Duke Research Blog

Following the people and events that make up the research community at Duke.

Category: Lecture (Page 2 of 19)

Does Digital Healthcare Work?

Wearable technologies like Fitbit have been shown to provide a short-term increase in physical activity, but long-term benefits are still unclear, even if recent studies on corporate wellness programs highlight the potential healthcare savings.

Headshot of Luca Foschini

Luca Foschini, PhD is a co-founder and head of data science at Evidation Health, and a visiting research scientist at UCSB. Source: Network Science IGERT at UCSB.

To figure out the effects of these technologies on our health, we need ways to efficiently mine through the vast amounts of data and feedback that wearable devices constantly record.

As someone who has recently jumped on the Fitbit “band” wagon, I have often wondered about what happens with all the data collected from my wrist day after day, week after week.

Luca Foschini, a co-founder and head of data science at Evidation Health, recently spoke at Duke’s Genomic and Precision Medicine Forum where he explained how his company uses these massive datasets to analyze and predict how digital health interventions — Fitbits and beyond — can result in better health outcomes.

California-based Evidation health uses real-life data collected upon authorization from 500,000-plus users of mobile health applications and devices. This mobile health or “mHealth” data is quickly becoming a focus of intense research interest because of its ability to provide such a wealth of information about an individual’s behavior.

Foschini and Evidation Health have taken the initiative to design and run clinical studies to show the healthcare field that digital technologies can be used for assessing patient health, behavioral habits, and medication adherence, just to name a few.

Foschini said that the benefits of mobile technologies could go far beyond answering questions about daily behavior and lifestyle to formulate predictions about health outcomes. This opens the door for “wearables and apps” to be used in the realm of behavior change intervention and preventative care.

Foschini speaks at Duke’s Genomic and Precision Medicine Forum

Foschini explains how data collected from thousands of individuals wearing digital health trackers was used to find a associations between activity tracking patterns and weight loss.

Evidation Health is not only exploring data based on wearable technologies, but data within all aspects of digital health. For example, an interesting concept to consider is whether devices create an opportunity for faster clinical trials. So-called “virtual recruiting” of participants for clinical studies might use social media, email campaigns and online advertising, rather than traditional ads and fliers. Foschini said a study by his firm found this type of recruitment is up to twelve times faster than normal recruitment methods for clinical trials (Kumar et al 2016). 

While Foschini and others in his field are excited about the possibilities that mHealth provides for the betterment of healthcare, he acknowledges the hurdles standing in the way of this new approach. There is no standardization in how this type of data is gathered, and greater scrutiny is needed to ensure the reliability and accuracy of some of the apps and devices that supply the data.

amanda_cox_100 Post by Amanda Cox

Diabetes — and Privacy — Meet ‘Big Data’

“Click here to consent forever.”

If consent to participate in medical research were that simple, Joanna Radin of Yale University would have to find a new focus for her research, and I would never have found the Trent Center for Bioethics, Humanities & History of Medicine.

Luckily for us both, this is not the case. Medical consent is a very complex issue that can, as Radin’s research attests, traverse generations.

joanna-radin-headshot

Joanna Radin’s reserach focuses on the intersection of medical history, anthropology and ethics at Yale University. Source: Yale School of Medicine

Radin is an Associate Professor of Medical History at Yale, the perfect fit for the Humanities in Medicine Lecture Series taking place this month at the Trent Center. Her research nails the narrow intersection of medical history, anthropology, bioethics and data analytics. In fact, Radin’s appeal is so broad that her visit to Duke was sponsored by no less than six Duke departments, including the Departments of Computer Science, History, Electrical and Computer Engineering, Cultural Anthropology and Statistical Science.

Radin’s lecture honed in on a well-known case in the realm of bioethics and medical history: the Pima Native American tribe in Arizona, which is known for unusually high rates of diabetes and obesity. The Pima were the first Native American tribe to be granted a reservation in Arizona—30,000 acres—at the beginning of the California Gold Rush. In 1963, following nearly half a century of mass famine among the Pima, the National Institute of Health (NIH) conducted a survey for rheumatoid arthritis in the Pima tribe, instead discovering a frighteningly high frequency of diabetes.

