Vertebral artery fusiform aneurysm geometry in predicting rupture risk

Abstract

Cerebral aneurysms affect a significant portion of the adult population worldwide. Despite significant progress, the development of robust techniques to evaluate the risk of aneurysm rupture remains a critical challenge. We hypothesize that vertebral artery fusiform aneurysm (VAFA) morphology may be predictive of rupture risk and can serve as a deciding factor in clinical management. To investigate the VAFA morphology, we use a combination of image analysis and machine learning techniques to study a geometric feature set computed from a depository of 37 (12 ruptured and 25 un-ruptured) aneurysm images. Of the 571 unique features we compute, we distinguish five features for use by our machine learning classification algorithm by an analysis of statistical significance. These machine learning methods achieve state-of-the-art classification performance (81.43 ± 13.08%) for the VAFA morphology, and identify five features (cross-sectional area change of aneurysm, maximum diameter of nearby distal vessel, solidity of aneurysm, maximum curvature of nearby distal vessel, and ratio of curvature between aneurysm and its nearby proximal vessel) as effective predictors of VAFA rupture risk. These results suggest that the geometric features of VAFA morphology may serve as useful non-invasive indicators for the prediction of aneurysm rupture risk in surgical settings.

Predicting the risk of rupture for vertebral aneurysm based on geometric features of blood vessels

“…= 10.12) with 12 ruptured and 25 unruptured aneurysms were included in the study. All three-dimensional images in stereolithography (STL) format of vertebral artery aneurysm images are processed and geometric features are extracted, following the procedure in [ 5 ]. The images are initially processed by MeshLab [ 19 ], an open source software, to derive geometric characteristics.…”

“…The curvature and torsion are evaluated by using Frenet–Serret formulae. Additional geometric indices are then generated, including maximum, minimum and equivalent diameters (the diameter of a circle with the same area as the region), cross-sectional area, area change rate, eccentricity, solidity (the proportion of the pixels in the convex hull) and extent (the ratio of pixels in the region to pixels in the total bounding box) [ 5 ]. Definitions of several important geometric indices are listed below and illustrated in figure 1 .…”

“…Based on the segmentation, we generated three groups of features: (i) the maximum, mean, standard deviation, integration and variation of the geometric index in each segment; (ii) the ratios of features in (i) between different segments; and (iii) the angle ( α ) and distance ( d ) between two centreline tangent vectors at the neck of the bulge ( figure 2 b ). We generated a new dataset (37 × 384) with less features than those in [ 5 ] (37 × 563). A complete list of features can be found in electronic supplementary material, appendix A.…”

“…Based on their shape, aneurysms are categorized into saccular and fusiform [5]. Previous studies have shown the importance of haemodynamic and morphological evaluation in predicting the rupture of saccular aneurysms [11][12][13][14][15][16].…”

“…Zhao et al [ 5 ] proposed a model based on five geometric features of VAFA morphology and achieved state-of-the-art classification performance ( ). The feature generation procedure and selected features depend heavily on the information of the aneurysm itself.…”