- clinical data
- dynamic time warping
- cost effective
- uncertain data
- non stationary
- random forests
- clinical setting
- models built
- conditional probabilities
- ensemble methods
- belief functions
- artificial neural networks
- clinical practice
- neural network ensemble
- disease diagnosis
- confidence intervals
- multivariate time series
- deep learning
- medical domain
- decision theory
- machine learning
- medical data
- multi class
- bayesian networks