Computational modelling of vascular interventions: endovascular device deployment

Thesis

Minimally invasive vascular interventions with stent deployment have become a popular alternative to conventional open surgery in the treatment of many vascular disorders. However, the high initial success rates of endovascular repairs have been overshadowed by reported complications that cause re-interventions and, in the worst case, morbidity and mortality. The dangerous complications could be mitigated by better choice of device design and by the appropriate positioning of the implant inside the vessel. However, there is currently no possibility for the interventionist to predict the resulting position and the expanded shape of the device for a given patient, before the actual procedure, within the clinical setting.

Motivated by this unmet clinical need and the lack of suitable methods, this thesis develops a methodology for modelling virtual deployment of implantable devices inside patient vessels, that features fast computational execution times and can be used in clinical practice. This novel deployment method was developed based on a spring-mass model and was tested in different deployment scenarios, expanding stents inside vessels in the order of seconds. Further, the performance of the novel method was optimised by calibrating a set of parameters with the help of a genetic algorithm, which utilises the outcomes of a finite element analysis as a learning reference. After the calibration, the developed stenting method demonstrated acceptable accuracy as compared to the "gold standard" of the finite element simulation. Finally, on a real patient case, 4 alternative stenting scenarios were investigated by comparing the subsequent blood flow conditions, via computational haemodynamics. The obtained results suggested that device design, dimensions, stiffness and positioning have important implications on the post-procedural haemodynamics of the vessel. Ultimately, the presented results can play a transformative role in aiding clinical decision-making and also give rise to overall improvements in implant design and deployment procedure.

Authors: Spranger, K

• Publication date: 2014
• DOI: 10.5287/ora-dpxqaxqkj
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Genetic algorithm; Modelling; Stent deployment; Virtual stenting; FEA
• Subjects: Biomedical engineering