The food of the future: algorithms define what to eat.

After 20 years of living with type 2 diabetes, Tom Idema had given up hope of controlling his disease. He had tried many diets that were unsuccessful and even considered weight loss surgery. When his employer offered him the chance to try a new diet app that uses artificial intelligence to monitor blood sugar, he took it.

Mr. Idema, 50, submitted a stool sample for sequencing of his microbiome and filled out an online questionnaire with his blood sugar level, height, weight and medical conditions. That data was used to create a profile for him, to which he added continuous blood sugar measurements for a couple of weeks. After that, the app, called DayTwo, rated different foods on how good or bad they might be for Mr. Idema’s blood sugar, to help him make better food choices.

After nearly 500 days using the program, her diabetes is in remission and her blood sugar levels have dropped to the upper limit of normal. And while DayTwo says the app isn’t targeting weight loss, it has gone from 300 to 225 pounds. “I wear pants sizes I haven’t worn since high school,” said Mr. Idema, who is an administrator at Central Michigan University in Mount Pleasant, Michigan.

DayTwo is just one of a number of apps claiming to offer AI power solutions. Instead of a traditional diet, which often has a fixed list of “good” and “bad” foods, these programs work like personal assistants that help someone quickly make healthy food choices. They are based on research showing that bodies react differently to the same foods, and the healthiest choices are likely to be unique to each individual.

It’s not yet clear if these AI nutritionists are ready for widespread use, and there is very little research available from sources outside the companies that sell apps. Users should beware of overly broad claims that go beyond predicting how foods affect blood sugar.

But advocates say that blood sugar is just the beginning and that artificial intelligence programs could target other aspects of metabolic health, such as obesity and heart diseaseand eventually help guide a person’s daily food choices.

How to make (artificially) smart food choices

The application DayTwo uses an algorithm based on the research of Eran Elinav and Eran Segal of the Weizmann Institute of Science in Israel.who co-founded the company in 2015. Last year, the company found that when they used their algorithm to match a diet to an individual’s microbiome and metabolism, it was better at controlling blood sugar than the Mediterranean diet, considered a of the healthiest in the world.

“Instead of measuring foods by their calorie content and trying to come up with a ‘healthy diet,’” said Dr. Elinav, “You must begin to measure the individual”.

This technology is relatively new and is only related to blood sugar. Meanwhile, the Mediterranean diet has decades of research behind it and will likely remain the gold standard for healthy eating for years to come. Still, for people like Mr. Idema, AI like DayTwo’s can make it easier to maintain healthy eating patterns.

The Mediterranean diet, which combines carbohydrates with legumes and vegetables, is considered one of the healthiest
The Mediterranean diet, which combines carbohydrates with legumes and vegetables, is considered one of the healthiestArchive

The app’s machine learning algorithm can identify patterns and learn from the data with human help. Analyze data on blood sugar responses from different individuals to tens of thousands of different foods to identify personal characteristics (age, gender, weight, microbiome profile, and various metabolic measurements) that explain Why does one person’s glucose rise with certain foods when another person doesn’t?. The algorithm uses these observations to predict how a particular food will affect blood glucose and assign each food a score.

The system still can’t take into account the chocolate bar someone ate two hours ago, but users can play around with food combinations to change the score for each meal. For example, the app gave cheese noodles, one of Mr. Idema’s favorites, a low score, but was able to improve it by adding protein. This is because adding protein or healthy fats can blunt the blood sugar spike from a high-carb meal like pasta.

“I thought they were going to say, ‘Oh my gosh, you just have to eat salads, and that hasn’t been the case.'”Mr. Idema said.

DayTwo, which is currently only available to businesses or health plans, not consumers, is one of the few AI-powered apps that recommend healthier meal options. Another company, ZOE, also generates meal scores and is available directly to consumers for $59 a month.

ZOE’s algorithm uses additional data, such as blood fat levels, in addition to microbiome and blood sugar tests. Algorithm was able to predict how a person’s blood sugar and fats respond to different foods in a large 2020 study led by one of the company’s founders, Dr. Tim Spector, professor of genetic epidemiology at King’s College London.

Currently, these algorithms focus primarily on blood sugar, but newer versions will incorporate more personal data and, in theory, recommend diets that lower cholesterol, blood pressure, resting heart rate, or any other measurable clinical indicator. .

“Incorporating all these different types of data is very powerful, and that’s where machine learning comes in,” said Dr. Michael Snyder, a professor of genetics at Stanford University who helped found the health company.

Buyer beware

The field of personalized nutrition is still in its Wild West phase, and experts say it’s important to get around the hype. Many companies are willing to assess your microbiome and offer AI-based dietary recommendations and sell you supplements, but few are based on scientifically rigorous trials. Last year, uBiome, which made one, was even accused of fraud. In general, the broader the health and weight loss claims that companies make, the less reliable the evidence to support them.

“I think everything is overrated right now, unfortunately,” said Dr. Eric Topol, a cardiologist and founder and director of the Scripps Research Translational Institute.

The data used by apps like DayTwo and ZOE also capture only a fraction of the interaction between the gut microbiome, our metabolism, and diet. Indeed there are many more factors, such as genetics, that affect metabolism and are ignored by current AI programs.

“It doesn’t tell you the whole story, and optimizing glucose alone will not be enough to create the perfect diet for you.i,” said Dr. Casey Means, co-founder and chief medical officer of a digital health company called Levels. AI apps could push users to eat foods that are good for preventing blood sugar spikes and diabetes, but may not be healthy in other ways.

For example, when Dr. Topol tested the DayTwo app, his recommendations to control your blood sugar, such as eating spinach and raspberries, were high in oxalic acid, which could have induced kidney stones. That’s because the app didn’t take into account your pre-existing risk for the condition.

Also, restrictive diets are increasingly seen as a bad way to change eating habits and often backfire. But many experts hope that custom AI applications will be easier to follow and build better behaviors in the long run.

For now, these apps could help nutritionists with meal suggestionsbut they won’t replace them, and both ZOE and DayTwo have regular virtual check-ins with a dietitian or nutritionist built into their programs.

According to Dr. Topol, larger, longer-term studies that incorporate more layers of data, such as sleep, exercise, or stress, into the algorithms could make these programs more precise and accurate for each individual. They could also help people see how short-term responses, such as glucose spikes after meals, influence long-term health.

What we don’t know is how or if daily improvements translate into long-term health.. Dr. Topol clarified about AI diet programs. “Can diabetes be prevented? Can heart disease and other chronic diseases be prevented?

These larger studios are coming. The Nutrition for Precision Health research program at the National Institutes of Health began a multiyear study in January to develop algorithms to predict individual responses to foods.

But for Mr. Idema, the effects of personalized diets are already tangible, most recently when his improved blood sugar levels allowed him to enjoy his daughter’s birthday cake. “I had the glucose monitor off at the time and stayed in range, so my body handled it well”, he claimed. “So now I’m in a better place, and in my opinion, this show definitely saved my life.”