Despite the large overall beneficial effects of endovascular treatment in patients with acute ischemic stroke, severe disability or death still occurs in almost one-third of patients. These... Show moreDespite the large overall beneficial effects of endovascular treatment in patients with acute ischemic stroke, severe disability or death still occurs in almost one-third of patients. These patients, who might not benefit from treatment, have been previously identified with traditional logistic regression models, which may oversimplify relations between characteristics and outcome, or machine learning techniques, which may be difficult to interpret. We developed and evaluated a novel evolutionary algorithm for fuzzy decision trees to accurately identify patients with poor outcome after endovascular treatment, which was defined as having a modified Rankin Scale score (mRS) higher or equal to 5. The created decision trees have the benefit of being comprehensible, easily interpretable models, making its predictions easy to explain to patients and practitioners. Insights in the reason for the predicted outcome can encourage acceptance and adaptation in practice and help manage expectations after treatment. We compared our proposed method to CART, the benchmark decision tree algorithm, on classification accuracy and interpretability. The fuzzy decision tree significantly outperformed CART: using 5-fold cross-validation with on average 1090 patients in the training set and 273 patients in the test set, the fuzzy decision tree misclassified on average 77 (standard deviation of 7) patients compared to 83 (+/- 7) using CART. The mean number of nodes (decision and leaf nodes) in the fuzzy decision tree was 11 (+/- 2) compared to 26 (+/- 1) for CART decision trees. With an average accuracy of 72% and much fewer nodes than CART, the developed evolutionary algorithm for fuzzy decision trees might be used to gain insights into the predictive value of patient characteristics and can contribute to the development of more accurate medical outcome prediction methods with improved clarity for practitioners and patients. Show less
3D Active Shape Modeling is a technique to capture shape information from a training set containing characteristic shapes of, e.g., a heart. The description contains a mean shape, and shape... Show more3D Active Shape Modeling is a technique to capture shape information from a training set containing characteristic shapes of, e.g., a heart. The description contains a mean shape, and shape variations (e.g. eigen deformations and eigen values). Many models based on these statistics, and used for medical purposes, are limited in their application by properties of the training set. These properties are, the radiological modality used for image acquisition (1), the orientation of the images (2), and the density of the data (3). In our model of the cardiac left ventricle, the limits above were relieved by not training on gray level information (1,2), using a Fuzzy Inference System for finding image features in CT and MR data (1), and weighted propagation of image features along the model surfaces (3). Evalutation of our model on both CT and MR showed almost equal performance. Evaluation of three different point-correspondences used for model- building did not clearly indicate one as best. By extended tuning using a grid-computing approach, and propagation of image features, we managed to achieve segmentation results on sparse data comparable to those of others on dense data. Our model is more widely applicable thanks to these developments. Show less