Page 153 - Visions of the Future Chemistry and Life Science

140 P. KOHL ET AL.

and electro-mechanical function, the whole sequence of the natural heart-
beat may be reproduced.
The same applies to pathologically-disturbed function. A simulated
reduction in coronary blood flow (heart attack) would lead to reduced
oxygen supply to the cells in the virtual heart, which would reduce effi-
ciency of cardiac contraction and possibly give rise to heart rhythm distur-
bances. Ventricular pressure development would be compromised, as
would the blood supply to all organs of the body, including the heart. All
these implications can be studied in a virtual heart.
This possesses an immense potential, not only for bio-medical
research, but also for clinical applications, including patient-specific mod-
elling of therapeutic interventions. For example, dynamic changes in a
patient’s cardiac anatomy can already be modelled on the basis of a non-
invasive technique called Magnetic Resonance Imaging. The location of
coronary vessels for that patient may be determined by 3D coronary angio-
graphy, a common procedure in the context of coronary surgery, and imple-
mented into the virtual heart. Integrated patient-specific virtual hearts of
this kind may, therefore, already be constructed – however, not in real time
yet. Once the interrelation between computational demand and comput-
ing power has sufficiently improved (and it does get better all the time),
virtual hearts will be used for the prediction of optimal coronary bypass
procedures, aid the surgeon’s decision on the operative approach, and even
predict the potential long-term consequences of various treatment strate-
gies.
So – what about the prospects of powerful computational equipment?
The calculations for Figure 8.2, for example, took about six hours on a 16-
processor Silicon Graphics Power Challenge, which is a fairly powerful
computer. Six hours to calculate a single heartbeat! However, there are
already much more powerful systems that could cope with the same task
in less than an hour. In December 1999, IBM announced the development
of a new generation of super computers (Blue Gene), and other major
systems manufacturers will have similar projects in the pipeline. The new
computer generation is said to provide a 500-fold increase in maximum
computing power over the next four to five years. Computation of the
above model would then be possible in real time.
Such are the prospects.
Patient-specific and modelling-aided clinical treatment could, there-
fore, become a reality within the first decade of this millennium!