A new partnership between Lakeridge Health and Durham College’s (DC) Hub for Applied Research in Artificial Intelligence for Business Solutions (the AI Hub) will test the power of artificial intelligence (AI) to predict how long an individual will wait in the emergency department (ED), improving decision making and the patient’s experience in the ED.
The 40-week project, led by DC faculty researcher Amit Maraj and a team of four research assistants from the Computer Programmer Analyst program, will result in a prototype for an AI-infused recommender system. If successful, this system would make individual wait-time predictions for ED visits based on the person’s condition, what ED they are visiting and the time of day and year. The predictions would also take into account a person’s individual health status and other factors, including staffing, the number of people waiting for care and the urgency of everyone’s needs.
“People waiting in the emergency department often feel frustration and anxiety because they don’t know how long they will be there or what to expect,” said Dr. Ilan Lenga, chief information officer and chief medical information officer of Lakeridge Health. “We are pleased to be able to harness the ingenuity of the students and faculty at Durham College to develop a system that will benefit the community and improve people’s experiences in the emergency departments.”
The program team will look at a set of actual, anonymized patient data from the past to build a system – using machine learning – that can predict with a high degree of accuracy their wait time given everything that was happening in the ED at the time that they visited.
“We are excited about this opportunity to collaborate with Lakeridge Health to optimize the use of our health-care system’s resources using artificial intelligence,” said Don Lovisa, president, DC. “The work we are doing in this project has the potential to be tremendously helpful for patients while showcasing the real-world application of AI in a health-care setting in a way that will positively impact the system as a whole.”
Personalized predictions are important because emergency departments must treat the most urgent patients first, and do not operate on a “first-come, first-served” basis.