Anyone who has had a prolonged stay at the hospital knows it can be an unpleasant experience. Aside from dealing with your illness or injury, you must also bear the frightening task of absorbing volumes of information from your doctors and nurses, most of which is new and confusing and ultimately only adds to your apprehension.
The staff in hospitals tend to a large number of patients, and can rarely ever devote their time and effort to one person’s specific needs. Doctors themselves can be impersonal individuals, who must convey similar information repeatedly to patients with the same illnesses or ailments. And health care budget cuts — more likely to occur in times of economic crisis — only exacerbate these conditions.
Fortunately for patients, researchers at the University of Plymouth in the UK are seeking to ameliorate some of these issues. With help from robots.
The project is called ALIZ-E, and involves the placement of human-resembling robots in hospitals to deliver health education to patients.
If successful, the project could set a new precedent for the role of technology in health management. It may also push for the emerging practice of computerized care-giving, a revolutionary integration of robotics and health care that has already seen human-controlled robots performing surgery at a distance.
The project was initially inspired by animal-assisted therapy, in which animals, including dogs, cats, elephants, and dolphins, become part of a patient’s treatment. This treatment poses certain constraints however, most of which having to do with hygiene. Some patients cannot be exposed to certain animals, and the presence of animals in hospitals poses certain sanitation risks.
The solution reached by project coordinator Dr. Tony Belpaeme was to use robots. Similar to animals, robots could take on the role of a “companion” to patients, one that you could have an actual conversation with. The robots would also take on the educational roles of medical practitioners, lightening the load on doctors and nurses.
So far, the project has been implemented at one institution, the San Raffaele Hospital in Milan. The target audience of the project is children aged eight years and under, who have been diagnosed with diabetes.
“The kids have often been admitted to the hospital for one week, and that’s no fun!” says Belpaeme. “The robots will help educate the kids on the effects of poor diet and lack of exercise, as well as help with teaching kids about how to measure glucose levels and how to inject insulin.”
Belpaeme says that so far the robots have worked very well with kids. “Children are open to anything. They have a willingness to suspend disbelief. Based on the trials, the children don’t seem to see the robots as metal — they see an actual living creature!”
An important goal of the project is to test human-robot interactions. “How do humans respond to robots? Will they get attached? Why is it that children often describe emotions to robots? Why is it when the robots have less of a resemblance to humans, the kids want to converse with them?” Belpaeme asks.
The robots themselves are about one metre tall, are composed of metal and plastic, and resemble the remote-controlled robots sold in toy stores. They have two cameras built into them, two microphones, and an Internal PC that connects to the internet over Wi-Fi.
The robots’ programming is much more complex than their structural components. The program relies on a cloud computing system, in which the robot’s actions are controlled by databases in several locations.
“Language is controlled in Germany. Mannerisms are controlled in Belgium. Emotions are controlled in the UK. The robot is merely an interface in which computers in different locations make calculations and send the info back to the robot,” explains Belpaeme.
The benefits of the robots’ use have already been well established. Not only do they lighten the load on medical staff; researchers have found that the children tend to learn better if taught by the robots.
“We want to make the children’s stay fun. The kids not only retain more information when it’s fun, they recover faster!”
Belpaeme describes one example in which the robots quiz the children on healthy eating. “Which is better, a lollipop or an apple?” If the kids choose the apple, they win a certificate. The robots also play physical games with the children. For example, the robot can perform a dance, and the children will imitate it.
The project has been running for a year, and still faces some challenges. One of these problems is speech recognition, since children’s voices differ greatly from those of adults, and propagate at a higher pitch. Although children are less discerning of the robots’ structure, adults are more sensitive to their build and would only briefly see them as entertaining, then recognize them as machines. At present, the robots are only capable of simple interactions, and can only work within limited scenarios.
The project is funded by the European Commission, and research is conducted at the University of Plymouth. Belpaeme’s research primarily focuses on developmental robotics and machine learning.
On how he got involved in such a project, and on the difficulties within the field of artificial intelligence, Belpaeme comments, “I wanted to understand the human brain, so I decided to build an artificial brain. The problem with A.I. is that we don’t know how the human brain works. We need to understand how the brain works for artificial intelligence to progress.”
