Quantification of subarachnoid hemorrhage: Prediction of delayed cerebral ischemia and outcome

Aneurysmal subarachnoid hemorrhage (aSAH) is a severe form of stroke that occurs at a relatively young age and has a poor outcome. Delayed cerebral ischemia (DCI) is a complication of aSAH, which is associated with high mortality and a poor functional outcome. The aims of this thesis are firstly to improve the prediction of DCI and secondly to improve the prediction of outcome in patients with aSAH. To achieve this goal we first systematically reviewed the literature on the association of currently commonly used methods of grading the amount of blood after aSAH on NCCT with DCI. Subsequently we developed and present a fully automated SAH segmentation method based on machine learning techniques. Finally, we utilized automated SAH segmentation techniques to develop and validate new prediction models for DCI and outcome using both regression and machine learning models. The most important conclusions of this thesis are that using
quantified total blood volume and machine learning only moderately accurate prediction models for predicting DCI were developed. On the other hand, we have developed accurate prediction models for clinical outcome after aSAH. These models could provide clinicians with valuable information in counselling patients and their families and therefore further optimize the process of shared-decision making.