In 1965, the NIH initiated a long-term observational study of the Pima that continued for about 40 years, though it was meant to last no more than 10. The goal of the study was to learn about diabetes in the “natural laboratory” of sorts that the Pima reservation unwittingly provided. The data collected in this study came to be known as the Pima Indian Diabetes Data set (PIDD).

Machine learning enters the story around 1987, when David Aha and colleagues at the University of California, Irvine (UCI) created the UCI Machine Learning Repository, an archive containing thousands of data sets, databases and data generators. The repository is still active today, virtually a gold mine for researchers in machine learning to test their algorithms. The PIDD is one of the oldest data sets on file in the UCI archive, “a standard for testing data mining algorithms for accuracy in predicting diabetes,” according to Radin.

pima_indian_man_miguel_a_farmer_pima_arizona_ca-1900_chs-3625

A Pima farmer in Pima, Arizona, circa 1900. Source: Wikimedia Commons

Generations’ worth of data on the Pima tribe have been publicly accessible in the UCI archive for over two decades, creating ethical controversy around the accessibility of information as personal as blood pressure, body mass index (BMI) and number of pregnancies of Pima Native Americans. Though the PIDD can help refine machine learning algorithms that could accurately predict—and prevent—diabetes, the privacy issues provoked by the publicness of the data are impossible to ignore.

This is where “eternal” medical consent enters the equation: no researcher can realistically inform a study participant of what their medical data will be used for 40 years in the future.

These are the interdisciplinary questions that Radin brought forth in her lecture, weaving together seemingly opposite fields of study in an engaging, thought-provoking presentation. No one who left that room will look at the Apple Terms & Conditions the same way again.

 

Post by Maya Iskandarani iskandarani_maya_100hed

Economics and Health: The Biases Behind Our Decisions

Eric Finkelstein of the the Duke Graduate Medical School in Singapore studies how economic principles might be used to improve individual healthcare.

At a talk last Friday, Finkelstein, who was selected by Thomson Reuters as one of the world’s most influential scientific minds of 2015, argued that the same biases that affect our economic decisions could also influence our healthcare choices, and that understanding these biases could help motivate individuals to live healthy active lives.

In theory, people should be able to make healthy choices, Finkelstein said. Under the utility maximization model, individuals have the ability to rationalize and recognize the benefits of taking particular actions for themselves. But often we are not rational beings, he said, and there are several “deviations” that steer us away from maximizing our utility.

One of these deviations is the “present bias” preference, which leads us to make decisions in the present that our future self will regret. He discussed a particular experiment in which people are asked to choose what they will eat in one week’s time: a candy bar or an apple. Most choose the apple, but after a week, when they were given the opportunity to reevaluate their choice and change it, most switch to the candy bar.

This experiment shows not only the dynamic, unpredictable nature of our decisions, but also highlights our tendency to overestimate the will power of our “future selves.”

Another interesting bias that prevents us from being rational is our probabilistic assessment bias, which describes our tendency to overestimate the probability of very unlikely events, while underestimating the probability of those that are likely. This bias directly relates to health and our tendency to ignore the possibility of suffering a detrimental health problem like a heart attack, when in reality it’s quite commonplace.

Eric Finkelstein’s research, which focuses on the intersection between economics and global health, has gained him renowned success nationally and abroad. Source: Duke NUS Medical School.

To understand how these biases might influence individuals suddenly diagnosed with a terminal illness, Finkelstein and his medical team in Singapore conducted their own study on healthcare choices. In the experiment, both healthy and sick individuals were asked to identify what treatments they would prioritize if diagnosed with terminal cancer: level of pain, hours of care required, potential to extend life, cost of treatment and location of death.

Most healthy individuals said they would want whatever treatment was cheapest, but showed very little interest in investing in extending their life or selecting where they died. When sick patients were asked the same questions, on the other hand, they valued place of death (home was preferred) and survival time above everything else. Such information indicates just how difficult it is for us to predict where to invest in healthcare for cancer patients.

From this study and several others, Finkelstein concludes that we are not rational beings, but are instead irrational ones that feed off of biases and change our opinions constantly. But, he suggests that through the use of incentives, we can mediate these irrational biases and ultimately improve health outcomes.

