Physiome is so big and so important that it needs an underlying open source framework

Submitted by Norm Roulet on Fri, 02/11/2005 - 07:21.

The Doctor Will See Your Prototype Now - Wired,
February 2005

The Physiome Project is assembling digital models of every system and anatomical feature of the human body - from large organs to tiny cellular and molecular functions.
The system would allows physicians to test various scenarios on your digital model - surgery, radiation, chemotherapy - and watch how your system reacts.

To Peter Hunter, the future of medicine looks like
this: You visit your doctor after weeks of feeling fatigued and
lethargic. She takes a blood sample, records your DNA profile, does a
quick CT body scan, then uploads the raw data to a workstation. Within
minutes, software stitches together a head-to-toe living, breathing
digital reproduction of your innards, which the doc can poke and prod
just like the real thing. Turns out you have lung cancer. Rather than
focusing on one treatment, your physician can test various scenarios on
your digital doppelgänger - surgery, radiation, chemotherapy - and
watch how your system reacts. The cure is the simulation that doesn't
kill the virtual you.

Hunter, director of the Bioengineering Institute in Auckland, New
Zealand, is an expert in biomechanics and in computational physiology,
an emerging field. He admits that his vision for health care might be a
decade or two away, but it's by no means science fiction. Bioengineers
in the institute's Physiome Project are assembling digital models of
every system and anatomical feature of the human body - from large
organs to tiny cellular and molecular functions.

Hunter and his colleagues have already finished a draft of the
skeletal system, and they recently built the first-ever digital human
heart and lungs. The lungs - with 300 million alveoli - inhale and
exhale just like flesh-and-blood ones. Meanwhile, work is under way on
a replica of the digestive system and a comprehensive database of
cellular functions. Other system models - nervous, endocrine, immune,
sensory, skin, kidney-urinary, reproductive - are coming.

The process for creating the models is surprisingly literal, even
crude. Hunter's team built a contraption for slicing off razor-thin
layers of cadaver flesh - heart and lung tissue, for instance - which
can be scrutinized under a microscope or fed through a digital scanner.
Another device, called a multiaxial rig, looks like a roulette wheel
made from an Erector set. It's used to assess the tensile strength of
human skin and other soft tissue.

The scientists in Auckland have plenty of help. The Physiome Project
is a worldwide initiative, akin to the Human Genome Project, with a
dozen research teams participating at labs across the US and in Israel,
Japan, and the UK. Hunter compares his project to the evolution of
Linux: "Physiome is so big and so important that it needs an underlying
open source framework."

Pharmaceutical companies are hovering over the research. Hunter says
he's met with execs from Aventis and Novartis. And no wonder: Once
perfected, the technology would allow drugmakers to develop and test
the effectiveness of medicines before forking out billions of dollars
for risky clinical trials. Physiome would also enable medical engineers
to fashion customized implants, such as pacemakers or artificial heart
valves. And it would let surgeons make dry runs on a digital replica of
their patient - determining which technique works best before ever
picking up a knife. "Today practically everything we see and touch was
prototyped by computers," says Randy Haluck, director of minimally
invasive surgery and surgical simulation at Penn State Hershey Medical
Center. "One glaring exception is in medicine - but there's no reason
to expect that the technology won't move there."

Hunter acknowledges that the Physiome Project is a long way from
fulfilling its potential. It needs more complexity on the molecular
level before it can simulate all the effects of an experimental drug or
a surgical procedure. "The biggest challenge," he says, "is being able
to understand gene regulation."And certain parts of the human body -
the brain, the immune system - are still very much a mystery. But after
all, that's what computers and Erector sets are for.

Michael Behar (michael [at] michaelbehar [dot] com) is a freelance writer in Washington, DC.