Artificial intelligence (better known as AI) is well on its way to becoming an important part of the healthcare industry. Researchers have already proven that AI can detect skin cancer with more accuracy than dermatologists, diagnose deadly blood diseases, assist in radiology, and even help develop new medicines. Now, AI is being used to predict when patients are at risk for having a heart attack or stroke.
A study published in Circulation on February 14, 2020, found that heart attacks and strokes can be predicted by an AI tool. Dr. Kristopher Knott of the University College London’s Cardiovascular Science Institute carried out the study together with a team of researchers using cardiovascular magnetic resonance imaging (CMR) in tandem with AI. A CMR scan measures blood flow to the heart using a contrast agent. The heart muscle reacts to the agent, influencing blood flow. The stronger the blood flow, the lower the chance of heart vessels being blocked.
However, manually analyzing the results of the CMR can be very difficult and time-consuming. The painstaking process often produces less-than-accurate results, thus making it harder to recommend treatment to patients suffering from cardiovascular issues. So Knott and his team decided to bring AI into the equation.
The researchers trained an AI program to read the CMR scans of 1,000 patients and determine if their blood was flowing properly. When the process was complete, the team found that the AI was proficient at predicting patients’ outcomes. Not only was it able to verify that patients with reduced blood flow were more likely to suffer from a heart attack or a stroke, it could also diagnose–oftentimes better than a doctor–which patients would die or suffer from heart complications.
“The predictive power and reliability of the AI was impressive and easy to implement within a patient’s routine care. The calculations were happening as the patients were being scanned, and the results were immediately delivered to doctors,” said Dr. Knott.
His colleague, UCL cardiology professor James Moon, added: “Artificial intelligence is moving out of the computer labs and into the real world of healthcare, carrying out some tasks better than doctors could do alone.”
“We have tried to measure blood flow manually before, but it is tedious and time-consuming, taking doctors away from where they are needed most, with their patients.”
This discovery could pave the way for more accurate and efficient methods of diagnosing cardiovascular diseases. Breakthroughs like this one are always needed. Heart disease is the leading cause of death for both men and women in the United States, with over 640,000 Americans dying every year from heart-related illness.
Written by Natan Rosenfeld
