Researcher harnesses AI to uncover hidden hospital risks and improve patient care

One in 10 Canadian patients experience an adverse event (AE), such as a fall or infection after surgery, while hospitalized. Not only do these events negatively impact patients, but they also place an economic burden on the health care system, estimated at $1.1 billion annually.

Most AEs are preventable and understanding more about how and why they occur could lead to an improvement in patient outcomes and cost savings for the health care system. However, traditionally, the data associated with these events is difficult for researchers to detect, especially at the population level.

That’s because traditionally AE detection relies on time consuming reviews of charts and coding systems that may not be consistent or accurate.

Electronic medical records, with their detailed data gathering and widespread usage across Canada, provide hope as a valuable resource in AE detection. However, they have not yet been paired with the technology that would allow researchers to assess the relevant information.

Dr. Guosong Wu, a postdoctoral fellow at the Centre for Health Informatics at the Cumming School of Medicine, is hoping to bridge that gap. He recently received a CIHR Postdoctoral Scholarship to develop electronic medical record-based computer algorithms that will automatically detect surgery related AEs.

Under the supervision of his mentors, Dr. Tyler Williamson, PhD, and Dr. Yuan (Shawn) Xu, PhD, Wu will use artificial intelligence techniques to mine data from linked medical record and coded data from Alberta. The resulting algorithms will be used to routinely detect AEs for patient safety surveillance and monitoring and to ultimately improve patient care.

“Electronic medical records have a lot of detailed data that can be used to improve patient care, but the data that is collected varies in format and detail, making it difficult for researchers to use it,” says Wu. “My work will streamline this process, allowing even text data (which is notoriously difficult to use) to be accessed for use in research.”

Wu earned his undergrad and master’s degree in China, working in public health focused research. He held a faculty position at Harbin Medical University for five years before starting his PhD in Health Services Research through a Cotutelle program between the University of Calgary and Harbin.  

Wu’s PhD training brought him to the University of Calgary, where he earned his PhD and is continuing his studies as a postdoctoral trainee.

Wu hopes to pursue an academic career in his chosen areas of focus: health services and health quality research.

“Population health research is important because it has the potential to impact so many people,” says Wu. “I am proud to be using my skills to improve our health care system and to benefit patients.”

The Libin Cardiovascular Institute’s Precision Medicine Initiative is harmonizing and combining Alberta’s world-renowned data sources into a single, 'super' database which will allow researchers to conduct life-changing cardiovascular research. Its goal is to predict future health, develop clinical decision-making tools and improve patient care.