A team of researchers developed an artificial intelligence algorithm that predicts potential hypotension — or abnormally low blood pressure — onset during surgery, according to a study published in Anesthesiology.
To create the algorithm, the researchers used machine learning, a type of artificial intelligence in which a computer learns over time, rather than having to be programmed like typical software. Using a dataset comprised of 1,334 patient records with 545,959 minutes of arterial pressure waveform recordings, researchers trained the machine learning model to predict hypotension onset based on patients' physiological data.
The researchers then validated the predictive algorithm with a second dataset of 204 patient records and 33,236 minutes of arterial pressure waveform recordings, including 1,923 episodes of hypotension. The algorithm accurately predicted a hypotensive event 15 minutes before it occurred in 84 percent of cases, 10 minutes before it occurred in 84 percent of cases and five minutes before it occurred in 87 percent of cases.
The researchers suggested the algorithm could prove helpful in avoiding complications associated with hypotension, such as postoperative heart attack and acute kidney injury.
"Physicians haven't had a way to predict hypotension during surgery, so they have to be reactive, and treat it immediately without any prior warning," Maxime Cannesson, MD, PhD, lead researcher and professor of anesthesiology and vice chair for perioperative medicine at UCLA Medical Center in Los Angeles, said in a June 11 statement. "Being able to predict hypotension would allow physicians to be proactive instead of reactive."
The FDA granted medical equipment company Edwards Lifesciences a De Novo request in March for software that uses the algorithm, which has been commercially available in Europe since 2016.