[Image from CDD20 on Pixabay]
LV dysfunction is preventable with timely detection and medication, according to Yale. However, identifying the disease before symptom onset has not been feasible.
The Yale Cardiovascular Data Science Lab (CarDS) Lab team, led by Dr. Rohan Khera, developed the AI-based ECG interpretation. They published their findings in the journal Circulation.
Related: AI breakthroughs in medtech: 7 ways to enhance healthcare
According to the Yale team, diagnosing LV systolic dysfunction requires cardiac imaging. Technology and available expertise limit the broad screening for the condition. However, they say ECG represents the most accessible cardiovascular diagnostic test in clinical practice around the world.
The team included nearly 400,000 ECGs paired with data on heart dysfunction from imaging tests. They tested the algorithm in different formats with data from U.S. clinics and hospitals and a large community cohort in Brazil.
Khera said the findings demonstrated that a simple photo or scanned image of a 12-lead ECG can provide key insights on cardiac structure and function disorders. The findings allow for early diagnosis and treatment, plus future LV dysfunction risk identification, Khera added.
“This opens up the possibility to finally bring a screening tool for such disorders that affect up to one in 20 adults globally,” Khera said. “Their diagnosis is frequently delayed as advanced testing is either unavailable or only reserved for those with symptomatic disease. Now we can identify these patients with a simple web-based or smartphone application.”
Related: Cough-counting is harder than it sounds, but AI and smartphones can help
