Antonio Tricoli and Noushin Nasri test their mobile sensor. Photo by Stuart Hay/ANU.

Smart phones are getting smarter by the minute. They already help us plan transport routes, check the weather and take pictures. What if they could also diagnose diabetes or lung cancer?

Dr Antonio Tricoli and his team in the College of Engineering and Computer Science are developing a technology to detect diseases by measuring markers in your breath – and that's small enough to fit in a mobile phone.

The implication for cheap, mobile, non-invasive diagnostics is potentially revolutionary. 

"If you are able to detect lung cancer at a very early stage the chances of recovering are much higher," says Tricoli.

"This solid-state technology is small enough to fit in a phone so it could be for personal use, or a bigger device could be used in pharmacies."

Working at the interface of biology and engineering, Tricoli and his team are using the National Computational Infrastructure's supercomputer to design a mobile sensor made of billions of interconnected nanoparticles.

"Your breath actually contains more than 10,000 compounds, which can tell us about what you've been eating, how much, and whether you might have an undiagnosed medical condition.

"For example, we know that acetone in your breath can mean diabetes. And the difference between healthy and diabetic patients is only about one particle per million.

"The question is: how do you make something which is sufficiently sensitive to detect this tiny concentration, but also selective enough that we can exclude the 9,999 other compounds you breathe out?"

The answer involves extremely sensitive ceramic nanoparticles, made up of as few as 10 atoms, forged at 2,400⁰C.

"By controlling what nucleates first and what other components are in the mix, we can tailor the specific crystal properties of the nanoparticles so they absorb a particular compound," explains Tricoli.

"We then make an aerosol cloud of nanoparticles, which we deposit on a substrate for self-assembly into a film or layer."

The adsorption of the target molecule on the nanoparticle's surface changes its electronic properties, causing a measurable shift in electrical resistance.

The smaller the semiconductor, the more sensitive its electronic properties, allowing detection of very small concentrations of the target molecule.

Optimising the design of the nanostructures is very challenging in the laboratory, says Tricoli. Simulating the process using computers is much quicker and more cost-effective.

"We use a computer simulation code where we can incorporate the size of the substrate, the size of the nanoparticles, and their velocity. We need a supercomputer to do this because the code calculates each step of the particles' movement, and we have millions and millions of particles to deposit for each simulation," says Tricoli.

PhD candidate Noushin Nasiri has spent many, many hours exploring simulation outputs, running more than 300 simulations – each composed of one million nanoparticle placements – in the past year. Each simulation takes up to one week to run.

Once the simulations have been performed, the team use the data to create and refine actual sensors in the laboratory. The project is supported by an ARC Discovery grant.

"Detection of lung cancer is an important goal but not the only disease that this technology will aim," says Tricoli. "There are many other diseases which would benefit from early-stage detection and a portable, low-cost diagnostics method could save many lives and also money."

Tricoli has already submitted a patent for this technology and has worked with Swiss company, Sensirion, to make a device just 300 micrometres wide, small enough to fit inside a mobile phone.

"It's not far off," he says.