a wave of new implantable sensors that change the way doctors and scientists gather information about the human body. “We are starting to bring to reality some of these more science-fiction ideas about how you can build electronic systems that interface with the body in fundamentally different ways than were possible in the past,” says John Rogers, an electrical engineer at the University of Illinois at Urbana-Champaign.

A Chip in the Ocean

The need for wireless implantable sensors is not just about data scarcity; it also stems in part from the flaws of their wired predecessors. In cases such as deep brain stimulation, in which an implanted electrode is wired to a device on the skull, the wires themselves can present an infection risk; often, they need to be surgically removed, too. This increases both the expense of the procedure and the risk of complications. Existing sensors are not always biocompatible, either, so they can trigger a foreign-body response, effectively inducing the host’s immune system to attack them.

The human body simply is not a
hospitable environment for electron-
ics. “We are basically giant bags of salt
water,” explains electrical engineer Mi-
chel Maharbiz, one of the leaders of the
neural dust project at the University of
California, Berkeley. “If I were to say I’m
going to take a chip and throw it into
the ocean and that it has to operate un-
der the ocean for 25 years, most people
would realize that’s pretty hard.”
An effective, long-term implant-
able sensor therefore has to be strong
enough to resist the corrosive effects of
the environment. It also has to be small
and biocompatible, so it will avoid
interfering with the body’s normal
functioning or provoking an immune
response. These requirements put
strong constraints on the design of the
sensors, the materials used, and more.

Untethered Sensors

At the University of Illinois at Ur-
bana-Champaign, Rogers has led a
multidisciplinary and institutional
team focused on building biocom-
patible sensors for the brain. The
idea stemmed in part from neurosur-
geons working with patients who suf-
fered from severe traumatic brain in-
jury (TBI). “One of the first things the
neuro-surgeon will do is insert sensors
for pressure and temperature because
those two parameters are critically im-
portant,” he says. “If the values fall out-
side a narrow, healthy range, there can
be brain damage.”
Current wired sensors capture the
necessary data, but they need to be
surgically removed, so Rogers and his
group designed a device that can be left
in place and absorbed into the body.
The millimeter-scale device is flexible
rather than rigid to be more biocom-
patible, and can be adapted to sense
a number of variables, including fluid
flow, temperature, and pressure. In the
most recent study in animals, the min-
iature sensor is attached to degradable
wires, but the group also demonstrated
that the sensors would work when com-
bined with an implanted, wireless data
transmitter, eliminating the need for
wires entirely.
The neural dust group has adopted
a different approach to data transmis-
sion. Their package consists of a small
electronic mote, which is surgically im-
planted near a nerve or muscle, and an
external, handheld ultrasound trans-
ducer. The mote is roughly two milli-
meters long and one-half millimeter
wide, and packs a piezoelectric crystal,
a transistor, and two electrodes. There
is no onboard power source. To activate
the mote, the researchers press a trans-
ducer against the skin, which sends
out six quick bursts of ultrasound. The
transducer then switches modes to re-
ceive signals, effectively listening for the
pulses to rebound. When these sound
waves strike the mote, the piezoelec-
tric crystal captures some of the energy,
while some of it bounces back toward
the transducer.
In the absence of neuronal activ-
ity, those rebounded signals will look
roughly the same each time. But if the
nerve being monitored fires during
this process, the two electrodes pick
up that electrical signal. The attached
transistor captures this jolt, which
modulates the current already flow-
ing through the piezoelectric crystal
from the ultra-sound. Once this cur-
rent changes, the amount of energy
reflected back to the handheld trans-
ducer changes as well. After some
computation to filter the received ul-
trasound pulses, the external devices
tease out the strength of the neuronal
signal. “Because you’re listening, you
can back out what’s happening at the
neurons,” Maharbiz says.
Although this version is coated
with epoxy, Maharbiz and the group
are now developing one coated with
a biocompatible thin film that could
function in the body for 10 years with-
out degrading.
Yet another key consideration is to
avoid disrupting function at the site of
the sensor—to protect the body’s sys-
tems, as well as the electronic ones.
Najafi and the team behind the Titan
sensor say this is a major factor, since
they have been testing their device in a
crucial region, the left side of the heart.
Their goal is to measure filling pres-
sure as an indicator of cardiovascular
health, or how well the heart is pump-
ing. In this setting, though, building
something that is biocompatible is
more challenging. “One element of
biocompatibility is the material you
use, but the element that is much
more important is whether your device
is disturbing the blood flow or not.”
The Titan, a cylindrical device with
a pressure-sensing module at one end,
must be surgically implanted, but in a
small human trial of 20 patients, the
device was inserted so that only the
pressure-sensing tip of the cylinder was
exposed to blood flow. The rest of the
device was buried in surrounding tis-
sue, and they went through a number of
design iterations to minimize possible
spots where bacteria could accumulate.
Once implanted, the Titan could
wirelessly trigger alerts through a
“If I think forward ...
I think the amount of
integration between
what we today
consider synthetic
and what we consider
organismal or
biological is going to
be extremely high.”