Making SNAP applications easier had positive economic effects

Making the application process for the Supplemental Nutrition Assistance Program more flexible generated “tangible downstream economic consequences”—like reduced credit card delinquencies—in Los Angeles and San Francisco, according to a new study.

The study assesses the impact of program designs aimed at making it easier to apply or recertify for SNAP, formerly known as the Food Stamp Program, and less prone to process-related denials.

More than 22.7 million households in the US receive SNAP benefits, or one in eight people. An estimated two-thirds are children, older adults, and the disabled.

The interventions in Los Angeles and San Francisco focused on giving applicants greater flexibility in scheduling their caseworker interviews, a mandatory component of the application process since missed interview appointments account for as many as half of all food stamp program denials nationwide.

The National Bureau of Economic Research (NBER) working paper estimates the effects of the Los Angeles and San Francisco application processes on financial health, using individual-level data derived from Californians’ credit reports (newly available in the University of California Consumer Credit Panel database) and SNAP administrative data for each of the two cities.

“[W]e find that reducing administrative burdens leads to significant improvements in the financial well-being of individuals seeking to gain or maintain access,” wrote researcher Tatiana Homonoff of NYU Wagner along with Min S. Lee and Katherine Meckel, both of the University of California-San Diego.

In the case of Los Angeles, people seeking food stamps were granted more autonomy over the scheduling of their interview through access to a new caseworker hotline. As a result of this procedural intervention, applicants were more likely to enroll successfully in the SNAP program.

That result led to meaningful improvements in their financial well-being: food stamp recipients were less likely to end up with a delinquent credit card account, and had higher credit scores. The amount of debt they carried through the year was hundreds of dollars less than those who were denied SNAP benefits solely due to the less flexible application process that was previously in effect, according to the paper.

The Los Angeles data covered 2020-21 and 65,000 who applied for SNAP.

The researchers noticed similar effects resulting from scheduling flexibility in the SNAP recertification process in San Francisco. People seeking food stamps renewals there were more likely to recertify for the program when they had additional time to reschedule missed caseworker appointments prior to the renewal deadline. This, in turn, improved financial health. The San Francisco data covered 2014-2016 and 40,000 people.

Taken as a whole, enhancing application and program integrity processes meant that SNAP applicants in the two counties were better able to pay their credit card balances or to carry less debt, according to the paper.

Source: NYU

The post Making SNAP applications easier had positive economic effects appeared first on Futurity.

from Futurity https://ift.tt/NwDVBAS

Fake beaver dams have a lot of environmental benefits

“Beaver mimicry” shows range of ecological benefits, researchers report.

The use of artificial beaver dams to replicate the ecological benefits created by the industrious rodents shows promise for offsetting damage to fish habitat, water quality, and biodiversity arising from climate change.

But as the use of such “beaver mimicry” spreads, particularly in the Pacific Northwest, there are key gaps in the research and a need for more studies that examine whether the outcomes seen in specific projects are broadly applicable.

That is a key takeaway from a new review of scientific literature by Washington State University researchers and others.

“There’s a fair amount of active research, but the extent to which this practice is being implemented is far outpacing the research on the subject,” says Jonah Piovia-Scott, associate professor in the School of Biological Sciences at Washington State University Vancouver and senior author of the paper.

“There’s a lot of good evidence coming out, but there’s still a lot of work to be done to bridge the gap between the potential benefits and what’s actually happening on the ground.”

The lead author on the paper in the journal Restoration Ecology was Jesse A.S. Burgher, who earned a PhD from WSU this year and is the wildlife program manager for the Cowlitz Indian Tribe. Other coauthors were Julianna Hallza, a current PhD student in Piovia-Scott’s lab, and Max Lambert, director of science for The Nature Conservancy in Washington.

The team evaluated 161 studies on a range of possible effects from beaver-related restoration efforts. They concluded there is substantial evidence that such efforts can make waterways more resilient to climate change, reducing summer water temperatures, increasing water storage, and enhancing flood-plain connectivity. They also can enhance biological diversity and build resistance to wildfires.

Beavers were once abundant in the Pacific Northwest, as well as across North America, but they were driven to near-extinction in the 18th and 19th centuries by the fur trade, which fueled economic expansion and early white settlement. Beaver populations have rebounded to a degree, but remain far below those earlier levels.

Efforts have been increasing to restore beaver populations and mimic the beneficial effects of the deep ponds created by their dams, as climate change threatens to diminish waterways and riparian areas, as well as the many species that rely on them.

