Lyu, Mengzhe; (2024) In-silico virtual planning of vascular interventional procedures. Doctoral thesis (Ph.D), UCL (University College London).

Text Lyu__thesis.pdf - Other Access restricted to UCL open access staff until 1 January 2025. Download (37MB)

Abstract

A virtual vascular intervention (VVI) tool, based on pre-operative imaging data such as computed tomography angiography (CTA), can be an effective method to predict post-interventional stent and vessel configurations. Additionally, in coronary arteries, by utilising computational fluid dynamics (CFD), the virtual fractional flow reserve (vFFR) can be calculated to measure the severity of coronary artery stenosis based on \added{such a} patient-specific 3D reconstructed coronary model. The combination of a fast virtual coronary intervention (VCI) \added{technique} and vFFR can be a powerful tool to support clinical decision-making by predicting the FFR before and after percutaneous coronary intervention (PCI) in a non-invasive mannor. Although existing VCI tools have succeeded to predict vFFR in PCI, the stenting process in those tools is typically over-simplified as a radius correction (RC) process without physical interaction between stents and vessel. Precise and effective methods to biomechanically simulate the stenting of vessels and post-PCI FFR are expected to improve the prediction, but such methods especially to be conducted in a clinically viable timeframe are lacking. Therefore in this thesis, a novel, fast and more biomechanically representative VCI tool is proposed such that more accurate prediction of stented vessels and post-PCI FFR could be acquired, and potentially predict the result of complex scenarios e.g., multiple stent deployment with rapid computation. The VCI algorithm was developed based on the mass-spring-damper model with calibrated system parameters, incorporating the capability to mimic various relevant materials such as metallic stents and vascular wall tissue. The accuracy of predicted post-VCI FFR, luminal cross-section area, and luminal centreline curvature was accessed by using post-PCI CTA datasets as the reference. Furthermore, as an extension of the VVI framework, the virtual intracranial intervention (VII) tool was developed to simulate intracranial interventional procedure and virtually compare different deployment strategies along with various devices. Also combined with CFD analysis of post-procedural flow, the tool successfully depict the differences between the flow characteristics due to device and deployment strategies and demonstrates its efficacy as a predictive tool to guide intracranial interventions.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item