Research aims to use human-like robot to reduce skin cancer in patients

Can robots transform the way we think and act? A team of researchers from McMaster University and Ryerson University are working with robots in a healthcare setting to see if they can encourage patients to modify their behaviour for the prevention of diseases and illnesses. Human-like robots have already made their way in many public settings around the world. From taking orders in restaurants, providing directions in airports and helping retailers in shopping malls, robots equipped with Artificial Intelligence and cognitive capabilities are changing the way we live.

The collaborative project with support from SOSCIP and IBM Canada Ltd., will integrate Pepper, a humanoid robot, into a clinical healthcare setting. The aim of the project is to educate patients on skin cancer and methods for preventing it, such as proper sunscreen application. In 2017 alone, 7,200 Canadians were diagnosed with melanoma skin cancer and 1,250 died from the same disease. Skin cancer is the most common cancer diagnosis and the most preventable cancer according to PreventCancer.org. A patient’s attitude and behaviour towards use of sunscreen and sun exposure is critical to long term reduction in this disease.

Interdisciplinary research team

Profs. David Harris Smith (Associate Professor, Department of Communication Studies and Multimedia, McMaster University), Frauke Zeller (Associate Professor, School of Professional Communication, Ryerson University), along with Hermenio Lima (Associate Professor, Head of Division Dermatologist and Clinical Immunologist, Faculty of Health Sciences at McMaster University) and Sue Becker (Professor, Department of Psychology Neuroscience & Behaviour, McMaster University) are aiming to achieve the first clinically tested implementation of a socially assistive robotics (SAR) innovation for healthcare communication.

The interdisciplinary collaboration came as a result of the successful hitchBot project, led by Profs. Harris Smith and Zeller, which saw an autonomous robot trek 10,000-kilometres on a hitchhiking journey across Canada. With its own social media account, the project garnered huge international interest and sparked discussions of its unique ability to build meaningful connections with humans. You can read more about hitchBOT here.

The duo has teamed up with Prof. Lima and Prof. Becker to better understand the implications of AI in improving communications around healthcare.

Meet Pepper

Pepper, which is manufactured by Softbank Robotics, is an endearing humanoid robot that has the ability to perceive emotions like joy and sadness and respond in kind. With its big eyes, gentle gestures and small stature (it stands at just four feet and weighs 62 lbs), Pepper commands feelings of protectiveness and trustworthiness from the people it interacts with.

“People tend to be quite charmed by Pepper,” said Prof. Harris Smith. “That’s really fascinating to me. When hitchBOT was destroyed, people shed tears. Meeting Pepper is like meeting a new type of being that withholds judgement.”

Access to partnership network and sophisticated computers

Outfitted with IBM Watson technology, Pepper will be able to carry on an intelligent conversation and interpret emotions and gestures. The first stage of the research is already underway with a small group of user studies to prototype the robot’s interactions with volunteer participants both at McMaster and Ryerson University. Users engage in a social conversation with Pepper designed to provide them with accurate and standard information, encouraging them to modify their skincare behaviour.

“The support of SOSCIP and IBM Canada has been critical to the project,” explained Prof. Harris Smith. “The SOSCIP infrastructure – providing access to the cloud computing and large memory system are vital. We are also working directly with IBM Canada to access the IBM Watson Cognitive Services for advanced speech recognition and processing.”

Virtual Pepper

Users are introduced to a virtual version of Pepper, accessed through a mobile application. The app uses SOSCIP’s cloud analytics platform to retrieve AI input and output, meaning patients can answer questions and interact with Pepper. The app allows users to set short and long-term goals and receive appropriate feedback, maintaining a rapport with Pepper and feeling a sense of reward for achieving their set goals.

Once the user trials have been completed and the ethics clearance is received, the team will enter clinical trials which will include two controlled groups of 120 patients. The first group will have consultations with a physician and receive the standard informational materials, such as brochures; the second group will interact with Pepper in a clinical setting and communicate with a virtual Pepper though a mobile app in which they can set and achieve goals.

Improving health communications

“It’s quite interesting to see how people behave in front of the robot,” admits Dr. Lima. “Pepper brings out the best in people. With Pepper, there is no judgement, patients tend to be honest with their responses about their behaviors.”

“Doctor-patient relationships are complex,” explain Prof. Zeller. “Some patients feel intimidated in a formal healthcare setting and don’t always communicate with their physician about their behaviours.”

Time is also a limitation for many physicians, and it’s not always clear that the information provided to the patient is understood and will be put into practice. There are also other factors – language and cultural differences – that aid miscommunication.

“So far everyone has been really positive about their interactions with Pepper, even to the point of somewhat adapting their style of communications,” explained Prof. Zeller.

The team insists that the project is about advancing our understanding of the kind of interactions that are possible with AI and social robotics and how they can enhance communications about health.

“Pepper is not about replacing doctors but about adding a valuable resource to the clinic – both for the clinic staff and the patients. Pepper can present data that is unbiased and solidly proven; in the future the presence of robots will help physicians in collecting information and data that can be used to improve the health outcomes of patients,” said Dr. Lima.

The research will be useful to the management of other types of diseases, such as hypertension and diabetes, which require patients to modify their lifestyle for better health outcomes.