Supercomputers: powering medical innovation to enable the
industry to achieve more
Karen Padmore, Operations Director at High
Performance Computing (HPC) Wales, explains how supercomputing can
be used to help speed-up breakthroughs in medical research.
11 June 2014
Ask any doctor how they would cope without a stethoscope
and there’s a good chance the general consensus would be ‘with
difficulty’. Now used by pretty much every medical speciality, this
versatile tool has revolutionised the way doctors operate, enhancing
their ability to detect, analyse and respond to issues early.
The medical industry has progressed a long way since the
stethoscope’s introduction almost 200 years ago — from the discovery
of the molecular structure of DNA to antibiotics, insulin,
vaccinations and countless ingenious medical devices.
Supercomputers are now powering the next generation of medical
innovation. The advanced technology is already being used across a
range of industries to address many of the challenges facing modern
society; whether it is predicting weather patterns and consumer
trends, mitigating the effects of climate change or, in the case of
the medical industry, developing new drugs and treatments to address
a wide range of debilitating diseases and conditions. The scale and
complexity of these problems mean that traditional problem-solving
techniques are both costly and time-consuming.
Even in an age of unprecedented technological advancement,
medical researchers working on cures for heart disease, cancer,
obesity and many other human ailments, still perform basic trial and
error experiments in laboratories. This approach yields
excruciatingly slow results. Supercomputers have the ability to
speed up these processes considerably.
So what is the definition of a supercomputer?
Also known as high performance computers, supercomputers can be
thought of as large collections of individual computers connected
together, working in parallel on a single problem and are capable of
performing complex and high-volume calculations and simulations at
top speeds, reducing analysis times from weeks to days, or even
HPC Wales’ supercomputing network can run 320 trillion operations
per second. To give you an idea of just how fast this is, imagine
the following: All seven billion people in the world have a
calculator and are asked to perform one calculation per second
twenty-four hours a day, nonstop. It would take the world’s
population approximately thirteen and a half hours to do what our
supercomputing network can do in just one second. If we think of a
regular computer travelling as fast as a snail, a supercomputer is a
The incredible processing speed of a supercomputer enables the
handling of ever-increasing medical data sets and the building of
predictive models of disease. The University of South Wales’
Genomics and Computational Biology researchers are using
supercomputing to understand bacterial evolution to help transform
the current state of antibiotic treatments.
With the recent rise in drug resistant infections, such as MRSA
and tuberculosis, many scientists and doctors are now concerned
about a potential antibiotic crisis. This research is aiming to
predict drug resistance and help GPs to select the most appropriate
antibiotics for their patients.
Researchers are using supercomputers to understand how generally
harmless bacterial strains can evolve into toxic strains, such as
E Coli 0157. The technology has enabled the researchers to
analyse vast amounts of data at incredible speeds, in this case
reducing the data analytics time from days to hours.
Farzana Rahman from the University of South
Wales, who is spearheading the research into the next-generation
Projects like this demonstrate the revolution occurring in
laboratory-based sciences and computing technologies. Researchers
can process enormous volumes of diverse data very efficiently. At
the same time, new experimental techniques are being developed.
Integrating these two sets of advances provides access to an
unprecedented amount of information about biological processes.
Experts predict that in the very near future, when every person
is genotyped, the notion of personalised medicine will become a
reality. Healthcare decisions and practices will be tailored to
individual patients by integrated use of genotypic and phenotypic
information as well as medical and family history.
However, access to the amount of newly available biological
information creates its own challenges – namely managing and
processing that information, and producing results that are
medically actionable. Overcoming these issues requires the use of
high performance computing systems.
Supercomputers have many applications that are equally invaluable
to medical appliance manufacturers, including advanced modeling and
simulation and the ability to render high-definition 3D graphics.
This means new designs can be tested with different parameters,
evaluated and refined without the need for physical prototypes. This
reduces the time, labour and cost involved in bringing products to
market, while improving research and development capabilities.
Advanced modeling and simulation enables these providers to
innovate and improve existing products and services, giving them a
competitive edge through more efficient product design and improved
Modelling a novel blood pump
Swansea-based Calon Cardio-Technology Ltd is a great example of
how supercomputing technology has been used to aid the development
of the next generation of affordable, implantable micro blood pumps
for the treatment of chronic heart failure. Ventricular Assist
Devices (VADs) have the potential to save many thousands of lives as
they provide an alternative for heart transplants.
While they are a viable treatment for heart failure, the current
generation of VADs are large, requiring highly invasive surgery
which is very costly and causes long term damage to blood that can
cause further health problems to the patient.
Calon Cardio is addressing these problems through the development
of a much smaller and cheaper VAD that can be directly implanted
into the heart, minimising the damage to the blood, offering an
extended life expectancy and helping to ensure an improved quality
of life for the patient.
When simulating the flow of blood inside the VAD, using a
standard desktop computer results in a very long cycle time for each
simulation. Typically, running just one case could take up to a
week, whereas with a supercomputer the process can be reduced to
less than a day — even a few hours. Results are generated at an
unprecedented speed, ensuring that the design process is as
efficient as possible.
Illustration of the Calon Cardio ventricular
and how it would be used in a patient
Until recently, the high costs associated with supercomputing has
meant that the technology has only been within reach of larger
institutions and corporations. However, that is changing fast. Not
only are costs for using supercomputers falling but in the UK there
are a number of associated grants available, both home-grown and
from the EU.
Businesses now do not have to invest in their own hardware and
software to benefit from supercomputing technology, with some
supercomputing providers now offering remote access to the
technology via a simplified web browser interface along with the
training and support required to exploit it effectively. Businesses
and academics now no longer require specialist HPC knowledge to
benefit from its power.
The message for medical researchers and product designers it that
supercomputing can transform research and modelling work, and there
are a number of funding options out there. The examples outlined
above provide only a snapshot of how supercomputers are powering
potentially life-saving work, both by businesses and academics. This
is why, I believe, it will not be long before the medical industry
finds HPC technology as valuable as the trusty stethoscope.
See a video of Farzana Rahman, University of South Wales’
Genomics and Computational Biology researcher, talking about her
inspiration to study antibiotics:
High Performance Computing (HPC) Wales: