The virtual doctor won’t see you just yet. But that day is getting closer.

Today’s health-care industry is making increasing use of Web-based virtual agents and avatars, or computerized assistants, not only to perform clerical duties but also to dispense medical information.

In Spain, an avatar named Osane for the past year has been helping visitors navigate the website of the Basque region’s public-health system. Osane, an animated character, mainly helps users with clerical matters such as changing their physician or booking an appointment.

But she also can give basic advice on healthy living and common illnesses. Type into the chat window that you think your son has varicella, for example, and Osane will respond with a brief description of the symptoms of chickenpox and offer to help book an appointment with a pediatrician.

But Osane knows her limits. Tell her only that your son has a fever, and the avatar will say she is not a doctor.

Osane runs on software from Anboto, a Spanish start-up with offices in Boston and Silicon Valley. Its technology is also used by a Spanish health insurer, IMQ, to help customers navigate its website.

Step by step Avatars helping hospital patients with discharge instructions can slash readmission rates

“Health care is a very broad sector with many potential applications for virtual agents,” says Anboto Chief Executive Xabier Uribe-Etxebarria.

Most avatars in use in health care today handle administrative tasks and website navigation. The U.S. insurer Aetna Inc. introduced an avatar from Next IT Corp., of Spokane, Wash., on its website in 2010, partly to reduce the load on its call center.

Nevertheless, some experts believe the technology is ready to be used in a clinical context—though these applications are mainly at the research stage.

Timothy Bickmore, an associate professor at Boston’s Northeastern University, used avatars in research aimed at helping hospital patients understand their discharge information. Many don’t, and so run a greater risk of being readmitted. Mr. Bickmore says his three-year trial at a Boston hospital reduced readmissions by 30%.

Craig Mundie, Microsoft Corp.’s chief research and strategy officer, last year demonstrated the use of avatars and Microsoft’s Xbox Kinect technology to facilitate group therapy sessions—the theory being that patients will feel less intimidated by an avatar than by a human group leader.

Cisco Systems Inc. has also demonstrated an avatar, named Patty, that counsels hospital patients. Patty gives the patients information on their treatment and allows them to ask questions they might be too embarrassed to ask a nurse or physician.

Northeastern’s Mr. Bickmore says avatars have an attribute that health-care professionals often lack—patience.

“Many people prefer avatars to nurses for counseling because they do not feel rushed and can ask questions,” he says

As new roles are explored for avatars in health care, experts say the key is to ensure the avatar recognizes its own boundaries and doesn’t overstep them.

“You have to be particularly careful,” Mr. Bickmore says, “with anything that involves human judgment or requires common-sense reasoning about the world.”

Article originally appeared in The Wall Street Journal

Just as the name implies, com­plex sys­tems are dif­fi­cult to tease apart. An organism’s genome, a bio­chem­ical reac­tion, or even a social net­work all con­tain many inter­de­pen­dent components—and changing any one of them can have per­va­sive effects on all the others. In the case of a very large system, like the human genome, which con­tains 20,000 inter­con­nected genes, it’s impos­sible to mon­itor the whole system at once.

But that may not matter any­more. In a paper pub­lished in the pres­ti­gious mul­ti­dis­ci­pli­nary journal Pro­ceed­ings of the National Academy of Sci­ence, North­eastern net­work sci­en­tists have devel­oped an algo­rithm capable of iden­ti­fying the subset of components—or nodes—that are nec­es­sary to reveal a com­plex system’s overall nature.

The approach takes advan­tage of the inter­de­pen­dent nature of com­plexity to devise a method for observing sys­tems that are oth­er­wise beyond quan­ti­ta­tive scrutiny.

“Con­nect­ed­ness is the essence of com­plex sys­tems,” said Albert-​​László Barabási, one of the paper’s authors and a Dis­tin­guished Pro­fessor of Physics with joint appoint­ments in biology and the Col­lege of Com­puter and Infor­ma­tion Sci­ence. “Thanks to the links between com­po­nents, infor­ma­tion is dis­trib­uted throughout a net­work. Hence I do not need to mon­itor everyone to have a full sense of what the system does.”

Barabási’s col­lab­o­ra­tors com­prise Jean-​​Jacques Slo­tine of M.I.T. and Yang-​​Yu Liu, lead author and research asso­ciate pro­fessor in Northeastern’s Center for Com­plex Net­work Research, for which Barabási is the founding director.

Using their novel approach, the researchers first iden­tify all the math­e­mat­ical equa­tions that describe the system’s dynamics. For example, in a bio­chem­ical reac­tion system, sev­eral smaller reac­tions between periph­er­ally related mol­e­cules may col­lec­tively account for the final product. By looking at how the vari­ables are affected by each of the reac­tions, the researchers can then draw a graph­ical map of the system. The nodes that form the foun­da­tion of the map reveal them­selves as indis­pen­sible to under­standing any other part of the whole.

“What sur­prised me,” said Liu, “was that the nec­es­sary nodes are also suf­fi­cient in most cases.” That is, the indis­pen­sible nodes can tell the whole story without drawing on any of the other components.

The meta­bolic system of any organism is a col­lec­tion of hun­dreds of mol­e­cules involved in thou­sands of bio­chem­ical reac­tions. The new method, which com­bines exper­tise from con­trol theory, graph theory, and net­work sci­ence, reduces large com­plex sys­tems like this to a set of essen­tial “sensor nodes.”

In the case of metab­o­lism, the researchers’ algo­rithm could sim­plify the process of iden­ti­fying bio­markers, which are mol­e­cules in the blood that tell clin­i­cians whether an indi­vidual is healthy or sick. “Most of the cur­rent bio­markers were selected almost by chance,” said Barabási. “Chemists and doc­tors found that they happen to work. Observ­ability offers a rational way to choose bio­markers, if we know the system we need to monitor.”

By now you’ve probably heard of the Boeing 787 Dreamliners and the problems they had in their first weeks in the air. Basically, the Dreamliner is an extremely fuel-efficient airliner. It was the first to use composite materials to reduce weight and the first to use “large format” lithium-ion batteries.

Due to fuel leaks and spontaneous fires in the batteries that exceeded the normal growing pains of any new, complex system, the entire fleet was grounded in mid January. Today, Chicago Tribune reports that the Federal Aviation Administration has permitted Boeing to perform a single “ferry flight” to relocate one of their planes so they can continue investigations. There is still no conclusion about the cause of the failures and FAA and National Transportation Safety Board officials don’t expect one for at least a couple months.

I wanted to better understand the problem so I asked Northeastern researchers K.M. Abraham and Peter Manolios for the takes on it all. Abraham is a research professor in the Northeastern University Center for Renewable Energy Technologies with 30 years of experience in the world of lithium batteries. He was quoted in two recent Wired articles about Boeing’s troubles, here and here. Manolios is an associate professor of computer and information science who has worked with Boeing and NASA for nearly a decade. The Dreamliner team commissioned Manolios to build an algorithm (dubbed CoBaSa) that can automatically integrate the various safety systems onboard the plane.

Abraham explained that lithium-ion batteries are such a hot topic (pun not intended) because they can store up to ten times more energy  than traditional batteries. This is what makes them so energy efficient but it is also precisely why they carry potential safety hazards. It’s the difference between taking a match to a couple grains of gunpowder or a hand grenade full of the stuff.

Now imagine the gunpowder is in the same vessel as a lit match, and a polymer membrane about half the thickness of a human hair is the only thing separating the two. This is akin to the situation in a lithium-ion battery, where the membrane separates two chemical reagents that are highly reactive with one another. Any failures in the system could allow the two chemicals to come into contact (internally short circuit), which would mean certain death for the battery…and a big explosion.

Lithium-ion batteries are pervasive these days. They’re in our cell phones and our computers. They’re the sole power storage devices in all electric vehicles like the Nissan Leaf and Tesla Roadster. All told, Abraham said there are more than 10 billion lithium-ion batteries out there, powering our digital world one chemical reaction at a time. But that’s no reason to get into a tizzy, as my mom would say. The batteries in these devices have gone through extensive optimization steps over the years. The chance of a fire in any of them is about one in 10 million.

According to the press, the Dreamliner batteries were also manufactured according to accepted specifications. It could be, Abraham speculated, that the specifications relevant in the small-format batteries in our cell phones and electric cars simply aren’t  enough at the larger scale.

The real problem may also have stemmed from how the batteries were used, said Manolios. “What seems to have happened is that there was a very large demand placed on the batteries and while they were charging they caught on fire,” he explained, pointing to a Time Magazine article on the topic.

While Manolios’ CoBaSa algorithm wasn’t designed to detect battery failures or fuel leaks, it’s possible that it could be used to prevent the former, he said. “We used my algorithm to synthesize software architectures, which involved figuring out which cabinets to place avionics code on, subject to a very large number of declarative constraints,” he said.

“It seems likely that we could use CoBaSa to express constraints saying that the power demands to the battery do not exceed a particular limit.” CoBaSa could then synthesize a system that doesn’t demand more power than the battery can provide, he explained.

Abraham is taking another approach, working on entirely new systems. He invented a battery called lithium-air, which uses oxygen from the atmosphere as the [cathode] and is significantly less hazardous than those described above. “It’s still in the early stages,” he said, “but there’s a worldwide effort in making it a practical battery.”

Ultimately, we’ll have to wait for the results of Boeing’s field tests to know exactly what happened. But I think it’s important we not put this down in the record books as a fundamental flaw in lithium battery technology. There are loads of researchers in the world, and several at Northeastern, all figuring out ways to make safe energy-efficiency a tangible goal.

Andrea Parker, a newly appointed assis­tant pro­fessor of per­sonal health infor­matics and human-computer inter­ac­tion, believes in the power of using tech­nology to pro­mote health and well­ness among low-income minority populations.

“Tech­no­log­ical inno­va­tions in health have the ability to facil­i­tate col­lec­tive mobi­liza­tion and sup­port behavior change in a great number of people,” she explained.

Parker’s exten­sive body of research bears this out: in 2009, for example, she designed a mobile game called OrderUp! in which a dozen low-income African-Americans in a south­west Atlanta com­mu­nity assumed the role of a server in a neigh­bor­hood restau­rant. The goal of the game, Parker said, was for par­tic­i­pants to serve their vir­tual cus­tomers as quickly and health­fully as possible.

In a paper on the health impact of playing mobile games, Parker revealed that OrderUP! shifted user per­cep­tion of what con­sti­tutes a healthy meal. The fun and easy-to-play game, she wrote, “helped par­tic­i­pants learn more about eating healthfully.”

Users, she added, “started to reassess their own behav­iors and began to see how they could make and eat healthier foods themselves.”

For another research project, Parker designed an appli­ca­tion that allowed some 40 mem­bers of the south­west Atlanta com­mu­nity to share text mes­sages doc­u­menting their eating habits. The mes­sages, she said, were visu­al­ized on a large touch screen dis­play appli­ca­tion installed in a local YMCA.

At the end of an exten­sive three-month study, Parker found that par­tic­i­pants had come to think of them­selves as com­mu­nity health advo­cates. “These kinds of tech­nolo­gies are not just helping people change their own habits, but they are also devel­oping par­tic­i­pants’ iden­ti­ties as advo­cates for change,” she explained. “It’s exciting to think about the influ­ence these people could have on their social networks.”

Prior to joining the North­eastern fac­ulty, Parker served as a post­doc­toral fellow in the Everyday Com­puting Lab at the Georgia Insti­tute of Tech­nology. She earned her doc­torate in human-centered com­puting from Georgia Tech in 2011 and her bachelor’s degree in com­puter sci­ence from North­eastern in 2005

“I feel like I’m coming home,” Parker said. “It’s def­i­nitely an honor and a priv­i­lege to be a pro­fessor where I began my aca­d­emic career.”

