Rupture Status Prediction for Intracranial Aneurysms

Alexandra Lauric^a, Eric Miller^a,b, Merih Baharoglu ^d, Adel M. Malek^c,d

^a Tufts University, Department of Computer Science

^b Tufts University, Department of Electrical and Computer Engineering

^c Tufts University, School of Medicine

^d Tufts Medical Center, Department of Neurosurgery

Intracranial aneurysms are polymorphic focal arterial dilations, which harbor a variable risk of rupture leading to high morbidity and mortality. Increased detection of incidental aneurysms by non-invasive imaging has created a need for rupture risk stratification tools, in addition to simple aneurysm size, to guide optimal treatment strategy. To this end, shape analysis has emerged as a possible differentiator of rupture likelihood.

A novel set of morphological parameters based on the writhe number are introduced here to describe aneurysms and predict rupture status. Classification in 117 saccular aneurysms (52 ruptured and 65 unruptured) was based on statistical analysis of writhe number distribution on the aneurysm surface. Aneurysms were analyzed both completely isolated and including their parent vessel. Sidewall and bifurcation aneurysm subtypes were found to be best described by disjoint sets of shape parameters, yielding a morphological dichotomy between the two aneurysm classes.

Using writhe number analysis resulted in a 20% increase in prediction accuracy for both sidewall and bifurcation subtypes, compared to the performance of the established aspect ratio index. These results support the utility of writhe number aneurysm shape analysis as a high order descriptor with potential clinical utility in rupture risk stratification.