How data science feeds into nutrition and health

1 Oct 2015

Nora Khaldi and Eileen Gibney tell us about the data that can be derived from food. Photo via science photo/Shutterstock

The food we consume is a treasure trove of data ripe for analysis. We speak to two experts delving deep into these datasets in different ways.

When Eileen Gibney started out researching nutrition, she was surprised at how much data was part of it.

“When you start as a nutritionist you don’t really think that data is going to play a significant role, but everything we do, every recommendation we make, is based on a body of evidence, which will come from data,” she told us at Predict 2015, a data science conference held recently in Dublin.

Nora Khaldi, founder and CSO of Nuritas, echoes this statement.

“What isn’t so obvious to most is that there is a vast amount of data present in everyday foods and, if you find the right way to interpret this data, it can provide incredible and, quite frankly, miraculous insights,” she said.

Data-driven diets

The research being done by Nuritas is fuelled by innovation in data science.

“We use data science to analyse billions of molecules in everyday foods that have the potential to provide an incredible array of health benefits, improving the lives of millions of people around the world,” said Khaldi. “Without the use of data science this will not be feasible.”

Meanwhile, scientists like Gibney, a lecturer in nutrition and genetics at University College Dublin, are constantly looking at association studies, intervention studies and even meta analysis of groups of studies; delving into datasets in an effort to understand how food and food consumption affects health and health outcomes.

“We’re surrounded by data and we use it to derive all of the public health recommendations that you will see given in any country across the world,” she said.

The future of food

Scientists are continuously examining and re-examining how nutrition affects all aspects of our health and lifestyle, and better analysis can provide better recommendations.

“We’re constantly re-evaluating and constantly challenging the ideas that are put out there and trying to find better evidence to support the recommendations that we’re going to give,” said Gibney.

With more data comes a growing body of evidence from which new findings can emerge. When someone like Gibney spots a new association between a nutrient and a health attribute, they then conduct intervention studies to confirm this assumption.

As well as guidance from Gibney and fellow nutrition researchers, our future diets may also include ingredients Nuritas is developing through data science.

Listed among our 20 Irish start-ups mining data science for business opportunities, Nuritas uses artificial intelligence (AI) and machine learning to identify functional ingredients and useful molecules in food sources. These molecules can then be added to food, cosmetics and dietary supplements.

Always innovating in AI

As AI and machine learning are still developing and improving, it will be interesting to see how both of these progress in the coming years and months, and how that can improve upon the work of Nuritas.

“Really, these two fields are influenced by the increasing cleverness of algorithms used and the size of datasets that can be handled,” said Khaldi.

“The frameworks for handling this data will continue to improve, with frameworks like Hadoop already able to handle vast amounts of data.

“For our platform, the machine-learning aspects mean our predictive power grows as we analyse more and more sources. What this equates to is more efficiency in terms of the time and cost of discovery. Our raison d’être is our focus on innovation. We are continually developing our capabilities – always coding! We will definitely look to incorporate new developments as they come.”

Food for thought

Nutrition is not the only area of health where increasing data and analysis can cook up new ideas.

“Health and data are huge,” said Gibney, who predicts an oncoming explosion.

“Nowadays, we’re not looking at health in just one parameter. We all have sensors that we walk around in. We know we can monitor our intake, our physical activity; we can monitor our sleep patterns, our blood pressure, our heart rate. And we can do all of this simultaneously, and we can do it continuously,” she said.

“So, the body of lifestyle, health and food intake data we’re going to get is going to allow us to really data mine these databases with huge power, because up to now we mightn’t have had the amount of data that we require, so we will begin to fine-tune this.”

Public and private impacts

For Gibney, this growing interest in public health presents opportunities for both consumers and industry, and Khaldi sees the same.

“I suppose when you look at data science and its impact, it’s all about the outputs. In the health sector, where we operate, we create a very tangible output from using data science that has the ability to transform human health and wellness.

“In terms of the general public, I feel that the use of data science will have a very significant impact,” she said.

In fact, this impact can already be observed.

“It has allowed access to an ingredient class that really has the ability to benefit wellness and facilitate another very effective preventative approach to chronic diseases,” said Khaldi, who sees this as an improvement on the “too-common reactive approach which is ‘cure it with drugs when you get sick’”.

From a business point of view, there are clear benefits for the health sector.

“When you look at the functional foods market – the fastest-growing sector of the food industry currently – product development is driven by ingredient innovation,” said Khaldi.

“It is data science that is providing access to this innovation, driving growth in this area.”

Siliconrepublic.com’s Data Science Week brings you special coverage of this rapidly growing field from 28 September to 2 October 2015. Don’t miss an entry worth your analysis by subscribing to our news alerts or following @siliconrepublic and the hashtag #DataScienceWeek on Twitter.

Women Invent is Silicon Republic’s campaign to champion the role of women in science, technology, engineering and maths. It has been running since March 2013, and is kindly supported by Intel, Eircom, Fidelity Investments, ESB, Accenture and CoderDojo.

Food science photo by science photo via Shutterstock

Elaine Burke is the host of For Tech’s Sake, a co-production from Silicon Republic and The HeadStuff Podcast Network. She was previously the editor of Silicon Republic.

editorial@siliconrepublic.com