Parker will hold joint appoint­ments in the Bouvé Col­lege of Health Sci­ences and the Col­lege of Com­puter and Infor­ma­tion Sci­ence. She hopes to uti­lize her inter­dis­ci­pli­nary exper­tise, she said, to tackle urban health research projects in Boston’s Rox­bury and Jamaica Plain neighborhoods.

“I have a real pas­sion for helping the under­served and under­standing why these pop­u­la­tions dis­pro­por­tion­ately expe­ri­ence dis­ease,” Parker explained, noting her interest in col­lab­o­rating on long-term field studies with Elmer Freeman, director of urban health pro­grams in Bouvé.

By bringing the world of com­puter sci­ence to life through fun and engaging class­room lec­turers, on the other hand, Parker hopes to expose stu­dents to the vast pos­si­bil­i­ties of an ever-expanding field of research.

“I want to help stu­dents under­stand how to design sys­tems that can mesh with people’s values and con­nect with them on an emo­tional level,” she said.

It’s game on at Northeastern’s new Playable Innovative Technologies (PLAIT, “Play It”) Lab. Combining expertise in art, design, computer and social sciences, engineering, and business from every college in the university, PLAIT represents a building movement in video-game design—one that will change the way we learn, train, and educate.

And it’s a field that has the attention of Washington. Last year, Secretary of Education Arne Duncan launched the “Digital Promise Initiative,” a national center created by Congress to advance technologies that transform teaching and learning. In addition, the White House tapped Constance Steinkuehler Squire, noted game-design expert, to become senior policy analyst in the White House Office of Science and Technology Policy.

Approximately 55 percent of the population plays video games, Squire says, and partnerships between academia and industry are crucial in developing games that will change the face of education.

To that end, Northeastern’s Magy Seif El-Nasr, an associate professor with dual appointments in the College of Arts, Media, and Design and the College of Computer and Information Science, was invited to the White House in July to consult at the Academic Consortium on Games for Impact. She was one of only 20 academic experts nationwide invited to do so.

Experts from industry, the National Science Foundation, National Institutes of Health, and the MacArthur and other foundations discussed strategies for building interdisciplinary collaboration in the serious-game industry.

“We need more people who design entertainment games professionally to get into this market,” says Seif El-Nasr, “and we need to partner academics with funding agencies so that we can push the frontier of academic research in serious games.”

With the establishment of the PLAIT Lab, eight combined game-design and interactive-media undergraduate degree programs, and a masters and doctorate in development, the university promises to play a leading role in bridging those arenas and advancing the field.

Why Gaming?

The idea that games can teach is not a new concept. Chess and the Chinese board game Go, for instance, have been used to practice conflict strategy and tactics for thousands of years.

Today’s digital descendants have already shown promise in crisis management, improving healthcare outcomes, and military training. They’ve also been used to harness the power of crowdsourcing to solve intricate scientific problems. Northeastern’s faculty experts believe games can do even more to address national challenges in health, education, and security.

Games Making an Impact

Take Seif El-Nasr’s research and game-design efforts, which will soon bear fruit. She has partnered with Vancouver-based IgnitePlay to fight obesity using real-time behavior tracking. She is collaborating with Bardia Aghabeigi, a PhD student in the College of Computer and Information Science, and with Mariya Shiyko, assistant professor of counseling and applied educational psychology, and Carmen Sceppa, associate professor of health science, both from Bouvé College of Health Sciences.

“It’s a platform to understand nutrition and adopt healthier behavior through exercise by using games as a motivation tactic for behavior change,” Seif El-Nasr says.

Anyone can play this online social-media game, but it’s targeted to women 30 to 50 years old. Users will track food and exercise and compare their progress against friends’. Users will also be able to set goals, or “quests,” for themselves, such as walking 10 minutes every day for the next 10 days, and get rewards for achieving such quests.

In addition to helping attack the national obesity problem, Seif El-Nasr is eager to evaluate the user data that will be generated to assess how and if the research tactics have the intended impact.

