AI and you: Is the promise of personalized nutrition apps worth the hype?
As a type 2 diabetic, Michael Snyder has long been interested in how blood sugar levels vary from one person to another in response to the same food, and whether a more personalized approach to nutrition could help tackle the rapidly cascading levels of diabetes and obesity in much of the western world.
Eight years ago, Snyder, who directs the Center for Genomics and Personalized Medicine at Stanford University, decided to put his theories to the test. In the 2000s continuous glucose monitoring, or CGM, had begun to revolutionize the lives of diabetics, both type 1 and type 2. Using spherical sensors which sit on the upper arm or abdomen – with tiny wires that pierce the skin – the technology allowed patients to gain real-time updates on their blood sugar levels, transmitted directly to their phone.
It gave Snyder an idea for his research at Stanford. Applying the same technology to a group of apparently healthy people, and looking for ‘spikes’ or sudden surges in blood sugar known as hyperglycemia, could provide a means of observing how their bodies reacted to an array of foods.
“We discovered that different foods spike people differently,” he says. “Some people spike to pasta, others to bread, others to bananas, and so on. It’s very personalized and our feeling was that building programs around these devices could be extremely powerful for better managing people’s glucose.”
Unbeknown to Snyder at the time, thousands of miles away, a group of Israeli scientists at the Weizmann Institute of Science were doing exactly the same experiments. In 2015, they published a landmark paper which used CGM to track the blood sugar levels of 800 people over several days, showing that the biological response to identical foods can vary wildly. Like Snyder, they theorized that giving people a greater understanding of their own glucose responses, so they spend more time in the normal range, may reduce the prevalence of type 2 diabetes.
The commercial potential of such apps is clear, but the underlying science continues to generate intriguing findings.
“At the moment 33 percent of the U.S. population is pre-diabetic, and 70 percent of those pre-diabetics will become diabetic,” says Snyder. “Those numbers are going up, so it’s pretty clear we need to do something about it.”
Fast forward to 2022,and both teams have converted their ideas into subscription-based dietary apps which use artificial intelligence to offer data-informed nutritional and lifestyle recommendations. Snyder’s spinoff, January AI, combines CGM information with heart rate, sleep, and activity data to advise on foods to avoid and the best times to exercise. DayTwo–a start-up which utilizes the findings of Weizmann Institute of Science–obtains microbiome information by sequencing stool samples, and combines this with blood glucose data to rate ‘good’ and ‘bad’ foods for a particular person.
“CGMs can be used to devise personalized diets,” says Eran Elinav, an immunology professor and microbiota researcher at the Weizmann Institute of Science in addition to serving as a scientific consultant for DayTwo. “However, this process can be cumbersome. Therefore, in our lab we created an algorithm, based on data acquired from a big cohort of people, which can accurately predict post-meal glucose responses on a personal basis.”
The commercial potential of such apps is clear. DayTwo, who market their product to corporate employers and health insurers rather than individual consumers, recently raised $37 million in funding. But the underlying science continues to generate intriguing findings.
Last year, Elinav and colleagues published a study on 225 individuals with pre-diabetes which found that they achieved better blood sugar control when they followed a personalized diet based on DayTwo’s recommendations, compared to a Mediterranean diet. The journal Cell just released a new paper from Snyder’s group which shows that different types of fibre benefit people in different ways.
“The idea is you hear different fibres are good for you,” says Snyder. “But if you look at fibres they’re all over the map—it’s like saying all animals are the same. The responses are very individual. For a lot of people [a type of fibre called] arabinoxylan clearly reduced cholesterol while the fibre inulin had no effect. But in some people, it was the complete opposite.”
Eight years ago, Stanford's Michael Snyder began studying how continuous glucose monitors could be used by patients to gain real-time updates on their blood sugar levels, transmitted directly to their phone.
The Snyder Lab, Stanford Medicine
Because of studies like these, interest in precision nutrition approaches has exploded in recent years. In January, the National Institutes of Health announced that they are spending $170 million on a five year, multi-center initiative which aims to develop algorithms based on a whole range of data sources from blood sugar to sleep, exercise, stress, microbiome and even genomic information which can help predict which diets are most suitable for a particular individual.
“There's so many different factors which influence what you put into your mouth but also what happens to different types of nutrients and how that ultimately affects your health, which means you can’t have a one-size-fits-all set of nutritional guidelines for everyone,” says Bruce Y. Lee, professor of health policy and management at the City University of New York Graduate School of Public Health.
With the falling costs of genomic sequencing, other precision nutrition clinical trials are choosing to look at whether our genomes alone can yield key information about what our diets should look like, an emerging field of research known as nutrigenomics.
