New research published in the Journal of Cardiovascular Development and Disease suggests artificial intelligence (AI) may be a game changer when it comes to recognizing and diagnosing cardiac amyloidosis (CA), including transthyretin amyloid cardiomyopathy (ATTR-CM).
Symptoms of CA, including shortness of breath, fatigue and swelling in the legs or abdomen, mimic other, more common health problems, leading to many cases being misdiagnosed until later stages of disease.
What’s more, traditional diagnostic methods like imaging, biopsies and blood tests may miss early signs of CA. This is where AI can step in to change that. Using smart tools like machine learning and deep learning models, a growing number of studies are demonstrating that AI can help spot signs of CA sooner.
Learn more about ATTR-CM testing and diagnosis
The researchers reviewed 30 studies on AI applications for traditional diagnostic tools like electrocardiogram (ECG), echocardiography, cardiac MRI, and nuclear scintigraphy.
In the ECG analysis, deep learning algorithms like LVH-Net achieved diagnostic accuracy as high as 95% in detecting CA. In some cases, these algorithms flagged the disease months before doctors made a formal diagnosis.
Other studies showed that deep learning models applied to echocardiograms could tell the difference between CA and other similar heart conditions like hypertrophic cardiomyopathy and heart changes related to hypertension.
In cardiac MRI and nuclear scans, deep learning algorithms were also found to offer fast, reliable and accurate insights in picking up signs of CA.
Of course, AI isn’t perfect, and the study authors point out important limitations. For example, most models were developed using retrospective data, often from single academic centers with limited study populations. This means the results might not yet be applicable to everyone in the general public.
Still, the researchers say the potential for AI in helping to diagnose CA is undeniable and call for more studies to better understand how it can improve screening and outcomes.
“We stand at the dawn of a new era in AI-driven healthcare, and this review underscores the dedication and perseverance of the early pioneers who successfully harnessed AI to diagnose CA,” the study’s authors concluded. “It is imperative that we build upon their foundational work to advance the seamless integration of AI into routine clinical practice.”
Sign up here to get the latest news, perspectives, and information about ATTR-CM sent directly to your inbox. Registration is free and only takes a minute.
