SPIT ON A STICK:
People with diabetes may have to prick their fingers several times a day to read off blood sugar levels. Now researchers in Iran have created a monitor to do the same reading from saliva. The glucose in saliva is 100 to 2000 times lower than in blood, which has been a problem until now.
uses an extremely sensitive technique with a more sensitive device. The researchers expect to patent the device and release it commercially by the end of April 2013. Stick a needle in your finger or spit on a stick? It'd be an
TEMPLE TOUCH: The Scanadu SCOUT is a tiny hardware device. Touch it to the temple and a few seconds later it reports pulse transit time, heart rate, electrical heart activity, temperature, heart rate variability and blood oxygenation to an app on your smartphone. The app then offers advice and guidance, such as suggesting a trip to the doctor or resting for 24 hours. The measurements are helpful, but the
advice is the gem.
BLUE ROADS: Calgary in Canada is trying to help drivers on one particular stretch of road, the Deerfoot Trail. Overhead displays show road congestion and estimated travel time to the next point. The system gathers its data from the MAC addresses of Bluetooth signals in cars using the road, while filtering out signals from pedestrians. 15 sensors and 7 permanent message boards gather and display the data. So it could be a public service to keep Bluetooth switched on, even if you're not using it.
NOT ANONYMOUS: Businesses in Newbury in the UK are trying to catch shoplifters as well as capturing just their image on security cameras.
Shopkeepers upload images from their in-store security cams to the Police and file a report. The Police then set up a gallery of images and ask the public to help identify the faces. The Facewatch system seems to be working in Newbury, with a drop in reported shoplifting. Crowdsourcing crime prevention takes another step.
EYES RIGHT: There's iris recognition, but then there's also recognising the pattern of veins in the whites of your eyes. EyeVerify hope their eye recognition might be used for secure access to online medical records or bank accounts. Gaze into your smartphone's high-res camera and look to the side to reveal the whites of your eyes. The captured image is compared to a stored eyeprint and access is granted if they match. The app can distinguish between a real eye and an image of an eye. In a small-scale test the app was 99.97% accurate. Which is pretty good, but it still means a lot of people locked out of their records.