If you’re a “Grey’s Anatomy” fan, chances are the word “LVAD wire” probably means something to you. Unlike the show, left ventricular assist devices (LVADs) possess capabilities far beyond the realm of hospital romance.
An LVAD is a mechanical pump designed for patients with advanced heart failure. They are implanted into the apex of the heart to assist the bottom left chamber (left ventricle) as it pumps blood out of the ventricle, through the aorta, and to the rest of the body. The pump is then attached to a cable leading out of the body and into an external computer, which provides alarms and messages that help operate the system. LVAD devices extend the lives of thousands of heart failure patients every year.
However, these devices do not come without the risk of serious complications. According to UW Medicine, upward to 20% of LVAD recipients experience right heart failure (RHF) due to the right ventricle not being able to withstand the sudden resurgence of blood flow from the pump. This results in a poorer chance of survival, or even immediate death, within days of implantation.
This outcome, often devastatingly unpredictable, piqued the curiosity of researchers at UW Medicine.
Through the use of a machine learning (ML) system trained to look for 186 different factors, experts have identified the top 30 pre-implant patient factors that are strongly associated with right-heart failure after LVAD implantation.
“A lot of patients, even though they survive, have a very poor quality of life and a major contributor to that is RHF,” Dr. Song Li, an assistant professor of cardiology at UW Medicine and one of the authors of this study, said. “It is difficult to predict beforehand, which is why we were interested in trying a new method to improve these predictions.”
This new method refers to the groundbreaking logistics of explainable ML. The ability to analyze hundreds of variables at the same time makes explainable ML far better equipped for the high-dimensional interactions between factors involved in this study.
“A lot of other AI machine-learning models are really just black boxes, limiting its usefulness in medicine,” Li said. “We need to have an explainable ML technique in order to apply ML properly.”
Standard ML models are notoriously limited to proving correlations without explanations, often referred to as black boxes.
Based on a sample population of 20,000 LVAD patients, the study found that the top five predictors of RHF are a patient’s INTERMACS profile, model for end-stage liver disease score, the number of inotropic infusions, hemoglobin, and race.
Written by Meha Singal via UW Daily.