The ASPIRE-DNA clinical trial at Imperial College London is aiming to see whether particular genetic variants can be used to classify individuals into two groups, those who are more glucose sensitive to fat and those who are more sensitive to carbohydrates. By following a tailored diet based on these sensitivities, the trial aims to see whether it can prevent people with pre-diabetes from developing the disease.
But while much hope is riding on these trials, even precision nutrition advocates caution that the field remains in the very earliest of stages. Lars-Oliver Klotz, professor of nutrigenomics at Friedrich-Schiller-University in Jena, Germany, says that while the overall goal is to identify means of avoiding nutrition-related diseases, genomic data alone is unlikely to be sufficient to prevent obesity and type 2 diabetes.
“Genome data is rather simple to acquire these days as sequencing techniques have dramatically advanced in recent years,” he says. “However, the predictive value of just genome sequencing is too low in the case of obesity and prediabetes.”
Others say that while genomic data can yield useful information in terms of how different people metabolize different types of fat and specific nutrients such as B vitamins, there is a need for more research before it can be utilized in an algorithm for making dietary recommendations.
“I think it’s a little early,” says Eileen Gibney, a professor at University College Dublin. “We’ve identified a limited number of gene-nutrient interactions so far, but we need more randomized control trials of people with different genetic profiles on the same diet, to see whether they respond differently, and if that can be explained by their genetic differences.”
Some start-ups have already come unstuck for promising too much, or pushing recommendations which are not based on scientifically rigorous trials. The world of precision nutrition apps was dubbed a ‘Wild West’ by some commentators after the founders of uBiome – a start-up which offered nutritional recommendations based on information obtained from sequencing stool samples –were charged with fraud last year. The weight-loss app Noom, which was valued at $3.7 billion in May 2021, has been criticized on Twitter by a number of users who claimed that its recommendations have led to them developed eating disorders.
With precision nutrition apps marketing their technology at healthy individuals, question marks have also been raised about the value which can be gained through non-diabetics monitoring their blood sugar through CGM. While some small studies have found that wearing a CGM can make overweight or obese individuals more motivated to exercise, there is still a lack of conclusive evidence showing that this translates to improved health.
However, independent researchers remain intrigued by the technology, and say that the wealth of data generated through such apps could be used to help further stratify the different types of people who become at risk of developing type 2 diabetes.
“CGM not only enables a longer sampling time for capturing glucose levels, but will also capture lifestyle factors,” says Robert Wagner, a diabetes researcher at University Hospital Düsseldorf. “It is probable that it can be used to identify many clusters of prediabetic metabolism and predict the risk of diabetes and its complications, but maybe also specific cardiometabolic risk constellations. However, we still don’t know which forms of diabetes can be prevented by such approaches and how feasible and long-lasting such self-feedback dietary modifications are.”
Snyder himself has now been wearing a CGM for eight years, and he credits the insights it provides with helping him to manage his own diabetes. “My CGM still gives me novel insights into what foods and behaviors affect my glucose levels,” he says.
He is now looking to run clinical trials with his group at Stanford to see whether following a precision nutrition approach based on CGM and microbiome data, combined with other health information, can be used to reverse signs of pre-diabetes. If it proves successful, January AI may look to incorporate microbiome data in future.
“Ultimately, what I want to do is be able take people’s poop samples, maybe a blood draw, and say, ‘Alright, based on these parameters, this is what I think is going to spike you,’ and then have a CGM to test that out,” he says. “Getting very predictive about this, so right from the get go, you can have people better manage their health and then use the glucose monitor to help follow that.”
Starting this summer, the public buses in the Oberhaching suburb of Munich, Germany, won’t have to be plugged in to charge overnight anymore. Stefan Schelle, the mayor of Oberhaching, is taking advantage of the fact that an innovative startup has its offices in his community: Magment, short for “magnetizing cement,” will install its underground charging pad in the coming months. As soon as that happens, the buses will charge while they wait at the city’s main station or while stored at their overnight quarters.
In his light-filled office, Magment’s co-founder and CEO, Mauricio Esguerra, demonstrates how the new technology works: The lights on his black model car only flash when he puts the miniature Porsche directly atop the induction plate. “This works just like when you charge your iPhone on its charging pad or heat a pot on an induction range. People don’t have to be afraid of magnetic fields or anything like that,” says the 60-year-old Colombia-born entrepreneur. “The induction only gets activated when the storage battery is placed directly on top.
Patented by Esguerra, the “magnetizing concrete” is able to target the charge quite precisely. The batteries will be mounted in a box underneath the vehicles such as the retrofitted public buses. “Look, here’s one passing by,” says Esguerra, pointing out the window as a blue city bus rides past his office.
