A new research published in European Heart Journalresearchers reported the success and validation of medical artificial intelligence (AI) models to screen for cardiac amyloidosis, a progressive, irreversible type of heart disease.
The results showed that patients can benefit from receiving appropriate treatment faster as AI tools are highly accurate, superior to existing methods and potentially more accurate diagnosis.
What is cardiac amyloidosis?
Heart amyloidosis is a heart disease in which abnormal proteins accumulate in the myocardium, which hardens its ability to pump blood. Multiple life-supporting drug treatments for this condition have recently become available, but without early diagnosis, doctors have missed the opportunity to expand patient survival and quality of life.
Unfortunately, cardiac amyloidosis can be difficult to diagnose. This is because it is often difficult to distinguish it from other heart problems without a burden-free amount of testing. ”
Jeremy Slivnick, Maryland, co-lead author, cardiologist, University of Chicago School of Medicine
Development of AI for cardiology
The AI model is Mayo Clinic and Ultromics, Ltd, an AI echo imaging company. was developed by researchers. They trained neural networks to detect cardiac amyloidosis using routine cardiac ultrasound images known as echocardiograms.
The resulting AI model can analyze a single echocardiographic video video of the apical four-chamber view of the heart to rapidly detect amyloidosis in the heart and distinguish it from other similar cardiac conditions.
Uchicago Medicine has participated in 17 other hospitals around the world to validate and test the results of the algorithm and test large, multiethnic patient populations. They found that the AI tool showed an accuracy of 85% to correctly identify patients with cardiac amyloidosis, and 93% to correctly eliminate it. This efficacy is true across multiple types of cardiac amyloidosis in diverse populations.
In their analysis, Slivnick and his colleagues compared AI models with existing clinical scoring methods commonly used to detect cardiac amyloidosis. Their results showed significantly outweigh these traditional approaches, allowing physicians to easily determine who needs sophisticated imaging tests or further evaluation.
“It was exciting to ensure that artificial intelligence can provide clinicians with reliable information and enhance the decision-making process of professionals,” Slivnick said. “New treatments for cardiac amyloidosis are the most effective in the early stages of the disease, so it is important to use all the tools at your disposal to diagnose them as quickly as possible.”
Bring your AI to the clinic
The AI model has been FDA cleared and has already been implemented in several hospitals across the country, and researchers hope that its use will eventually spread to everyday cardiac care.
“This AI model provides a practical solution,” says Slivnick. “It automatically analyzes common echocardiogram views, allowing easy integration into everyday clinical practice without any hassle or sacrificing diagnostic accuracy.”
sauce:
University of Chicago Medical Center
Journal Reference:
Slivnick, JA, et al. (2025). Detection of cardiac amyloidosis from a single echocardiographic contrast video clip: a new artificial intelligence-based screening tool. European Heart Journal. doi.org/10.1093/eurheartj/ehaf387.