Such projects have become especially common in the Pacific Northwest with the use of human-made “beaver-dam analogs.” In some cases, the dams are meant to simply replicate the effects of beaver structures; others are intended to attract beavers to recolonize waterways.

“This practice has become more and more widespread in our region,” Piovia-Scott says. “These are going in all over the place in the Pacific Northwest.”

One such example is a project along Oregon’s Bridge Creek, a tributary to the John Day River. Ecologists there have implemented a series of artificial dams—woven from willows and other plant materials—that have led to growth in the population of threatened steelhead and other fish.

It’s unknown whether those results would apply in other places with different ecological characteristics or for projects of differing sizes.

“That series of studies is fantastic and it’s really a great example of how these beaver-mimicry practices can improve fish habitat and contribute to fish-population growth,” Piovia-Scott says. “It’s not at all clear whether those benefits will accrue in other systems.”

Piovia-Scott says that his research work, which focuses on the effects of beavers and beaver dam analogs on a variety of wildlife, has led to important partnerships with tribes, non-profits and other groups focused on beaver restoration that may lack the resources to conduct scientific research themselves. Such partnerships are crucial to developing the best ecological approaches, he says.

“I work with a lot of restoration practitioners, and they have way more on-the-ground expertise in so many things than I do as a researcher,” he says.

“What they don’t have is a lot of capacity to conduct large-scale research projects—they may have the skills, but it’s usually not what their organizations are funded to do. So, there’s a tremendous amount of knowledge around restoration practices that can be generated through these partnerships.”

Source: Washington State University

The post Fake beaver dams have a lot of environmental benefits appeared first on Futurity.

from Futurity https://ift.tt/Fl3TZrU

Why people fall for fake news

In a world where misinformation spreads faster than fact, a new study is offering insight into why so many people fall for fake news, even when they suspect it’s false.

Researchers from Georgia State’s Robinson College of Business, Kennesaw State University, and the University of Tennessee have developed a model that explains how emotional cues, rather than accuracy, shape the way we consume and share news on social media.

The study, co-authored by Aaron French, Amrita George, Joshua Madden, and Veda C. Storey, appears in Information Systems Frontiers.

At the heart of the research is a simple question: Why do people believe and spread fake news, and do people consume fake news in the same way they consume tabloids?

Previous studies largely pointed to belief in fake news as confirmation bias, which is the tendency to believe information that supports your existing worldview. But this new study suggests something deeper is going on, especially during times of uncertainty like the COVID-19 pandemic.

“We found that people do consume fake news differently than tabloid news, which is largely consumed for entertainment and not taken seriously. With fake news, people are believing and sharing it because it feels useful either emotionally or informationally,” says Amrita George, co-author and clinical assistant professor of computer information systems(CIS) at Robinson.

In other words: fake news scratches an emotional itch. And in anxious, unstable times, that emotional itch is more powerful than truth.

For their study, the researchers defined fake news as news articles posing as legitimate news, but stemming from non-institutional journalistic sources that contain verifiably false information with the intention to deceive. To explore how people consume fake news, the researchers created the Content Dimensions–Overton Window–Perceived Utility or COP Model.

The model looks at three main factors in any piece of news:

• Veracity (how true it is)
• Emotional appeal (how it makes you feel)
• Relevance (how closely it connects to your life)

These factors shape how people judge whether a story is worth reading, liking, or sharing. Overlaying this is the Overton window, a political science concept that describes the range of ideas the public considers acceptable at a given time. If fake news falls within that window, or pushes its boundaries just enough, it’s more likely to be embraced.

To test their theory, the researchers analyzed more than 10,000 tweets about COVID-19. They looked at which tweets were “liked” and which were “ratioed” (received more negative comments than likes, signaling public disapproval). They also ran emotion and sentiment analyses to gauge tone, trust, and relevance.

“We found the Overton window plays a significant role in the response to fake news. It determined whether the fake news would be acceptable or unacceptable to people,” George says.

The researchers found people are highly sensitive to emotional tone, especially negative emotions like fear, anger, and disgust.

Even when a tweet was less truthful, if it hit the right emotional chord and felt relevant to a person’s life, it was more likely to be liked and shared. And interestingly, users were more forgiving of false information if the story felt emotionally satisfying. This tendency was much stronger with fake news than with traditional tabloid journalism, where readers typically know they’re not getting hard facts.

“A really interesting finding was that rather than providing information, fake news provided more emotional support in uncertain times, given we were analyzing fake news data from the COVID-19 pandemic,” says George.

This research lands at a critical moment. With AI-generated content flooding our feeds, understanding how and why fake news spreads is more urgent than ever.

The study offers practical insights. For example, the “ratio” of likes to replies on social media could help platforms flag potentially misleading or inflammatory content. Emotional tone, not just fact-checking, should be part of the detection process, according to the authors.

The findings also reinforce the importance of media literacy and the need to teach people not just how to spot falsehoods, but also how to recognize when their emotions are being manipulated.

Countries like Finland already include media literacy in school curricula starting in kindergarten. The researchers suggest similar programs could help inoculate the public against emotionally-driven misinformation.

Perhaps most importantly, the study shows how fake news can shift the boundaries of public discourse. When emotional stories are widely accepted, they slowly stretch the Overton Window, making extreme or previously unthinkable ideas feel normal.

“We’re not just talking about what people believe,” George says. “We’re talking about what becomes acceptable to believe. And that’s a much bigger deal.”

Source: Georgia State University

The post Why people fall for fake news appeared first on Futurity.

from Futurity https://ift.tt/NE4fCBz

Unlike birds, these dinos evolved to fly in a ‘burst’

Pterosaurs rapidly evolved flight, in contrast to modern birds, a new study suggests.

The study of fossils suggests that a group of giant reptiles alive up to 220 million years ago may have acquired the ability to fly when the animal first appeared, in contrast to prehistoric ancestors of modern birds that developed flight more gradually and with a bigger brain.

A report on the study, which used advanced imaging tools to examine the brain cavities of pterosaur fossils, and was funded in part by the National Science Foundation, appears in Current Biology.

The findings add to evidence that enlarged brains seen in modern birds and presumably in their prehistoric ancestors were not the driver of pterosaurs’ ability to achieve flight, says evolutionary biologist and Johns Hopkins Medicine assistant professor Matteo Fabbri.

“Our study shows that pterosaurs evolved flight early on in their existence and that they did so with a smaller brain similar to true non-flying dinosaurs,” he says.

Fabbri says the pterosaur was a force to be reckoned with in dinosaur skies, weighing up to 500 pounds and with a wingspan of up to 30 feet in some species. It is known to be the oldest of three groups of flying vertebrates (in addition to birds and bats) that independently evolved self-powered flight.

To learn whether pterosaurs acquired flight differently than birds and bats, the scientists studied the reptile’s evolutionary tree to pinpoint the evolution of pterosaur brain shape and size, looking for clues that may have led to the development of flight. They focused particularly on the area involved in vision, the optic lobe, the growth of which is thought to be associated with flying abilities.

Using CT scans and imaging software that allowed the scientists to extract information about the nervous systems of fossils, the researchers honed in on the pterosaur’s closest relative initially described by a team of researchers in 2016: the flightless, tree-dwelling lagerpetid that originated during the Triassic period 242 to 212 million years ago. In 2020, another group of scientists characterized the lagerpetid’s close relation to the pterosaur.

“The lagerpetid’s brain already showed features linked to improved vision, including an enlarged optic lobe, an adaptation that may have later helped their pterosaur relatives take to the skies,” says corresponding author Mario Bronzati, a researcher at University of Tübingen, Germany.

A larger optic lobe was also present in pterosaurs, Fabbri says. However, he says there were otherwise very few similarities in the shape and size of pterosaur brains and that of the flying reptile’s closest relative, the lagerpetid.

“The few similarities suggest that flying pterosaurs, which appeared very soon after the lagerpetid, likely acquired flight in a burst at their origin,” Fabbri says. “Essentially, pterosaur brains quickly transformed acquiring all they needed to take flight from the beginning.”

By contrast, modern birds are believed to have acquired flight in a step-by-step, more gradual process, inheriting certain features, such as an enlarged cerebrum, cerebellum, and optic lobes from their prehistoric relatives, and later adapting them to enable flight, says Fabbri. This theory is supported by 2024 findings from the lab of Amy Balanoff, assistant professor of functional anatomy and evolution at Johns Hopkins Medicine, that point to the expansion of the brain’s cerebellum as a key to bird flight. The cerebellum, located at the back of the brain, regulates and controls muscle movement among other activities.

“Any information that can fill in the gaps of what we don’t know about dinosaur and bird brains is important in understanding flight and neurosensory evolution within pterosaur and bird lineages,” Balanoff says.

In further studies, the scientists analyzed brain cavities of fossils from crococdylians (crocodile ancestors) and early, extinct birds, and compared these with pterosaur brain cavities.

They determined that the pterosaur’s brain had moderately enlarged hemispheres, similar in size to other dinosaurs—including two-legged bird-like troodontids living during the Late Jurassic to the Late Cretaceous periods 163 to 66 million years ago, and the oldest-known bird, Archaeopteryx lithographica from 150.8 million to 125.45 million years ago—compared with the brain cavities of modern birds.

In the future, Fabbri says that better understanding how the structure of the pterosaur brain, in addition to the size and shape, enabled flight will be the most important step to better infer the basic biological laws of flight.

Funding support for this research came from the Alexander von Humboldt Foundation, Brazilian Federal Government, The Paleontological Society, Agencia Nacional de Promoción Científica y Técnica, Conselho Nacional de Desenvolvimento Científico e Tecnológico, the European Union NextGeneration EU/PRTR, the National Science Foundation, and the Swedish Research Council

Additional researchers from the New York Institute of Technology; the American Museum of Natural History; Federal University of Santa Maria, Brazil; the University of Ohio; Bernardino Rivadavia Museum of Natural Science; São Paulo State University; Yale University; Universidad Nacional de La Plata, Argentina; Museo Nacional de Ciencias Naturales, Spain; Universidade de São Paulo, Brazil; University of Birmingham; Virginia Tech; Stony Brook University; and Eberhard Karls University of Tübingen contributed to the work.

Source: Johns Hopkins University

The post Unlike birds, these dinos evolved to fly in a ‘burst’ appeared first on Futurity.

from Futurity https://ift.tt/YpFAwrh

Happy hours invites take a toll on some workers

New research suggests that there are both positives and negatives to getting an invite to after-work events, but that impact depends on the person.

While extroverts tend to feel good about themselves and their workplace when they get after-work invites, those with more reserved personalities can become more withdrawn and stressed when invited.

“We always think that social activity is so great, right? If you’re social with your coworkers, you feel energized and connected. But those invitations are not necessarily always good,” says Joanna Lin, corresponding author of the study and a professor of management in the University of Georgia Terry College of Business.

“There’s a social pressure that makes people feel like they have to say yes and need to be there. These outings seem like an obligation, even if they’re supposed to just be something fun.”

Across multiple experiential and field studies and surveys with hundreds of full-time employees, Lin found that social invitations extended to coworkers don’t always create a feeling of belonging.

Some workers who receive invitations to things like dinners, parties or activities such as bowling after-work experience a profound sense of gratitude. The researchers found that these folks tend to already be more social beings, with a higher social confidence than others.

Contrarily, the study showed that workers who identified as shy or less confident in social settings felt pressured after being asked to hang out in order to appease their coworkers at the cost of their own stress levels.

“If you’re a social butterfly, you’re really good at interacting with others, so that doesn’t cause strife. When you already have a hard time being social, however, that sense of expectation contributes to the stress. You feel generally grateful for their gesture to include you in this social event but are worried about your social skills,” Lin says.

Those who anticipated that added pressure to perform socially after work also became more anxious regardless of if they accepted the invite, causing them to be tense and unproductive at work in that waiting period.

“There’s also that uncertainty. ‘If I say yes, how long will it take? Who else is going? Or if I say no, what are the consequences? Maybe my coworker will be mad at me or I will feel left out.’ There are lots of psychological decision-making points that a simple invite causes,” Lin says.

Acknowledging your own preferences and well-being in the workplace can make you feel better overall, the researchers found. That could make you more equipped to handle the possible added stress of social gatherings.

It’s also important for those in workplaces to know each other and to consider the potential impact of invites before they’re given. It doesn’t hurt to consider when you ask someone to hang out either.

“When we initiate this invitation, it has unintended consequences to employee performance for that day,” she says. “I think we should be mindful of when we give it. You may think you’re going to help this person by inviting them, but on the other hand, you could be making that person think ‘Oh my god, what am I going to do?'”

The researchers also say there is a need to understand additional factors in the invite process, like if it’s a supervisor giving the invite or if you say yes when you don’t want to, if you feel more socially confident, and likely to say yes again after the event concludes.

This study appears in Personnel Psychology.

Additional coauthors are from National Taiwan Normal University, Indiana University, and National Sun Yat-sen University.

Source: University of Georgia

The post Happy hours invites take a toll on some workers appeared first on Futurity.

from Futurity https://ift.tt/DR0ECf7

New work paves way for better infection treatments

A new way to make phage DNA lays groundwork for better infection treatments, researchers report.

Researchers developed a method to construct bacteriophages with entirely synthetic genetic material, allowing researchers to add and subtract genes at will.

The findings open new ways to understand how these bacteria-killing viruses work and to create potential therapies to fight the worsening problem of antibacterial resistance.

There is massive variation among phages, but researchers don’t know the roles played by many individual genes, says Graham Hatfull, a professor of biotechnology at the University of Pittsburgh and one of the study’s lead researchers.

“How are the genes regulated? What happens if we remove this one or that one? We don’t have the answers to those questions, but now we can ask—and answer—almost any question we have about phages,” he says. “This will speed up discovery.”

The team constructed synthetic DNA modeled after two naturally occurring phages that attack Mycobacterium, which include the pathogens responsible for tuberculosis and leprosy, among others. They then added and removed genes, successfully editing the synthetic genomes of both.

They published their work in the Proceedings of the National Academy of Sciences.

Biologists are already able to create synthetic DNA, but the process is notoriously difficult for a certain class of phages due to their structure. DNA is built of two pairs of chemical building blocks, or base pairs, represented by the letters A, T, C, and G. The phages that attack Mycobacterium are about 65% G and C.

“Traditional methods of synthesizing DNA have technical problems with so-called ‘high GC’ DNA,” Hatfull says, as opposed to more easily editable genomes like those of E. Coli, which have ratios of base pairs that are closer to even.

To move past these obstacles, Hatfull worked with Greg Lohman of New England Biolabs, a company known for techniques enabling the design and assembly of synthetic DNA. Also integral to the work was Ansa Biotech, which has developed a process that overcomes the hurdles of synthesizing high GC DNA. The paper’s first author, Ching-Chung Ko, is a research associate in Hatfull’s lab.

The team was able to chemically synthesize DNA identical to two naturally occurring phages: BPs—a 40,000-base-pair virus used clinically to treat a bacteria that often infects people with cystic fibrosis—and Bxb1, which has 50,000 base pairs. They constructed the DNA in 12 sections and inserted them into a cell, which followed the instructions from its new genome: make phages.

Researchers and clinicians have become increasingly interested in phages as a response to fighting antibiotic-resistant bacterial infections. For perhaps as long as three billion years, phages and bacteria have evolved alongside one another, resulting in niches not unlike Darwin’s finches; one phage will only attack one specific kind of bacteria.

Exactly how phage genomes codify these relationships remains mostly a mystery.

Hatfull’s lab has freezers full of 28,000 phages that were found in dirt, ponds or even on rotting fruits. Finding one that will attack any particular strain of bacteria is a process of directed trial and error. When a clinician sends a sample from a sick patient, researchers use their experience, their library of about 5,500 phage genomes and plenty of petri dishes to search for a match tailored specifically to that patient.

Precisely altering phage genomes and observing how those changes affect behavior will be a game-changer that will both inform researchers’ understanding of how the phages work and, later, may allow them to engineer phages with broader applications.

“We’ve been surrounded by questions that we can’t always answer because we didn’t have the technologies to be able to do so,” Hatfull says. “This is a technological advance that enables us in principle to begin to answer many questions in a much simpler way than we could have done before.”

In addition, the ability to create entirely synthetic genomes will alleviate the need to keep tens of thousands of phages on ice, with duplicates of some, and sufficient backup power sources. Instead of storing biology, Hatfull hopes one day phages can be stored simply as information.

“And then, the sky’s the limit,” Hatfull says. “You can make any genome you want. You’re only limited by what you can imagine would be useful and interesting to make.”

Source: University of Pittsburgh

The post New work paves way for better infection treatments appeared first on Futurity.

from Futurity https://ift.tt/Wx0sQFp

What are the benefits of AI-powered devices for diabetes care?

A new study explores value and shortcomings of AI-enhanced wearable devices for Type 2 diabetes and prediabetes care.

Artificial intelligence-enhanced wearable devices, such as continuous glucose monitors (CGMs), have dramatically improved the ability of people with diabetes and even prediabetes to better understand and control their blood sugar.

But research on AI-enhanced wearable devices has been uneven, often focused on just a few kinds of devices, data types, and AI models.

