Artificial intelligence technology is already being used in Ontario to predict people’s future health based on existing health data
Ontario doctors expect the introduction of artificial intelligence (AI) technologies to improve health care and make it easier to detect changes in patients’ health.
The Ontario Medical Association (OMA) held an online press conference Wednesday to outline how AI technology has been introduced into some areas of health care and how bigger changes are on the way.
OMA President Dr. Andrew Park said many Ontario doctors are already at the forefront of research into artificial intelligence and how to apply it to improve patient care and outcomes.
“We’re trying to use the power of artificial intelligence to make predictions about which individuals or populations are likely to get certain diseases, predict which patients will have complications, and optimize our scarce healthcare resources,” Park said.
He said one of the key points is that AI technology can quickly absorb reams of information and give doctors a more accurate picture of a patient’s condition.
“Artificial intelligence technologies can quickly and accurately analyze real-time patient data, including medical records, genetic information and lifestyle factors, and identify patterns that doctors might miss,” Park explained.
He said that along with improving patient outcomes, the technology could reduce the burden on the healthcare system by preventing diseases from progressing to more advanced stages.
One example was provided by Dr. Amol Verma, Professor of Artificial Intelligence Research and Medical Education at the University of Toronto.
He said there are concerns about unplanned transfers to the intensive care unit when a patient suddenly worsens and requires more focused care. Verma said that the St. Michaela in Toronto is using a new artificial intelligence tool called CHARTWatch Surgical that can monitor a patient in real time, minute by minute, around the clock.
“The question is whether there are any subtle clues or cues that AI could pick up earlier to alert us to patients who might be deteriorating and provide an early warning system,” he said.
Verma said all patient information is stored as an electronic medical record (EMR), which is then analyzed using CHARTWatch. Every hour, the program generates a new patient assessment that helps the clinical team get a better idea of the patient’s progress—or lack of progress.
“And so we implemented this CHARTWatch tool actually in about October 2020. And it’s been running for almost three years now. And what we’ve seen is compared to when we had the tool on our units, where the tool is present, we’ve seen A 26 percent reduction in unexpected deaths,” Verma said.
He said the tool also allows the healthcare team to speak with the patient and explore whether the patient wants to progress with treatment if they are ill or make a decision with regards to palliative care.
Dr. Ibukun Abejirinde, a scientist at the Women’s University Hospitals Institute for Health Systems Solutions and Virtual Care, outlined how she is part of a healthcare team that uses artificial intelligence to predict whether a person will develop diabetes 2 years ahead. The project is funded by the Canadian Institute for Advanced Research (CIFAR).
She said the development of artificial intelligence in predicting the possibility of future disease has created problems that the medical team is working on.
“One of them is the issue of social acceptability,” Abejirinde said.
“There are questions about the implicit bias of AI; questions about the ethics of using it; concerns about privacy and data breaches and who benefits,” she added.
“The second problem has to do with the fact that there is documented evidence of the potential and power of artificial intelligence. There are also documented bugs. a very large population,” Abejirinde said.
Another challenge, she said, is how to restructure Ontario’s health care system to be able to apply new knowledge in a way that provides earlier intervention to people who are likely to develop conditions like type 2 diabetes.
“So the way we see it is that if we have responsible AI that can be deployed sustainably at scale, we’ll be able to make better health care decisions that will better serve the population.”
One of the questions raised during the press conference was whether AI will eventually replace doctors. Park said he didn’t believe it would ever happen.
“I think there’s always a fear around emerging technologies that we have to address,” Park said. He cited a recent remark by the president of the American Medical Association.
“He said that AI will not replace doctors, but doctors who use AI will replace doctors who do not use AI.
Len Gillis covers healthcare and mining industry stories for Sudbury.com.
#Big #Doctors #improve #healthcare
