Doctors turning to artificial intelligence to determine life expectancy

By  | 

More than half a million people will be diagnosed with congestive heart failure this year, and more than half of those people will die within five years.

Once diagnosed with congestive heart failure, doctors are now turning to artificial intelligence (AI) to determine how long patients can expect to live.

Stewart Seaward spends several hours a week working out in the gym.
Ironically, he was on a treadmill a few years ago when he found out he had a heart condition.

"I went to my family physician and he put me on a treadmill and said, looks like you've had a heart attack," Stewart said.

Every day, cardiologist Eric Adler sees patients like Stewart, suffering with congestive heart failure.

"My day job is deciding who needs heart transplants. And you can imagine, we don't want to be wrong," Adler said. "We don't have a crystal ball and that's often the most difficult situation we're put into."

Until now, cardiologists would compare tests themselves and make their best educated guess.

Avi Yagil wanted to know.

In the ICU and suffering from congestive heart failure, this physicist came up with the idea of using the same machine learning technology that Amazon and Google use to learn about their users and patients like himself.

Adler said, "It takes the information we have, and it looks at the relationship between them."

The algorithm creates a risk score from eight variables already collected for the majority of patients.

"Are you going to live longer than three months and are you going to live longer than two years?" he asked.

Giving doctors a multi-dimensional picture of each patient, the algorithm achieved an 88 percent success rate compared to just 50 percent before.

"That changes what resources we apply to them," Adler adds.

Stewart's working out hard, hoping to add years to his life and says the more information available, the better.

Doctor Adler and his team believes this is just the tip of the iceberg for using AI in the medical field.

Before it can be used in other hospitals, they're working on ways to protect the information.

They say, just like all your health information, there should be no way for anyone outside the hospital to access it.

MEDICAL BREAKTHROUGHS
RESEARCH SUMMARY
TOPIC: AI PREDICTS HOW LONG YOU HAVE TO LIVE
REPORT: MB #2722

BACKGROUND: When congestive heart failure (CHF) happens, blood and fluid can back up in your body and make it harder for your kidneys to flush out sodium and water. That makes you hold on to too much fluid, which then causes swelling. There is no cure for CHF. Your doctor may recommend medication to help lower blood pressure, relax your blood vessels, make your heart beat stronger, or ease swelling. Diet and lifestyle changes are known to help, too. Nearly 5 million Americans are currently living with CHF, and approximately 550,000 new cases are diagnosed in the U.S. each year. CHF affects people of all ages, from children and young adults to the middle-aged and elderly. Almost 1.4 million people with CHF are under 60 years of age; and 2 percent of persons age 40 to 59; and more than 5 percent of persons age 60 to 69 have CHF. (Source: https://www.webmd.com/heart-disease/heart-failure/congestive-heart-failure-facts and https://www.emoryhealthcare.org/heart-vascular/wellness/heart-failure-statistics.html)

SYMPTOMS, DIAGNOSIS AND TREATMENT: The diagnosis of congestive heart failure is based on knowledge of the individual's medical history, a careful physical examination, and selected lab tests. The treatment can include lifestyle modifications, addressing potentially reversible factors, medications, heart transplant, and mechanical therapies. It is important to know that both sides of the heart may fail to function at the same time, and this is called biventricular heart failure. This often occurs since the most common cause of right heart failure is left heart failure. The most common symptom of left heart failure is shortness of breath and may occur while at rest, with activity or exertion, while lying flat, while awakening the person from sleep, fluid (water, mainly) accumulation in the lungs or the inability of the heart to be efficient enough to pump blood to the organs of the body when called upon in times of exertion or stress. Chest pain or angina may be associated, especially if the underlying cause of the failure is coronary heart disease. (Source: https://www.medicinenet.com/congestive_heart_failure_chf_overview/article.htm)

AI USED TO PREDICT HEART ATTACKS: New research says artificial intelligence (AI) could be used to predict those at risk of a fatal heart attack up to five years in advance. Experts at the University of Oxford have developed a "fingerprint", or biomarker, using machine learning. The fat radiomic profile (FRP) detects biological red flags in blood vessels that supply blood to the heart, identifying inflammation, scarring and changes to the vessels. All these red flags point to a future heart attack. There's currently no routine method used by doctors to spot underlying red flags of a future heart attack. Researchers hope their AI tool will bridge this gap. Lead researcher, Charalambos Antoniades, professor of cardiovascular medicine and British Heart Foundation senior clinical fellow at the University of Oxford, said, "By harnessing the power of AI, we've developed a fingerprint to find 'bad' characteristics around people's arteries. This has huge potential to detect the early signs of disease, and to be able to take all preventative steps before a heart attack strikes, ultimately saving lives." (Source: https://www.digitalhealth.net/2019/09/artificial-intelligence-predicts-heart-attacks/)