University at Buffalo researchers have published in NPJ Digital Medicine the first comprehensive meta-review (a study of studies) of AI-enhanced wearables for people with prediabetes and Type 2 diabetes.

Their conclusion is that these devices have enormous potential that will be realized once certain challenges are overcome.

Just how much AI-enhanced wearable devices could help patients with diabetes became clear to Raphael Fraser, corresponding author and associate professor of medicine in the Jacobs School of Medicine and Biomedical Sciences at the University at Buffalo, when he saw that AI-enhanced CGMs can provide data every few minutes, instead of just a few readings per day.

“As newer AI models demonstrated the ability to recognize patterns and predict glucose changes before they happened, it became clear that diabetes care could shift from reacting to problems after they occur to anticipating and preventing them,” Fraser says.

“That was the moment when I realized AI could genuinely transform daily management and long-term outcomes. AI turns CGMs from a rear-view mirror into a heads-up display.

“For people living with diabetes, AI-enabled wearables have the potential to provide more timely and personalized guidance, helping them avoid glucose swings and manage daily decisions with greater confidence,” says Fraser.

“For clinicians, the key takeaway is that these tools may help identify risks earlier and support more efficient care.”

While larger studies are needed to evaluate the benefit of CGMs for people with prediabetes, early use of wearables combined with AI could support lifestyle changes and perhaps delay or prevent the progression to diabetes.

Research in this area is growing rapidly, but the studies done to date have been scattered across different devices, data types, and AI models, making it difficult to see the bigger picture.

“We wanted to bring everything together to understand what we actually know, what has been consistently shown and where the evidence is still thin,” explains Fraser.

“Our goal was to identify which approaches seem most effective, where the limitations are and what gaps need to be addressed before AI-enabled wearables can become routine tools in clinical care.”

The researchers selected 60 out of 5,000 peer-reviewed studies that examined the integration of AI and wearable technology in diabetes management.

There were many positive findings. “AI-enhanced wearables can predict glucose changes up to one to two hours in advance, helping individuals maintain steadier control and receive personalized guidance that reflects their daily routines, activity levels and sleep patterns,” Fraser says.

These systems also have the potential to reduce clinical workload by sorting through large streams of data and highlighting what requires attention.

But the researchers also found aspects of AI-enhanced wearables that were problematic; for example, AI-enhanced wearables are based on different AI models, which must be transparent and validated before they are widely adopted, Fraser says.

“Many AI models operate as ‘black boxes,’ making it difficult for clinicians and patients to understand or trust their recommendations,” Fraser explains, “which limits their usefulness in guiding day-to-day decision-making for people with prediabetes and Type 2 diabetes.”

For example, he says, an AI-enabled glucose app may warn a user that their blood sugar is likely to rise in the next 30 minutes but provide no insight into what triggered the prediction, whether that is a recent meal, reduced physical activity, elevated stress, poor sleep, or normal day-to-day fluctuations.

“When people cannot see the ‘why’ behind the alert, they struggle to decide what action to take, making the tool far less helpful in real life,” says Fraser.

He adds that in some of the studies, limited sample sizes and narrow demographic representation reduce how broadly the findings can be applied. In addition, the lack of standardized benchmark datasets means results are not always easy to compare across studies. Practical barriers, such as inconsistencies in the quality of the data included in the studies, limited integration into clinical workflows, and the cost and accessibility of wearable devices, also curb widespread adoption of these devices, says Fraser.

Another factor that influences how well a CGM will work for patients and clinicians is what type of AI model it uses.

“Different AI models are suited to different kinds of data and prediction tasks,” explains Fraser. “Models designed to learn patterns over time, such as long short-term memory networks or similar architectures, tend to perform better with continuous glucose data because they can track trends and anticipate future changes,” he says.

“Newer models like transformers are particularly good at integrating multiple forms of data, such as glucose, heart rate, sleep and physical activity, which can give a more holistic understanding of the body’s metabolic state.”

Nevertheless, he adds, sometimes simpler models are easier for clinicians to interpret. “So the challenge is not just choosing the most powerful model but choosing one that performs well while also being understandable and clinically trustworthy,” he says. “The ‘right’ AI is the one that fits the data and can be explained in the doctor’s office.”

The research was supported by the American Diabetes Association, the National Institute of Diabetes and Digestive Kidney Disease, and the National Institute for Minority Health and Health Disparities.

Source: University at Buffalo

The post What are the benefits of AI-powered devices for diabetes care? appeared first on Futurity.

from Futurity https://ift.tt/URWraZ8