African man looking at mobile phone

Health diagnostic tools for developing countries. (Image:

Dial "D" for Diagnosis

Researchers are developing an inexpensive mobile-phone-based sensor that will allow users in developing countries to rapidly diagnose infectious disease such as malaria or tuberculosis. (7:05)

  • 26 February 2013

The player will show in this paragraph

Find out more about Biomedical Materials and Devices.


Glen Paul: G'day, and welcome to CSIROpod.  I'm Glen Paul.  When we visit the doctor with symptoms of disease infection the diagnosis will usually include the taking of blood or other samples for the culturing of microorganisms, or in some cases DNA based methods can be used.  While we may take this for granted, in the developing world getting access to these types of diagnostic tools is not quite as simple.

With the assistance of the Science and Industry Endowment Fund CSIRO is partnering with scientists from The Nossal Institute for Global Health at the University of Melbourne to establish a proof of concept for using mobile phones and inexpensive sensors as health diagnostic tools for developing countries.

Joining me on the line to discuss the technology is CSIRO's Dr Scott Martin.  Scott, it seems there isn't anything smartphones can't do with the right attachment, but just how will this device work?

Dr Martin: Well that's right, mobile phones can do an incredible amount, and it may surprise you to learn that the rate of ownership of mobile phones in developing countries is actually very high.  We intend to take advantage of that fact by making sensors that can be attached to their mobile phones using what we're calling a smart cable that's being developed by The Nossal Institute for Global Health in order to attach a whole range of different types of sensors which eventually will be able to diagnose a lot of different sorts of diseases or conditions, some quite simple, some more complex, down the track.  And when new sensors are developed for new conditions we can then just tack them onto the smart cable for a different purpose.

Glen Paul: And where are the sensors placed to obtain the best reading, under the tongue, like a thermometer, for example?

Dr Martin: We envisage it being used in a range of different ways.  In fact some are already in existence, and some, like ours, are still in development.  So the ones that we are developing at the moment are based on urine or breath analysis, so at the end of the smart cable we would envisage some sort of device which would either be urinated on, or breathed into – I don't think it would be the same one in each case – and the mobile phone would then receive the data through the smart cable, which would process some of the measurements, in order to give the user a diagnosis of whether they have the particular condition, be it an infectious disease or whatever, and then they would get an immediate result of whether they were suffering from that condition.

Glen Paul: So would it be as simple as a message popping up saying, “You have malaria.  Please consult your doctor.”

Dr Martin: That sort of thing, yes.  But they may not actually have a trained doctor there in the location at the time, so it may actually tell them what they have and then go ahead using technologies that The Nossal Institute already have for mobile phones in order to tailor a regime of treatment for that particular person.  And that might take into account what's available locally, but it also might take into account the size, the sex, and so forth of the person, and tailor the dosage particularly for them.

Glen Paul: It is real 21st Century stuff, a fantastic idea.  Where did the idea for a diagnosing mobile phone come from?

Dr Martin: That's an interesting question.  It's not a conventional approach to measuring disease.  In the conventional ways you've mentioned before we go looking for a specific indicator, so if we're looking for an infectious disease which is generated by a bacterium, the conventional way is to go looking for the bacterium itself, and as you can imagine these things are very small, and although there are lots of them in the body, they're in very, very low concentrations, which makes detection very, very challenging indeed.

What we are looking for in this approach is the host response, so it's the body's response to the bacterium which can be quite large, which we are measuring through fingerprinting of urine or breath, in order to give a measurement and an indication of whether the person is suffering from a particular condition.

Glen Paul: Hmm.  And what range of diseases then will it be capable of diagnosing?

Dr Martin: Again that's another interesting question.  We know of people around the world who are using similar approaches and they claim, for example a group in Israel, to be able through a very simple measurement of breath to determine whether you're suffering from cancer or not, and this is not something that you would immediately think you could get a sense of from something as simple as somebody's breath.

And again, although it sounds fantastic, there have been quite a lot of credible reports in the literature that indicate that people's dogs are able to smell cancer in their owners, so it's that sort of approach that we're taking here.

Glen Paul: Man's best friend is a good approach.  And what about testing the device, would you say hang out in the waiting room of the local GP, for example?

Dr Martin: That sort of approach, yes, although we have to be a lot more diligent about it than just hanging around.  Obviously we have to consider the safety of our own workers.  We are looking at the moment at infectious disease, and so the first thing to do is to make sure that all of our Researchers are safe.  We don't ourselves hang around for samples, but we do have arrangements with two hospitals in Melbourne, we have employed nurses at the two emergency departments in Melbourne, and those fully trained nurses recruit people who volunteer to supply both breath and urine samples to us.  And we're interested in two particular cohorts of people – those people who come in with a clear infectious disease, and then a controlled sample of people who come in with no signs of disease, so they might come in with a sports injury or something like that – and we can use those two groups to work out whether our sensors can tell the difference between the two types of people, those with an infectious disease and those without.

Glen Paul: Brilliant.  And how long do you think it might be before we can download the App from iTunes and plug the attachments in?

Dr Martin: Oh, that's a very good question.  There's a lot of work to do just to get the technology to a point of proof of concept, which we're undergoing at the moment, and then we need to develop it into a fully fledged medical device which could then be sold.  This typically takes a number of years, and I would be quite nervous about guessing a number of years on that, but I would say at least five years.

Glen Paul: And of course smartphone technology will develop alongside of that as well.  As I say, fantastic stuff.  Thank you very much for discussing the technology with me today, Scott.

Dr Martin: Thank you very much.

Glen Paul: Dr Scott Martin.  And to find out more about the research, or follow us on other social media, visit