Is artificial intelligence the pill that healthcare needs?

Scheduling nurses in the emergency department at St. Michael’s Hospital used to be a tedious four-hour-a-day job.

Now, it’s done in 15 minutes through an automated program designed by data scientists at Unity Health, where a team of more than 25 employees harnesses artificial intelligence and machine learning to improve care. Unity Health includes St. Mike’s, St. Joseph’s Health Center and Providence Healthcare.

The team also created an early warning system that alerts doctors and nurses if a patient is at risk of going to intensive care or dying.

The programs are just two of more than 40 that have launched since 2019 when the analytics service was founded, thanks in large part to Unity CEO Dr Tim Rutledge, who believes the technology can change. radically health care.

« It’s crystal clear in my mind that by harnessing data and using advanced analytics and artificial intelligence, we’re going to be able to transform healthcare globally, » Rutledge says. « And I really think we’re at an inflection point and we’re really going to start harnessing that data and making a huge difference. »

Countries around the world are exploring how they can use artificial intelligence to improve medical staff productivity and create better patient outcomes in the face of a shortage of doctors, nurses and clinicians.

The technology uses “computers and machines to mimic the problem-solving and decision-making abilities of the human mind,” as defined by IBM.

But there have been a number of concerns in healthcare about data privacy, security and quality, and the technology is still in its infancy.

“We have oceans of data in healthcare,” Rutledge says. « And really, until recently in recent years, that data hasn’t been used to its full potential. »

It is estimated that 15-35% of healthcare hours overall could be automated, according to a 2020 report by McKinsey and Co., a consulting firm, in collaboration with EIT Health, a division of the European Institute of Innovation. and technology

But where that number ultimately lands will depend on whether the medical community is willing to embrace artificial intelligence, say the report’s authors.

At St. Mike’s, the use of analytics has not only reduced human effort by automating tasks, but early data shows it has played a role in decreasing fatalities, says Muhammad Mamdani, vice president data science and advanced analytics at Unity.

« The benefits we’re seeing are quite substantial, » he says.

Mamdani heads one of the few hospital-based data science departments in Canada.

Unity founded the department with a $10 million donation from a Hong Kong philanthropist. The department is now funded by the hospital’s charitable foundation, a break from research that is usually privately or government funded.

Elsewhere in the province, artificial intelligence is being used in a project called GEMINI to measure the quality of care in the general medicine departments of 30 hospitals.

General practice wards see patients with multiple conditions, which, due to advances in medicine and an aging population, is typical of the majority of patients admitted to hospital.

The research, which began in 2015, found that in these services, there are « great variations across our province and even within hospitals in how patients are cared for depending on who their doctor happens to be, who happens to be their care team, or maybe which hospital they go to,” says Dr. Amol Verma, internal medicine physician at St. Michael’s Hospital and co-lead of GEMINI.

The data also showed a 56% increase in the number of GP patients between 2010 and 2017, an increase that far outpaced the growth in capacity, both in beds and in human resources, Verma says. This has contributed to long waits for admission from emergency departments as well as so-called hallway medicine.

As a result of GEMINI’s research, the provincial government, through Ontario Health, is funding the General Practice Quality Improvement Network to examine different aspects of clinical practice, such as length of stay of a patient, unplanned readmissions and the use of imaging and blood tests – all of which affect patient outcomes in hospitals.

Both projects are led by Verma and Dr. Fahad Razak, an internist at St. Mike’s Hospital.

“This highlights many opportunities for us to learn from each other, for us to identify providers who provide excellent care, and to try to spread and standardize very high quality care,” says Verma. .

Below, we take a closer look at how data science and artificial intelligence are used at Unity Health and Gemini.

CHARTWatch: Operational in St. Mike’s Internal Medicine Unit, CHARTWatch runs hourly in the background, gathering more than 100 variables about a patient, such as blood pressure, heart rate, lab results and demographic data. The data goes through a machine learning algorithm that predicts the risk of the patient going to the intensive care unit or dying in the next 48 hours. If the patient is determined to be high risk, the medical team is called and responds within an hour to determine next steps.

“All of this is driven by doctors,” says Mamdani. “The way our model works is that our doctors actually build these solutions with our data science team. We never do it in isolation.

Preliminary data shows that the algorithm predicts patient outcomes at least 15% better than clinicians and has resulted in a 15% reduction in mortality in high-risk patients.

MuskRAT: Developed for the St. Mike’s Multiple Sclerosis Clinic, the program summarizes a patient’s medical history over several years in seconds. The program summarizes key points, including symptoms and treatments, in a one-page visual that provides links to more detailed information.

Emergency room volume forecasting tool: A patient volume forecasting tool developed for St. Mike’s Emergency Department that helps determine needed staffing levels using historical data. “What he tells us is when we are going to have really bad days,” Mamdani says. « So number one we are mentally prepared, but also two, if we can recruit staff – and that is our biggest problem at the moment, although we know it is difficult to have the right number of people – at least we’re trying to prepare better. »

COBRA: A three-day advance warning system that predicts a shortage of beds due to patient demand, giving the hospital time to plan more efficiently and offload patients where they can.

Planning tool: Creating a schedule for St. Mike’s emergency department nurses was a constant stressor due to staffing rules, such as pairing a junior nurse with a senior nurse and a 48-hour limit about the time a nurse can work with another nurse in the same area of ​​the ward. The analysis team created a program that, with the click of a button, produces an optimized schedule, using the rules, for the next four days. The error rate, which refers to failure to follow scheduling rules, fell from over 20%, which Mamdani said is typical for many hospitals, to less than 5%.

GEMINI: GEMINI has partnered with the U of T and the Vector Institute to develop a tool, using artificial intelligence, to identify hospital rates of delirium, a condition that doubles the risk of in-hospital death. Delirium is an acute confusion that affects up to 40% of older people hospitalized for other illnesses. It’s usually caused by issues like bladder infections, dehydration, or medications and could be prevented in 20 to 40 percent of cases with intervention, Verma says.

Rates of delirium have typically been difficult to measure due to inconsistent terms used to document patient status, such as « confused » or « drowsy, » he says. The new tool captures these terms and uses information from electronic medical records to predict whether or not delirium has occurred.

« What we’re most excited about now is to start being able to provide hospitals with information on delirium rates in different units in their hospitals, » Verma says, and to work with partners to « really make sure that the systems and services needed to prevent delirium are deployed where they are needed.

It will be « one of the first times artificial intelligence has been used to measure healthcare quality and identify a preventable opportunity to improve healthcare quality. »

TO SHARE: