Artificial intelligence is used to help follow diets

After 20 years of living with type 2 diabetes, Tom Idema had given up hope of controlling the disease. He tried many diets with no success and even thought about having weight reduction surgery. When his employer offered him the opportunity to try out a new diet app that uses artificial intelligence to control blood sugar, he accepted.

Idema, 50, sent a stool sample to have her microbiome sequenced and filled out an online questionnaire with her blood sugar, height, weight and medical conditions. This data was used to create a profile of him, to which she added continuous blood sugar measurements for a few weeks.

After that, the app, called DayTwo, ranked different foods according to whether they were good or bad for Idema’s blood sugar, to help him make better food choices.

After nearly 500 days of using the program, her diabetes is in remission and her blood sugar levels have dropped to the upper limit of normal. And despite the DayTwo app saying it doesn’t aim to reduce weight, Idema went from 145 kg to 103 kg.

“I’m wearing pants in sizes I haven’t worn since high school,” said Idema, who is a trustee at Central Michigan University in Mount Pleasant.

DayTwo is just one of a series of apps claiming to offer AI power solutions.

Instead of a traditional diet, which often has a set list of “good” or “bad” foods, these programs are more like personal assistants that help a person make healthy food choices quickly. 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.

How to make (artificially) smart food choices

The DayTwo app uses an algorithm based on research done by Eran Elinav and Eran Segal of the Weizmann Institute of Science in Israel, who founded the company in 2015. Last year, the company discovered that when they used their algorithm to find a diet compatible with an individual’s microbiome and metabolism, it was better for blood sugar control than the Mediterranean diet, considered one of the healthiest in the world.

“Instead of measuring foods by calorie content and trying to create a ‘healthy diet,’ you need to start measuring the individual,” Elinav said.

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

The app’s algorithm can identify patterns and learn from the data, with human help. It analyzes data from different individuals’ blood sugar responses to tens of thousands of different meals to identify personal characteristics — age, sex, weight, microbiome profile and various metabolic measures — that explain why a person’s glucose rises with age. certain foods and someone else’s is not.

The algorithm uses this analysis to predict how a particular food will affect blood sugar and assign a score to each meal.

DayTwo, currently only available to employers or health plans, not consumers, is one of several AI-powered apps that recommend healthier meals.

Another company, ZOE, also generates meal notes and is available directly to consumers for $59 a month. ZOE’s algorithm uses additional data such as blood fat levels as well as microbiome and blood sugar tests.

The 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, Doctor Tim Spector, professor of genetic epidemiology at King’s College, London.

buyer beware

The field of personalized nutrition is still in the pioneering phase, and experts say it’s important to avoid fads. Many companies offer to test microbiomes and offer AI-driven dietary recommendations — in addition to selling supplements — but few are based on scientifically rigorous testing.

Last year, uBiome, which made one, was even accused of fraud. In general, the more comprehensive the health and weight loss claims made by companies, the less reliable the evidence in support of them.

“I think it’s all too hyped up now, unfortunately,” said Dr. Eric Topol, cardiologist and founder and director of the Scripps Institute for Translational Research.

The data used by apps like DayTwo and ZOE also only captures a fraction of the interaction between the gut microbiome, our metabolism and diet. There are certainly many other factors, including genetics, that affect metabolism that are ignored by current AI programs.

“That doesn’t tell the whole story, and just optimizing around glucose won’t be enough to create the perfect diet for you,” said Dr. Casey Means, co-founder and chief medical officer of digital health company Levels. AI apps can lead users to eat foods that are good for avoiding blood sugar spikes and diabetes, but can be harmful in other ways.

For example, when Topol tried the DayTwo app, its recommendations for controlling 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 this condition.

Also, restrictive diets are increasingly considered a bad way to change eating habits, and often backfire. But many experts expect custom AI apps to be easier to follow and create better long-term behavior.

For Idema, the effects of personalized diets are already tangible, most recently when her improved blood sugar levels allowed her to taste her daughter’s birthday cake.

“I was on the glucose monitor at the time and I was well within range, so my body handled it well,” he said. “I’m in a much, much better situation now, and in my mind, this show definitely saved my life.”

^{Translated by Luiz Roberto M. Gonçalves}