She also directs the Games User Experience and Research Lab at Northeastern, which consists of multidisciplinary research teams of graduate and undergraduate students developing new interaction techniques, or game mechanics, for healthcare and education games. In addition, they research and analyze data from game-user interactions to help create games with greater impact.

“For example, if we can understand users’ behaviors and feelings with a game like the IgnitePlay one, then we can increase the value of the game on health outcomes,” she says. “‘Remission’ is one such example, which has helped young people understand their cancer and increase their adherence to treatment meds in some cases.”

Casper Harteveld, who joined Northeastern this fall from the Delft University of Technology (The Netherlands) as assistant professor of art and design in the College of Arts, Media, and Design, integrates theories from organization science, psychology, and gaming into his research. He has collaborated with Dutch governing bodies responsible for levee maintenance on a game called Levee Patroller. He describes the crisis-management game as an example of “sense gaming”—games that let users prepare for disasters or situations that don’t occur often.

In his game, players deal with disastrous levee failures in The Netherlands, which occur only once every 100 years or so. Players have to make sense of the failures by first finding them and then reporting their observations, assessing and diagnosing the situation, and taking action if necessary.

“This type of serious game confronts players with certain phenomena and offers the tools to help them make sense of these phenomena,” he says. “With ‘Levee Patroller,’ it’s about failures; with another game, it might be training doctors to deal with rare diseases or unusual medical complications.”

The Right Mix

It’s not enough, though, to create a cool-looking game. Users must want to play while they’re learning a new skill or changing a behavior, and that’s what’s driving game-user research, analytics, and advances in the serious-games model. It’s also why programmers, engineers, and artists are only part of the design picture, and why it must also include experts in various subject domains. It takes an interdisciplinary village to create an effective video game.

Toward that goal, Harteveld uses a three-pillar model that he calls triadic game design. A game needs subject-matter experts to provide valid criteria (reality); storytellers or teachers to provide strategies and purpose (meaning); and programmers and artists to create an immersive, fun, engaging world (play).

Gillian Smith, assistant professor in the College of Arts, Media, and Design with a joint appointment in the College of Computer and Information Science, joined Northeastern from the University of California, Santa Cruz because of that very collaborative culture. “In game design, you need artists to be able to communicate with engineers, and engineers to be able to communicate with designers,” she says. “In the case of games for impact, we need to communicate with experts in health and education. Northeastern has that diversity of expertise.”

Seif El-Nasr, an international authority on digital-game research, embodies the multiple-discipline ethos. She earned graduate and undergraduate degrees in computer science, worked as a graphic designer, and took psychology courses to understand emotions. For other ways of expressing behavior, she took years of acting to be able to incorporate theater techniques in building better systems. She is currently working with Matthew Gray, assistant professor in the theater department, and with Derek Isaacowitz, associate psychology professor in the College of Science, to develop methodologies to evaluate emotion modulation and attention for use in game design.

In January, Northeastern will add even more firepower to its game-design movement. Two renowned international research experts will join the team as associate professors in the College of Arts, Media, and Design: Alessandro Canossa, who will focus on game design, psychology of play, and game-user research, and Anders Drachen, with expertise in game analytics and game-user research.

The PLAIT Lab gives Northeastern impressive momentum in this emerging field. “In the next several years, as a result of our interdisciplinary collaboration, we’ll be seeing innovative games addressing a variety of domains come out of the lab, as well as an improved understanding of how people learn through games and the technologies that make it easier to create games and interpret data coming from them,” says Smith.

Northeastern has built-in advantages that give the university an edge, says Harteveld: the culture of collaborative, interdisciplinary research, the graduate-campus system that’s expanding industry connections, and experiential learning, which gets students deeply involved in game design.

Above all is the need for innovative, relevant, and rigorous research, which is a principal mission of the PLAIT Lab.

“Serious-game design, once it matures a little more, could have a pervasive effect, transforming educational research and bringing new insights into how we learn. And because of our momentum, Northeastern can become a world player,” says Harteveld.