| pubmed-article:21309494 | pubmed:abstractText | The aim of this study was to develop an improved binary logistic regression model for predicting the risk of intracranial aneurysm rupture. A cohort of patients (n=37) with aneurysms underwent three-dimensional digital subtraction angiography examination to measure several morphological parameters of the aneurysm. The aspect ratio (height/neck size) and the size ratio (length/mean diameter of parent vessel) were also calculated. All the morphological parameters combined with the aneurysm location and the patient's baseline data were used to derive a backward binary logistic regression model. In order to validate the model, it was applied to another independent cohort of 19 patients with aneurysms. The model had sensitivity, specificity and accuracy of 84.6%, 66.7% and 78.9%, respectively. This binary logistic regression model of aneurysm rupture risk identified the status of an aneurysm with high accuracy and could form the basis of more complex models in the future. | lld:pubmed |
