Background: Accurate prediction of tissue outcome in acute ischaemic stroke is critical for effective treatment and can significantly influence clinical outcome. Traditional deterministic prediction models are unable to account for or communicate uncertainty.
Aim: The aim of this paper is to investigate and evaluate methods for quantifying uncertainty in tissue outcome prediction, in particular its communication through visual processing.
Methods: Different prediction models based on high-dimensional perfusion data will be investigated to predict lesions in the brain after stroke. These models will be validated with real patient data from several clinical centres to compare their predictive performance.
Results: The different models show different levels of uncertainty assessment. Their pictorial representation helps to provide a multifaceted context in a condensed form and to refine the ground truth.
Conclusions: The results highlight the importance of further development and application of advanced statistical methods for uncertainty quantification in stroke care. Future research should focus on integrating these models into clinical decision-making processes to improve treatment outcomes.