AUTHOR=Bizjak Žiga , Pernuš Franjo , Špiclin Žiga 

TITLE=Deep Shape Features for Predicting Future Intracranial Aneurysm Growth

JOURNAL=Frontiers in Physiology

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/physiology/articles/10.3389/fphys.2021.644349

DOI=10.3389/fphys.2021.644349

ISSN=1664-042X

ABSTRACT=Introduction: Intracranial aneurysms (IAs) are a common vascular pathology and are associated with a risk of rupture, which is often fatal. Aneurysm growth is considered a surrogate of rupture risk, therefore, the study aim is to develop and evaluate prediction models of future AI growth based on baseline aneurysm morphology as a computer-aided treatment decision support.

Materials and Methods: Follow-up CTA and MRA angiograms of 39 patients with 44 IAs were classified into growing and stable (25/19). From the angiograms vascular surface meshes were extracted and the aneurysm shape characterized by established morphologic features and novel deep shape features. The features corresponding to the baseline aneurysms were used to predict future aneurysm growth using univariate thresholding, multivariate random forest and multi-layer perceptron learning and deep shape learning based on PointNet++ model. 

Results: Proposed deep shape features learning method achieved accuracy of 0.82 (sensitivity=0.96, specificity=0.63), while the multivariate learning and univariate thresholding methods were inferior with accuracy up to 0.68 and 0.63, respectively.

Conclusion: High performing classification of future growing IAs renders the proposed deep shape features learning approach as the key enabling tool to manage rupture risk in the ”no treatment” paradigm of patient follow-up imaging. 

Results show, that proposed deep learning method is superior over univariate and multivariate methods for aneurysm growth classification, and can improve the process of aneurysm monitoring, consequently improving patient safety.