An invention finds its purpose
Esguerra grew up in Bogotá, studied physics at the Technical University Munich where he fell in love with a German woman, and started a family in her home country. For 15 years, he developed magnetic products, including the magnetizing cement, for Siemens, Europe’s largest industrial manufacturing company. The patent belonged to Siemens, of course. “But there were hardly any electric vehicles yet,” Esguerra says, “and Siemens didn’t quite know what to do with this invention.”
Esguerra changed companies a few times but, in 2015, he got an offer from Siemens. The patent for the magnetizing cement was expiring and Siemens wasn’t interested in keeping it. Would he, as the inventor, want it back? “I did not hesitate a second,” Esguerra remembers with a smile. “I’m a magnetician at heart.” That same year, he founded Magment to finally make use of the technology he created 20 years ago.
To demonstrate how his cement is made, he opens the lid of a plastic bucket filled with cement powder. Mixed in are fingernail-sized black pieces, so-called ferrites, mainly consisting of three ceramic oxides: iron, nickel and zinc. Conventionally, they are used in electronics such as cell phones, computers and cables. Molded in concrete, ferrites create a magnetic field that can transport charge to a vehicle, potentially eliminating range anxiety for EV drivers.
Molded in concrete, ferrites create a magnetic field that can transport charge to a vehicle, potentially eliminating range anxiety for EV drivers.
Magment
“Ferrites have extremely high rejection rates,” Esguerra adds. “It’s comparable to other ceramics: As soon as there is a small tear or crack, the material is rejected. We are talking about a rejection pile of 500,000 tons per year worldwide. There are mountains of unused materials.”
Exactly this fact was the starting point of his research at Siemens: “What can we do with this energy-intensive material? Back then, it was crushed up and mixed into the cement for building streets, without adding any function.” Today, too, the Magment material can simply be mixed with the conventional material and equipment of the cement industry. “We take advantage of the fact that we don’t have to build factories and don’t have high transportation costs."
In addition to saving resources, recycled ferrite also makes concrete more durable.
No plugs, no charging breaks
A young intern in the office next door winds cables around a new coil. These coils will later be lowered underground in a box, connected to the grid and encased in magnetizing concrete. The recipient box looks similar; it’s another coil but smaller, and it will be mounted underneath the carriage of the vehicle. For a car, the battery box would be 25 by 25 centimeters (about 10 inches), for a scooter five by five centimeters (about two inches).
Esguerra pushes an electric scooter into a cemented scooter rack next to his office. The charging pad is invisible. A faint beep is the only sign that it has started charging. “Childs play!” Esguerra says. “Even when someone puts in the scooter a little crooked, the charge still works. Our efficiency rate is up to 96 percent.” From this summer on, hotel chains in Munich will try out this system with their rental scooters, at a price of about 500 Euros per charging station.
Compared to plug-in charging, Magment’s benefits include smaller batteries that charge slower and, therefore, gentler, so they may last longer. Nobody needs to plug in the vehicles manually anymore. “Personally, I’ve had an EV for six years,” Esguerra says, “and how often does it happen that I forgot to plug it in overnight and then start out with a low charge in the morning? Once people get used to the invisible charging system, it will become the norm.“
There are also downsides: Most car companies aren’t ready for the new technology. Hyundai is the first carmaker that announced plans to equip some new models with inductive charging capability. “How many cars are electrified worldwide?” Esguerra asks and gives the answer himself: “One percent. And how many forklifts are electrified? More than 70 percent!” Therefore, Magment focuses on charging forklifts, e-scooters and buses.
Magment has focused most of its efforts on charging forklifts and other vehicle types that are entirely or predominantly electric, unlike cars.
Magment
On the morning of my visit to Esguerra’s office, a developer of the world’s third-biggest forklift manufacturer is there to inspect how the technology works on the ground. In the basement, a Magment engineer drives an electric forklift over a testbed with invisible charging coils, turning on the green charging light. Esguerra opens the interior of the forklift and points out the two batteries. “With our system, the forklift will only need one battery.” The savings, about 7,000 Euro per forklift, will pay for the installation of Magment’s charging system in warehouses, Esguerra calculates. “Less personnel and no unnecessary wait times for charging will lead to further savings,” he says.
To implement the new technology as efficiently as possible, Magment engineers began recording the transport routes of forklifts in warehouses. “It looks like spaghetti diagrams,” Esguerra explains. “Soon you get the areas where the forklifts pass or wait most frequently. This is where you install the chargers underground.” The forklifts will charge while in use, without having to pause for charging breaks. The method could also work for robots, for instance, in warehouses and distribution centers.
