Image-based CFD: Confronting the clinical realities

Nearly half a century has passed since the first modern hemodynamic hypotheses of vascular disease were framed. More than a quarter of a century has passed since engineering experiments confirmed direct links between hemodynamic forces and vascular pathology. And well over a decade has passed since image-based computational fluid dynamics (CFD) was introduced as a means of bringing hemodynamic knowledge from bench to bedside. Yet, despite the fact that hemodynamic disturbances are almost universally acknowledged to play a central role in vascular (patho)physiology, clinical decision-making still rarely incorporates this knowledge in a systematic way. Reasons for this have much to do
with the clinical requirements of fast, straightforward and affordable approaches, and trials proving their efficacy and safety.
In this presentation I will discuss our experiences and efforts in confronting these clinical realities.
For example, “virtual imaging” –
recognizing and exploiting clinical visual vocabularies – has proven to be a useful adjunct to our usual engineering visualization conventions. Our software tools, crafted by engineers for engineers, can be intimidating or inscrutable when presented to a clinical
audience, which serves to widen rather than bridge the gap between the lab and the clinic. Most important, we have come to appreciate that
our engineering models may be relied upon in a qualitative, but not necessarily quantitative sense. As a result, we can (and should) be
willing to sacrifice accuracy for efficacy in the same way that clinicians must balance the desire for sensitivity and specificity
with practicalities of obtaining it.