AI-based Echocardiographic Quantification in Heart Failure

Launched by MACKAY MEMORIAL HOSPITAL · May 30, 2025

Keywords

ClinConnect Summary

The AIED Study is a clinical trial aimed at improving the diagnosis of a type of heart failure known as heart failure with preserved ejection fraction (HFpEF). This condition is often challenging to identify because its symptoms are similar to other types of heart failure. The study is exploring how artificial intelligence (AI) can help analyze echocardiogram images—tests that use sound waves to create pictures of the heart. By using AI, researchers hope to more accurately assess various heart functions and ultimately improve the way doctors diagnose and treat patients with HFpEF.

To be eligible for this study, participants must be at least 18 years old and have been admitted to the hospital for acute or chronic heart failure. They need to have undergone a specific type of heart ultrasound within 48 hours of admission. The trial is not yet recruiting participants, but it will include both men and women aged 65 and older who meet certain heart function criteria. If you take part in this study, you can expect to contribute to important research that may enhance how heart failure is understood and managed in the future.

Gender

ALL

Eligibility criteria

• Inclusion Criteria:
• • 1. Age ≥ 18 years.
• • 2. Admission for acute or chronic heart failure between January 1, 2017, and April 30, 2024.
• • 3. Transthoracic echocardiography completed ≤ 48 h after admission with diagnostic-quality DICOM cine loops (parasternal long/short axis and apical 2-/3-/4-chamber views plus Doppler and tissue Doppler).
• 4. Meets one of the two predefined phenotypes:
• • HFpEF: LVEF ≥ 50 % + typical HF signs/symptoms + objective diastolic dysfunction.
• • HFrEF: LVEF \< 40 % in keeping with guideline-defined systolic HF.
• Exclusion Criteria:
• • 1. Mid-range LVEF 40-49 %.
• • 2. Significant native or prosthetic valvular heart disease (moderate-to-severe) requiring surgery or trans-catheter therapy.
• • 3. Congenital heart disease, hypertrophic cardiomyopathy, restrictive or constrictive pericardial pathology, or prior cardiac transplantation/LVAD.
• • 4. Inadequate echocardiographic image quality (e.g., missing views, severe acoustic shadowing) precludes automated analysis.
• • 5. Hemodynamic instability preventing standardized imaging or data collection.
• • 6. Pregnancy.
• • 7. Concurrent enrollment in another interventional trial that may confound results of imaging or biomarkers.