Roads of the future could be electric
Potential disadvantages might become apparent once the technology is more broadly in use. Therefore investors were initially reluctant, Esguerra admits. “Some are eager to be the first but most prefer to wait until the technology has been extensively used in real life.”
A clear hurdle today is that electrifying entire freeways with induction coils would cost at least 1 to 1.5 million Euros per kilometer. The German Department for Transportation even calculates overall costs of 14 to 47 million Euros per kilometer. So, the technology may only make sense for areas where vehicles pass or dwell the longest, like the Oberhaching train station or a busy interstate toll booth.
And yet, Magment is ramping up to compete with other companies that build larger inductive charging pads. The company just finished the first 20 meters of a testbed in Indiana, in partnership with the Purdue University and the Indiana Department of Transportation. Magment is poised to build “the world’s first contactless wireless-charging concrete pavement highway segment,” Purdue University announced.
The project, part of Purdue’s ASPIRE (Advancing Sustainability through Powered Infrastructure for Roadway Electrification) program, is financed by the National Science Foundation. “Indiana is known as the Crossroads of America, and we’re committed to fortifying our position as a transportation leader by innovating to support the emerging vehicle technology,” Governor Eric J. Holcomb said. If testing is successful, including the concrete’s capacity to charge heavy trucks operating at higher power (200 kilowatts and above), Indiana plans to identify a highway segment to install Magment’s charging pads. The earliest would be 2023 at best.
In the meantime, buses in the Californian Antelope Valley, trams at Hollywood's Universal Studios and transit buses in Tampa, Florida, are already charging with inductive technology developed by Wave, a company spun out of Utah State University. In Michigan, Governor Gretchen Whitmer announced plans to build a test route for vehicles to charge while driving, in collaboration with the Israel-based company Electreon, and this year contracted to build the first road-based charging system in the U.S. The state is providing support through an innovative grant program.
Costs remain one of the biggest obstacles, but Esguerra’s vision includes the potential that toll roads could charge a premium for inductive charging capabilities. “And in reverse, a driver who has too much energy could feed his surplus into the grid while driving,” Esguerra dreams.
Meanwhile, Wave’s upcoming big projects are moving trucks along a route in Southern California and running a UPS route between Seattle and Portland. Wave CTO Michael Masquelier describes the inductive power transfer his company champions as “similar to a tuning fork. By vibrating that fork, you sent energy through the air and it is received by another tuning fork across the room. So it’s similar to that, but it’s magnetic energy versus sound energy.”
He hopes to partner with Magment, saying that “the magnetizing cement makes installation easier and improves the energy efficiency.” More research is needed to evaluate which company’s technology will prove to be the most efficient, practical, and cost-effective.
Esguerra’s vision includes the potential that toll roads could charge a premium for inductive charging capabilities. “And in reverse, a driver who has too much energy could feed his surplus into the grid while driving,” Esguerra dreams.
The future will soon arrive in the idyllic town of Bad Staffelstein, a quaint tourist destination in the Upper Franconia region of Germany. Visitors will be taken to and from the main station and the popular thermal bath by driverless shuttles. Together with the University of Wuppertal, the regional government of Upper Franconia wants to turn its district into “the center of autonomous driving.” Magment is about to install inductive charging pads at the shuttle stations and the thermal bath, eliminating the need for the shuttles to stop for charging times. No more drivers, no cable, no range anxiety. Masquelier believes that “wireless and autonomous driving go hand in hand.” Science fiction? It will become science reality in spring 2023.
CORRECTION: An earlier version of the story erroneously mentioned that Electreon required overhead cables.
Smartwatches can track COVID-19 symptoms, study finds
If a COVID-19 infection develops, a wearable device may eventually be able to clue you in. A study at the University of Michigan found that a smartwatch can monitor how symptoms progress.
The study evaluated the effects of COVID-19 with various factors derived from heart-rate data. This method also could be employed to detect other diseases, such as influenza and the common cold, at home or when medical resources are limited, such as during a pandemic or in developing countries.
Tracking students and medical interns across the country, the University of Michigan researchers found that new signals embedded in heart rate indicated when individuals were infected with COVID-19 and how ill they became.
For instance, they discovered that individuals with COVID-19 experienced an increase in heart rate per step after the onset of their symptoms. Meanwhile, people who reported a cough as one of their COVID-19 symptoms had a much more elevated heart rate per step than those without a cough.
“We previously developed a variety of algorithms to analyze data from wearable devices. So, when the COVID-19 pandemic hit, it was only natural to apply some of these algorithms to see if we can get a better understanding of disease progression,” says Caleb Mayer, a doctoral student in mathematics at the University of Michigan and a co-first author of the study.
