Dr. Antonis Armoundas was studying electrical engineering at the National Technical University of Athens in his native Greece when he decided the subject was too dry for him.

“I couldn’t find meaning in what I was doing,” he said. So he started looking into biomedical engineering, which applies the problem-solving tools of engineering to biology and medicine to improve people’s health. His interest piqued, he dove into a senior thesis on the subject and went on to earn a master’s degree in biomedical engineering from Boston University and a doctorate in nuclear engineering from the Massachusetts Institute of Technology.

Gradually, Armoundas’ passion for medicine grew.

“I felt like I was serving a purpose that was more meaningful,” he said. “Medicine became my greatest passion, but using engineering principles in medicine gave it meaning.”

Armoundas now studies the ways artificial intelligence, or AI, can improve health care. He is a principal investigator at the Cardiovascular Research Center at Massachusetts General Hospital in Boston and an associate professor of medicine at Harvard Medical School. He spoke about his AI work with American Heart Association News as part of “The Experts Say,” a series in which experts explain how they’re applying what they’ve learned to their own lives. The following interview has been edited.

How is AI used in cardiology?

We know that AI is being used in every aspect of cardiology, including the diagnosis, classification, and treatment of cardiovascular disease. I chaired the writing committee for a scientific statement from the American Heart Association earlier this year, where we discussed the key areas where it’s being applied. Some examples of how it’s being used include interpreting cardiac imaging and matching genetic variants with health-related traits to better understand genetic risk for disease.

We know that in some cases, AI has even surpassed the performance of medical experts in interpreting cardiac imaging. It can detect subtle, non-visually detected signatures in data to make a more accurate diagnosis.

How do you use AI in your work at Massachusetts General Hospital?

I focus on improving the monitoring of patients in and outside the hospital.

In terms of in-hospital monitoring, we want to integrate all available patient-related information to make the best possible decisions with these patients. That means being able to predict and prevent adverse events.

We also want to use AI to help determine the optimal time to discharge a patient. We don’t want to keep patients longer than they should or discharge them sooner than they should. AI can help make those decisions. It will integrate information that is gathered from different sources that changes dynamically during a patient’s stay in the hospital, such as electrocardiograms, blood pressure, and electronic health records.

Because we can’t have a doctor next to the patient all the time, AI can collect and integrate all this data, see if certain parameters are exceeded. Then it can make a recommendation, for example that patient X should stay an extra day.

In terms of out-of-hospital monitoring, we know that many, many people use wearable devices, such as smartwatches, that collect data, such as whether someone has an irregular heart rhythm. Many of the devices are not medical grade, so we can’t fully trust this data yet, but we are moving towards better technologies.

We want to use the data that these devices collect to make decisions about when a patient should come to the hospital. We don’t want to just bring someone to the hospital for a check-up, but we also don’t want to bring them too late if their condition has deteriorated. AI can help us make better decisions about when to come, which can not only improve care but also reduce costs, which is another major benefit.

How does AI take into account non-measurable data, such as how someone feels?

We need to acknowledge and integrate how the patient feels – their symptoms – into this process. We need more than just a person’s vital signs and how they might fluctuate.

But there are apps that can capture this by asking questions. How do you feel during or after exercise? What is your mental state? Are you hungry? It can be a questionnaire and the person can log into the app every day to report how they feel. All of these states of being can be integrated with the other data that we collect.

Do you use AI outside of work?

AI has permeated every aspect of our lives, whether we know it or not.

It is used in car navigation systems to help us assess traffic and choose the optimal route to get somewhere. It is used to determine whether an email is spam. It is used in social media to target our interests to certain topics. We may not pay attention to it or understand how it is used, but it has come into our lives to stay.

The fundamental principles behind AI are not new. They have been taught for years, but they were not called AI or machine learning back then. Today’s AI research and development is made possible by an explosion in the amount of data we collect, as well as the vastly increased computing power available to process that data.

When a new technology comes along, I might not use it directly in my work, but I do experiment with it. For example, I might take a paper I’ve written and I want to see how well a new application of AI can understand it, or whether it can generate an accurate summary of what I’ve written.

I’ve also had conversations with my son, Alkinoos, a 22-year-old second-year medical student, about a project he’s doing to compare a doctor’s opinion in making a diagnosis with how well AI can make a correct diagnosis based on the patient’s symptoms, physical examination, and diagnostic and laboratory tests. More importantly, he wants to know if AI can guide a doctor’s decision-making process. Could AI help improve prognosis or case management?

How can AI be used in medicine in the future?

The future is the integration of all of this data. We want to integrate all of these different aspects of clinical knowledge, genomics, data from wearable devices and electronic health records, and our medical history into systems that can make optimal predictions about a person’s health.

I am fascinated by the integration of data for multi-organ systems research – that is, how medicine is approached holistically, not focusing on a single organ problem, but on the body as a whole. For example, how a person’s mental state relates to cardiovascular disease.

We are all different and we need that personalized care to improve outcomes. This is the hope of AI in medicine, the transition to delivering equitable care and improving outcomes on an individualized basis. (American Heart Association News via AP)