 

Post by Lola Sanchez-Carrion

lola_sanchez_carrion_100hed

Violence: Risk vs Protection Factors

On Oct. 3, in place of a typical Interdisciplinary Discussion Course (IDC) for the Focus program, we were brought together in the White Lecture Hall on East to hear Dr. Jeremy G. Richman give a lecture on “Violence, Compassion, and the Brain”.

Dr. Jeremy G. Richman

Dr. Jeremy G. Richman

Since his daughter Avielle was killed in the Sandy Hook Elementary School shooting, Dr. Richman has been studying violence and the brain with the Avielle Foundation. Using his extensive research experience in neuroscience and neuropsychopharmacology, Dr. Richman has been working with a team on understanding the risk and protective factors for violence.

untitled-3

Dr. Richman lecturing in the White Lecture Hall on East Campus

 

 

 

 

 

 

The brain is very complex, and its processes and connections are still very much veiled. As a result, The Avielle Foundation currently conducts research on understanding violence by “bridging biochemical and behavioral sciences” using functional MRI brain scans, biochemistry, and genetics and epigenetics.

Based on the research he has analyzed, Dr. Richman identified four types of adverse childhood experiences (ACEs) that are risk factors for violence: psychological, physical, sexual, and household dysfunction/neglect.

Interestingly enough, research has shown that physically abused children are not necessarily more likely to be arrested for a violence-related crime. While it can happen, the situation is complicated because there is more than one factor involved. Adverse childhood experiences typically lead to violence towards oneself. When someone has at least four ACEs, their risk for alcoholism increases seven-fold. When males have more than five experiences within the ACE categories, their risk for drug use increases by 46 times. The more ACEs experienced, the greater exponentially the percentage of lifetime history of attempted suicides.

Studies have shown that firearm access in the home “is associated with an increased risk of firearm homicide and firearm suicide in the home.”

The debate of Nature vs. Nurture also comes into play. Humans have a monoamine oxidase A (MAOA) gene, also known as the Warrior Gene. MAOA is also associated with low dopamine levels. When males experience physical abuse and have low MAOA, the chance for them to have psychopathology increases. The interesting thing is that there is one main difference between a violent psychopath and a resilient leader: childhood experiences. A violent psychopath likely had an adverse childhood experience; the resilient leader had a nurturing childhood. Richman noted that studies involving neuroscience and psychology always have shortcomings because of the brain’s complexity. The brain’s processes and connections are still very much veiled.

Despite all these risk factors to violence, there are protective factors in place. One is less at risk when in a compassionate, kind, resilient, and connection-building environment with family and peers who embody these ideals. Healthy habits, good nutrition, and exercise help reduce stress, which in turn helps reduce the chances for violence.

Post by Meg Shiehmeg_shieh_100hed

Students Mine Parking Data to Help You Find a Spot

No parking spot? No problem.

A group of students has teamed up with Duke Parking and Transportation to explore how data analysis and visualization can help make parking on campus a breeze.

As part of the Information Initiative’s Data+ program, students Mitchell Parekh (’19) and Morton Mo (’19) along with IIT student Nikhil Tank (’17), spent 10 weeks over the summer poring over parking data collected at 42 of Duke’s permitted lots.

Under the mentorship of graduate student Nicolas-Aldebrando Benelli, they identified common parking patterns across the campus, with the goal of creating a “redirection” tool that could help Duke students and employees figure out the best place to park if their preferred lot is full.

A map of parking patterns at Duke

To understand parking patterns at Duke, the team created “activity” maps, where each circle represents one of Duke’s parking lots. The size of the circle indicates the size of the lot, and the color of the circle indicates how many people entered and exited the lot within a given hour.

“We envision a mobile app where, before you head out for work, you could check your lot on your phone,” Mo said, speaking with Parekh at the Sept. 23 Visualization Friday Forum. “And if the lot is full, it would give you a pass for an alternate lot.”

Starting with parking data gathered in Fall 2013, which logged permit holders “swiping” in and out from each lot, they set out to map some basic parking habits at Duke, including how full each lot is, when people usually arrive, and how long they stay.

However, the data weren’t always very agreeable, Mo said.

