Some interesting things seem to be happening in the sensor space recently, particularly in area of food scanning and nutrition monitoring. Monitoring food intake is a common feature of many life-logging, exercise tracking, and quantified self apps these days but it has always felt clunky and unsatisfactory. The standard approach is to expect people to simply log what they eat. Some apps take a bit of the guess work out of this by providing users with access to a database of common foods to choose from. Others go a step further by, for example by allowing users to take a photo of their food which is then analysed by experts or crowdsourced calorie counters; check out the Meal Snap app for instance.
Over the last couple of weeks I’ve come across two related product ideas that promise an entirely new approach to this, and one that has some pretty far-reaching implications if the technology stands up to scrutiny. First up is TellSpec, a device that is quickly drawing a crowd on Indiegogo. It’s a handheld scanner that promises to count the calories and assess the composition of plateof food with the simple wave of your hand. It uses raman spectroscopy to analyse the chemical composition of a plate of food. Basically a laser scans your food and measures the light that is reflected back to determine characteristic chemical signatures, while a smartphone app crunches the data to determine the food type and quantities. It will be interesting to see if this works. At all or well enough to have practical value for the diet conscious.
And if you think that sounds like the stuff of science fiction then take a look at AIRO. On the face of it this is a more conventional wearable sensor (a la the Jawbone UP or Nike Fuelband). It promises to track exercise, activity and calorie intake. But rather than using the standard battery of accelerometers and manual input to track activity and calorie intake, ARIO uses a combination of heart-rate sensing and spectroscopy. Instead of waving it over your food AIRO is worn on your wrist and spectroscopically scans your blood to analyse the metabolites that are produced during and after eating to estimate calories and nutritional content.
I must say I find myself to be quite sceptical that either of these devices will work as advertised. However, they are clearly pointing to something quite interesting in the wearables space that goes far beyond simple movement sensing. Perhaps the spectrographic analysis does not need to be pinpoint accurate if it is matched up with common combinations of food for example? Time will tell.