Author(s): Ong, Gabriela M. (2018)
Abstract:
Among the challenges in the simulation of haemodynamics in the human brain is the problem of connecting the raw data from medical imagery to a quantitative, patient-specific prediction of the flow in order to provide information assisting practitioners when deciding on a treatment. In this study a sequence of steps is presented to facilitate such connection, focusing on an endovascular treatment known as flow-diverter stenting. Starting from DICOM data, the format changes and preparations based on Horos and Blender are described, as pre-cursors to an immersed boundary (IB) representation of the geometry. On the IB representation, a fluid mechanics analysis within the CFD toolbox OpenFOAM is done, providing both qualitative and quantitative descriptions of the flow. This sequence of steps is discussed and illustrated for flow in a model geometry with the typology of a side-wall aneurysm, under steady-state and pulsating flow conditions. The flow diverting effect of two different stents is presented, showing a significant blocking of flow from entering the aneurysm sac after stent placement. Under suitable spatial resolution conditions accurate results are obtained and we observe that both a fine-mesh flow diverter and an eCLIPS device perform virtually identically in terms of almost completely halting the flow in the aneurysm sac. A slight preference for an eCLIPS can be motivated, showing a more uniform flow blocking also in the immediate vicinity of the device, combined with more flexibility of merging it with the diseased vessel.
Document(s):
Ong_MA_EEMCS.pdf