People may not internally sense COVID-19’s direct impact on the heart, but “heart rate is a vital sign that gives a picture of overall health," says Daniel Forger, a University of Michigan professor.
Millions of people are tracking their heart rate through wearable devices. This information is already generating a tremendous amount of data for researchers to analyze, says co-author Daniel Forger, professor of mathematics and research professor of computational medicine and bioinformatics at the University of Michigan.
“Heart rate is affected by many different physiological signals,” Forger explains. “For instance, if your lungs aren’t functioning properly, your heart may need to beat faster to meet metabolic demands. Your heart rate has a natural daily rhythm governed by internal biological clocks.” While people may not internally sense COVID-19’s direct impact on the heart, he adds that “heart rate is a vital sign that gives a picture of overall health.”
Among the total of 2,164 participants who enrolled in the student study, 72 undergraduate and graduate students contracted COVID-19, providing wearable data from 50 days before symptom onset to 14 days after. The researchers also analyzed this type of data for 43 medical interns from the Intern Health Study by the Michigan Neuroscience Institute and 29 individuals (who are not affiliated with the university) from the publicly available dataset.
Participants could wear the device on either wrist. They also documented their COVID-19 symptoms, such as fever, shortness of breath, cough, runny nose, vomiting, diarrhea, body aches, loss of taste, loss of smell, and sore throat.
Experts not involved in the study found the research to be productive. “This work is pioneering and reveals exciting new insights into the many important ways that we can derive clinically significant information about disease progression from consumer-grade wearable devices,” says Lisa A. Marsch, director of the Center for Technology and Behavioral Health and a professor in the Geisel School of Medicine at Dartmouth College. “Heart-rate data are among the highest-quality data that can be obtained via wearables.”
Beyond the heart, she adds, “Wearable devices are providing novel insights into individuals’ physiology and behavior in many health domains.” In particular, “this study beautifully illustrates how digital-health methodologies can markedly enhance our understanding of differences in individuals’ experience with disease and health.”
Previous studies had demonstrated that COVID-19 affects cardiovascular functions. Capitalizing on this knowledge, the University of Michigan endeavor took “a giant step forward,” says Gisele Oda, a researcher at the Institute of Biosciences at the University of Sao Paulo in Brazil and an expert in chronobiology—the science of biological rhythms. She commends the researchers for developing a complex algorithm that “could extract useful information beyond the established knowledge that heart rate increases and becomes more irregular in COVID patients.”
Wearable devices open the possibility of obtaining large-scale, long, continuous, and real-time heart-rate data on people performing everyday activities or while sleeping. “Importantly, the conceptual basis of this algorithm put circadian rhythms at the center stage,” Oda says, referring to the physical, mental, and behavioral changes that follow a 24-hour cycle. “What we knew before was often based on short-time heart rate measured at any time of day,” she adds, while noting that heart rate varies between day and night and also changes with activity.
However, without comparison to a control group of people having the common flu, it is difficult to determine if the heart-rate signals are unique to COVID-19 or also occur with other illnesses, says John Torous, an assistant professor of psychiatry at Harvard Medical School who has researched wearable devices. In addition, he points to recent data showing that many wearables, which work by beaming light through the skin, may be less accurate in people with darker skin due to variations in light absorption.
While the results sound interesting, they lack the level of conclusive evidence that would be needed to transform how physicians care for patients. “But it is a good step in learning more about what these wearables can tell us,” says Torous, who is also director of digital psychiatry at Beth Israel Deaconess Medical Center, a Harvard affiliate, in Boston. A follow-up step would entail replicating the results in a different pool of people to “help us realize the full value of this work.”
It is important to note that this research was conducted in university settings during the early phases of the pandemic, with remote learning in full swing amid strict isolation and quarantine mandates in effect. The findings demonstrate that physiological monitoring can be performed using consumer-grade wearable sensors, allowing research to continue without in-person contact, says Sung Won Choi, a professor of pediatrics at the University of Michigan who is principal investigator of the student study.
“The worldwide COVID-19 pandemic interrupted a lot of activities that relied on face-to-face interactions, including clinical research,” Choi says. “Mobile technology proved to be tremendously beneficial during that time, because it allowed us to collect detailed physiological data from research participants remotely over an entire semester.” In fact, the researchers did not have any in-person contact with the students involved in the study. “Everything was done virtually," Choi explains. "Importantly, their willingness to participate in research and share data during this historical time, combined with the capacity of secure cloud storage and novel mathematical analytics, enabled our research teams to identify unique patterns in heart-rate data associated with COVID-19.”