TAKING IT PERSONAL
Today, 422 million people are living with diabetes and 2 billion people are overweight. That’s not news. But for the first time in human history, the lancet showed that there are more overweight than underweight adults on this planet.
If any of you has ever been on a diet, or trained for a marathon, you probably noticed that it’s impossible to know what nutrition advice to follow because the advice goes something like this: Eat carbs! No, wait – don’t eat carbs! Stop eating fat! Eat all the fat! It’s a $400 billion industry and it doesn’t work.
Building the FoodPrint
We each have a unique fingerprint, and we also have our own FoodPrint ™. FoodPrint is a digital signature of how our body reacts to different foods. It is contextually driven and provides correlations, insights and predictions that become the underpinning for personal and continually improving nutrition recommendations.
An individual’s FoodPrint is based on two buckets of information: data about food and data about individuals.
On the food side, we are building the world’s largest nutritional database. Nutrino’s food analysis system uses techniques such as machine learning and natural language processing. We understand nutritional recommendations and who they might be relevant for. We also understand foods: elementary foods like an apple and packaged foods like crackers. We have analyzed recipes and prepared foods and restaurant menus to understand the nutritional breakdown and parse the micronutrients inside.
At the same time, we have built tools that allow our platform to quickly become smart about individuals. We collect individual inputs from users like activity, stress, sleep, and we also go pretty deep with medical data like continuous glucose and insulin, and even DNA. We aspire to take into account everything that is either affected by nutrition or may affect metabolism.
When we bring together these two vast buckets of data, we are able to quickly make sense of them and how they work together, in real time. We use predictive analytics and optimization theory to provide personalized and contextual recommendations for users and larger trends and insights about populations.
Today and the future
People and companies use FoodPrint in different ways. Some examples including partnering with IBM Watson to develop nutritional recommendations for pregnant women. We work with Welltok to support the success of their corporate wellness programs. And, over the past two years we’ve developed deep expertise in using our platform to improve diabetes management through understanding a patient’s Glucose FoodPrint. We’ve developed partnerships with Medtronic, Abbott and Dexcom to improve the usage of their devices and the success of their patients.
Next, we’ll be applying FoodPrint to other diseases like hypertension. We are going to bring FoodPrint to other industries like the supplements industry, and consumers will be able to hone in on the supplements that will actually work for them. We’ll serve athletes and military personnel with food that increases performance. We’ll enable smart kitchens that will trigger insights for consumers, based on their personal needs.
To get there, we continue to collect online and offline data, organizing and analyzing it into a platform, growing our food ontology and adding layers of metadata, adding rules, classes, attributes and relations of food. The scalable nature of the platform and the machine learning that supports it means that the more data we feed it, the better and more precise it becomes.
Knowing your own FoodPrint, knowing people’s FoodPrints will completely change the way food decisions are made every day. And when that happens, we’ll be living in a different world.