“One of the things we got was a historical occupancy count, which is exactly what we wanted – the number of cars in the facility at a given time – but we were seeing negative numbers,” said Mo. “So we figured that table might not be as trustworthy as we expected it to be.”

Other unexpected features, such as “passback,” which occurs when two cars enter or exit under the same pass, also created challenges with interpreting the data.

However, with some careful approximations, the team was able to estimate the occupancy of lot on campus at different times throughout an average weekday.

They then built an interactive, Matlab-based tool that would suggest up to three alternative parking locations based on the users’ location and travel time plus the utilization and physical capacity of each lot.

“Duke Parking is really happy with the interface that we built, and they want us to keep working on it,” Parekh said.

“The data team worked hard on real world challenges, and provided thoughtful insights to those challenges,” said Kyle Cavanaugh, Vice President of Administration at Duke. “The team was terrific to work with and we look forward to future collaboration.”

Hectic class schedules allowing, the team hopes to continue developing their application into a more user-friendly tool. You can watch a recording of Mo and Parekh’s Sept. 23 presentation here.

The team's algorithm recommends up to three alternative lots if a commuter's preferred lot is full. In this video, suggested alternatives to the blue lot are updated throughout the day to reflect changing traffic and parking patterns. Video courtesy of Nikhil Tank.

Kara J. Manke, PhD

Post by Kara Manke

 

How to Get a Lemur to Notice You

Duke evolutionary anthropology professor Brian Hare studies what goes on in the minds of animals.

Duke evolutionary anthropology professor Brian Hare studies what goes on in the minds of animals.

Duke professor Brian Hare remembers his first flopped experiment. While an undergraduate at Emory in the late 1990s, he spent a week at the Duke Lemur Center waving bananas at lemurs. He was trying to see if they, like other primates, possess an important social skill. If a lemur spots a piece of food, or a predator, can other lemurs follow his gaze to spot it too?

First he needed the lemurs to notice him. If he could get one lemur to look at him, he could figure out if other lemurs then turn around and look too. In similar experiments with monkeys and chimps, oranges had done the trick.

“But I couldn’t get their attention,” Hare said. “It failed miserably.”

Hare was among more than 200 people from 25 states and multiple countries who converged in Durham this week for the 50th anniversary celebration of the Duke Lemur Center, Sept. 21-23, 2016.

Humans look to subtle movements in faces and eyes for clues to what others are thinking, Hare told a crowd assembled at a two-day research symposium held in conjunction with the event.

If someone quickly glances down at your name tag, for example, you can guess just from that eye movement that they can’t recall your name.

We develop this skill as infants. Most kids start to follow the gaze of others by the age of two. A lack of interest in gaze-following is considered an early sign of autism.

Arizona State University graduate student Joel Bray got hooked on lemurs while working as an undergraduate research assistant in the Hare lab.

Arizona State University graduate student Joel Bray got hooked on lemurs while working as an undergraduate research assistant in the Hare lab.

“Gaze-following suggests that kids are starting to think about the thoughts of others,” Hare said. “And using where others look to try to understand what they want or what they know.”

In 1998 Hare and researchers Michael Tomasello and Josep Call published a study showing that chimpanzees and multiple species of monkeys are able to look where others are looking. But at the time not much was known about cognition in lemurs.

“When you study dogs you just say, ‘sit, stay,’ and they’re happy to play along,” Hare said. Working at the Duke Lemur Center, eventually his students discovered the secret to making these tree-dwelling animals feel at home: “Lemurs like to be off the ground,” Hare said. “We figured out that if we just let them solve problems on tables, they’re happy to participate.”

Studies have since shown that multiple lemur species are able to follow the gaze of other lemurs. “Lemurs have gone from ignored to adored in cognitive research,” Hare said.

 

Ring-tailed lemurs are among several species of lemurs known to follow the gaze of other lemurs. The ability to look where others are looking is considered a key step towards understanding what others see, know, or might do. Photo by David Haring, Duke Lemur Center.

Ring-tailed lemurs are among several lemur species known to follow the gaze of other lemurs. The ability to look where others are looking is considered a key step towards understanding what others see, know, or might do. Photo by David Haring, Duke Lemur Center.

Robin SmithPost by Robin A. Smith

Page 2 of 19

Powered by WordPress & Theme by Anders Norén