‘Patient-specific computer models target individuals’
Dr Raoul van Loon is a lecturer in Biomedical Engineering at Swansea University.
It became clear to me that airflow around wings, water flow through rivers or the blood flow in an artery, are very similar and can be described with the same equations.
With the immense increase in computerpower, large complex calculations have become feasible and, as a result, these computer models are nowadays an essential part in the design and analysis of buildings, bridges or aeroplanes.
My research exploits the strengths of computer modelling with the aim of enhancing artificial heart valve design or to improve diagnostics of arterial disease.
One of the challenges in this research is caused by the large variability between humans, while often the diagnostic criteria currently used by clinicians in hospitals are based on averages.
Also, the human body is alive and changing, which is a pain for computer models.
In my work I create sophisticated computer models that overcome these difficulties and are tailor-made for a given patient by extracting the arterial geometry from ultrasound scans, MRI scans or CT scans of that patient.
Using these geometries we develop computer models that predict blood flow, blood pressure or heart valve motion in much higher detail than can be measured.
Finally, abnormalities in the flow can be examined to predict growth of calcifications or plaque rupture.
This process is called patient-specific modelling, which means the results are very detailed and targeted to one individual patient.
These very large computer models require several days of calculations on many computers, which is perfectly acceptable for research purposes. However, it would be impossible to do this for every single patient in the hospital.
Therefore, we need to translate our in-depth understanding of the complex three-dimensional flow models into more simplified models that cardiologists can easily use on a day-to-day basis.
A project I am working on with consultant cardiologist Adrian Ionescu, from Morriston Hospital, in Swansea, aims to develop such simplified models.
Assessing the efficiency of one diseased heart valve is quite straightforward, but when you have two diseased valves the recommendations are not very clear.
The reason is that the interactions between multiple diseased valves are extremely difficult to interpret.
However, computer models are particularly good for analysing these complex interactions and they could be a great addition as part of the software in ultrasound machines, which are daily used by cardiologists.
One of the joys of being a biomedical engineer is the interaction with specialists in hospital who work in a different world, speak a different language and have a different thinking pattern.
Breaking down these engineering-medicine barriers is a big challenge on its own, but can have a crucial impact on health care in Wales and beyond.
To contact Raoul please email r.vanloon@swansea.ac.uk.
This article first appeared in the Western Mail‘s Health Wales supplement on the 3rd October 2011, as part of the Welsh Crucible series of research profiles.