Exploring the limitless potential of artificial intelligence in health research
Academic spotlight: Professor of e-Health, Shang-Ming Zhou
Fuzzy logic. A phrase many may have heard without fully appreciating its meaning. An internet search offers the definition that it highlights how people make decisions based on imprecise and non-numerical information. But for Shang-Ming Zhou , it is more than just a phrase – it is two words that have defined his career. As we sit in his office on the University campus, with little other than a handful of books on a shelf and musings on a whiteboard for company, his enthusiasm for the topic more than adequately fills the room.
“When I first heard about fuzzy logic, my initial reaction was ‘wow’,” he says. “Until that point, everything about mathematics had been presented as black and white – that there was either a right answer or a wrong one. Now I understood there was a scale of possibilities when people were making decisions. It ignited a fire in me that still burns today.”
Even before that, the potential for Shang-Ming to excel had been evident. His hometown, Tai’an in China’s Shandong Province, is known for Mount Taishan, a UNESCO World Heritage Site and the most famous sacred mountain of China.
Shang-Ming also lived around 80 km from the birthplace of Confucius. Stories of the great philosopher were recounted in schools, and his influence was evident across the region. Against this backdrop, Shang-Ming won his first academic awards at primary school, added to his collection in high school, and secured a place to study mathematics and education at a local university focused predominantly on teacher training. However, it was in his first days as an undergraduate that his mind was really let loose.
In addition to timetabled studies, if lecturers mentioned something he found interesting, he used his spare time to learn about it. It was in such a context he first heard talk of fuzzy logic, and with no modules covering it he decided to teach himself. He also quickly established that he wanted to follow up his undergraduate degree with a masters. He set himself the ambitious goal of becoming the first student from his region to earn a masters place at Beijing Normal University, the country’s leading university at the time for mathematics. It was a goal he achieved and there, and in subsequent work with the Chinese Academy of Sciences, where he began to focus solely on fuzzy logic and applied artificial intelligence.
“The reason for looking mainly at its potential applications was quite simple – no-one else was,” Shang-Ming says. “I realised that in addition to it being an interesting theory, fuzzy logic and neural networks could be used to solve real world problems. At first, I looked at ways to use it in intelligent control systems and image processing. But in my mind, the potential for its use was almost limitless.”
The first opportunity to put this to the test in a health context came shortly after Shang-Ming arrived in the UK. After initially working at the University of Essex, he began working with De Montfort University on a study analysing health data associated with breast cancer patients. After that, he joined the medical school in Swansea for over ten years and the Digital Health & Care Innovation Centre at the University of Strathclyde for nearly half a year, and began using artificial intelligence to analyse the data captured routinely by GPs, hospitals and other healthcare settings. His aim was to identify patterns which could influence how people were treated, and ultimately deliver better care outcomes.
This dynamic inter-disciplinary environment also provided Shang-Ming with his first opportunity to work directly alongside practising clinicians. However, while he now had a clear picture in his mind of the difference artificial intelligence could make for them, it took a little longer to develop the same flow of trust in the opposite direction.
“From the very beginning, I felt so lucky to work with frontline health professionals and get the chance to use my skills to understand their problems,” he says. “They, on the other hand, were wondering how artificial intelligence would help them. It took them a little time to see its potential. But eventually, it became something they were genuinely interested in as well.”
Through these collaborations, and as artificial intelligence has grown in scope and sophistication, Shang-Ming – now part of the University’s Centre for Health Technology – has worked on some of the 21st century’s most pressing healthcare challenges. He has developed models, based on aspects of fuzzy logic and artificial intelligence, that are influencing the treatments of people with chronic conditions. He has also explored the potential for them to be used in fields from public health to emergency medicine.
Those trends continue in his current research. As Director of the NHS Kernow Datalab, he is leading a team developing AI and machine learning techniques to improve emergency services and health outcomes for respiratory patients. Funded by the Health and Care Research Wales, he is looking at ways to apply machine learning techniques to predict responses to chemotherapy in metastatic breast cancer. Building on previous work, he is looking for patterns associated with polypharmacy in epilepsy patients. And he is looking at effective ways to update the BabyCheck app, which aims to improve early identification of severe illnesses in infants under the age of six months.
In addition to helping improve outcomes for patients generally, one of Shang-Ming’s key focuses is to enable clinicians to treat people as individuals. With such enormous banks of health data, that has always been a major challenge. The use of AI provides a means to overcome that, to identify patterns in the data that ensure people are not just treated as an average but as a unique individual. Through his teaching on the University’s postgraduate programmes, across the Faculty of Health and the School of Engineering, Computing and Mathematics , Shang-Ming also aims to foster among his students not only a love of their subject but also the vision to see where it could take them. It is the same grounding, the same inspiration, offered by his high school teachers right at the beginning of his academic journey.
During the course of our conversation, there are several words that pop up repeatedly. Passion. Inspiration. Luck. And, of course, fuzzy logic. I challenge him over his use of the word ‘luck’, instead suggesting that everything he has achieved is in fact down to hard work, but he corrects me.
“I was lucky to develop a vision very early on that the medical domain was perhaps the best area to apply AI techniques,” he says. “It has given me the opportunity to work alongside amazing people in truly multidisciplinary teams. And the work I am doing can make a genuine difference to people’s lives. I think that counts as being pretty lucky whatever way you look at it.”
Centre for Health Technology
Bringing together digital health and health technology expertise from across the University to drive the development, evaluation and implementation of innovative technologies, products, services and approaches to transform health and social care.
