Investigating the effect of cannulation on blood flow during extracorporeal membrane oxygenation using positron emission particle tracking

Extracorporeal membrane oxygenation (ECMO) is a life-saving technique that can provide support for patients with both cardiac and respiratory dysfunction. Different variations of ECMO are used depending on the needs of the patient, but these all rely on the cannulation of particular blood vessels. Although ECMO has been successfully used as a form of life support for over 30 years, there are complications that can arise during treatment, which increase the mortality rate significantly. For many of these complications, there is a limited understanding of the causes, making it challenging to propose and find solutions. When patients are treated using ECMO, the angle of insertion of the cannula will vary on a case-by-case basis. This will impact the flow from, and around, the cannula, and thus finding an optimal angle could minimise the turbulence in the flow, and therefore reduce the occurrence of complications, such as blood clots. Additionally, the datasheets for the cannulae used in clinical settings only provide measurements of water flow through the cannula, in coarse units of L/min.
The aim of this project is to investigate the blood flow in such a system with higher precision, and in turn, provide insight on potential causes of some of the common complications. To do this, experimental flow measurements through a cannula and model vascular system can be acquired using positron emission particle tracking (PEPT). PEPT is an imaging technique, pioneered at the University of Birmingham, that is used to track the motion of individual radioactively labelled tracer particles. The first run of experiments has been carried out while systematically varying both the angle of insertion of the cannula and the fluid used in the model system.
In parallel to the experimental work, numerical simulations of the system using computational fluid dynamics (CFD) are being developed. These simulations will be compared to the experimental results, and subsequently used to test more complex systems. Initially, both the experimental and simulation data will be analysed using existing methods. Following this, work will be carried out to explore the potential benefits of incorporating topological methods into the reconstruction and analysis algorithms.