Publikationstyp: Konferenz: Poster
Art der Begutachtung: Peer review (Abstract)
Titel: What's ugly? : shape-based analysis of intracranial aneurysms
Autor/-in: Juchler, Norman
Schilling, Sabine
Vartan, Kurtcuoglu
Hirsch, Sven
Angaben zur Konferenz: Latsis Symposium 2016 : Personalized Medicine, Zurich, 27-29 June 2016
Erscheinungsdatum: 29-Jun-2016
Sprache: Englisch
Schlagwörter: Shape-based risk assessment; Intracranial aneurysms
Fachgebiet (DDC): 616: Innere Medizin und Krankheiten
Zusammenfassung: It is exceedingly challenging to assess the clinical significance of intracranial aneurysms. Currently, clinicians associate aneurysm shape irregularity with wall instability. However, there is no consensus on which shape features reliably predict aneurysm rupture risk. Here we present a machine learning approach to tackle this problem: We implemented a classification pipeline to identify shape features with predictive power of aneurysm instability. 3D models of aneurysms are extracted from medical imaging data (mostly 3D rotational angiography) using a standardized vessel segmentation protocol. A variety of established representations of the 3D shape are calculated for the extracted aneurysm segment. These include the calculation of Zernike moment invariants (ZMI) and simpler geometry indices such as undulation, ellipticity and non-sphericity. Feature reduction techniques (for ZMI) and classification methods are applied to find patterns linking shape features to aneurysm stability in an exploratory way. This processing pipeline was applied to a clinical dataset of approximately 250 aneurysms registered in the AneurysmDataBase (SwissNeuroFoundation) and AneuriskWeb database. Classification based on ZMI alone allowed us to distinguish between sidewall and bifurcation aneurysms, but failed to forecast an aneurysm’s rupture status reliably. Remarkably, simpler geometry indices performed similarly well in rupture status prediction. It remains to be investigated whether further stratification of the aneurysms in terms of location, size and clinical factors will increase the robustness of the applied classification methods. This study was performed within the scope of the AneuX project, funded by, and received support by SNSF NCCR Kidney.CH. 
Volltext Version: Publizierte Version
Lizenz (gemäss Verlagsvertrag): Lizenz gemäss Verlagsvertrag
Departement: Life Sciences und Facility Management
Organisationseinheit: Institut für Computational Life Sciences (ICLS)
Publiziert im Rahmen des ZHAW-Projekts: AneuX
Enthalten in den Sammlungen:Publikationen Life Sciences und Facility Management

Dateien zu dieser Ressource:
Es gibt keine Dateien zu dieser Ressource.
Zur Langanzeige

Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.