New AI That Predicts Heart Attacks Could Save Patients From Cardiac Death For 10 Years

AI-powered heart attack prediction is soon to come and could save a million lives.

The Bionics Institute 2022 Graeme Clark Oration, Melbourne's top science event, will feature a demonstration of this ground-breaking technology on Tuesday.

The artificial intelligence and bioengineering technique, created by American researcher Professor Natalia Trayanova from Johns Hopkins University, may save the lives of more than four million Australians who suffer from cardiovascular disease (CVD).

Digital Heart Twins

Professor Trayanova has developed "digital heart twins", which are virtual replicas of a person's heart that may be used to forecast the progression of heart disease, quantify the risk of heart attacks, and guide treatment decisions with the use of data-driven machine learning and biophysics-based modeling.

The leading cause of death worldwide continues to be heart disease. Up to 20% of fatalities can be attributed to sudden cardiac death (SCD) alone. SCD is an electrical problem that prevents the heart from beating normally rather than a heart attack, which occurs when arteries are clogged.

To estimate a patient's risk of sudden cardiac death over ten years, Professor Trayanova explained that the AI technology combines algorithms generated from MRI and PET scans in conjunction with deep learning of clinical data.

Trayanova said in a statement to Fast Company that doctors implant a defibrillator if a patient's ejection fraction is less than 30 percent; otherwise, they will not add one.

She noted that the 30 percent ejection fraction rule of thumb is also resulting in many people purchasing defibrillators when they don't require them.

Fast Company pointed out that according to one study, 40% of patients at some healthcare facilities received these devices even though only 23% of them were necessary.

Bridging Prognostic Gaps in SCD

The AI developed by Trayanova's lab aims to close the significant prognostic gaps in SCD. Her team built the system using patient records from 156 persons with cardiac issues over a ten-year period who volunteered to disclose their health information.

They supplied every piece of information from their treatment records, including 22 other items that might be pertinent, such as nationality, weight, medication usage, and hypertension, in addition to the MRIs of their hearts.

Researchers were able to identify underlying patterns, such as how scar tissue and other elements of a person's heart make them more susceptible to SCD, by entering all the MRI scans into a machine learning system, according to Fast Company's report.

After creating their software, researchers tested their technology against patient records from 60 healthcare facilities located around the United States. In its diagnoses, the AI performed better than doctors.

According to the Australian Institute of Health and Welfare, the country's economy loses almost $12 billion annually (according to 2018-2019 data) from CVD costs. This new AI tool may be essential to easing the strain on Australia's healthcare system.

By optimizing the patient's condition for the future rather than focusing on the patient's current condition, Trayanova envisions a future in which re-hospitalizations and repeat treatments are decreased with the introduction of her new invention.

This article is owned by Tech Times

Written by Joaquin Victor Tacla