MedPath

An Observational Study Using Artificial Intelligence (AI) Algorithms on Electrocardiography (ECG), Point-of-care Ultrasound (POCUS), and Transthoracic Echocardiophy (TTE) to Estimate the Under-diagnosis of Transthyretin Amyloid Cardiomyopathy (ATTR-CM) Across a Diverse Range of US Health Systems.

Active, not recruiting
Conditions
Transthyretin (TTR) Amyloid Cardiomyopathy
Registration Number
NCT07062848
Lead Sponsor
Yale University
Brief Summary

This is a multi-center, observational study with the overall objective to examine the scale of under-diagnosis for transthyretin amyloid cardiomyopathy (ATTR-CM) across a broad range of diverse health systems in the US using a fully federated deployment of an artificial intelligence (AI) toolkit of algorithms that detect ATTR-CM on electrocardiography (ECG), point-of-care ultrasound (POCUS), and transthoracic echocardiography (TTE).

Detailed Description

Not available

Recruitment & Eligibility

Status
ACTIVE_NOT_RECRUITING
Sex
All
Target Recruitment
1500000
Inclusion Criteria
  • Age 50-95
  • At least one retrievable ECG and/or 2D echo file (DICOM or equivalent video file) from EHR.
Exclusion Criteria
  • Unavailable key demographics (age, gender, race, ethnicity)
  • Individuals who have opted out of research studies

Objective-specific inclusion and exclusion criteria:

Primary Objective:

Additional exclusion criteria:

  • For subgroup analyses: when evaluating the prevalence of probable ATTR-CM status across demographic groups, we will exclude those with missing baseline demographic information (age, sex, race, geographic region).

Secondary Objective 1:

Additional inclusion criteria:

  • 'Cases': ATTR-CM diagnosis defined by ICD-10 codes (Table 1) OR abnormal bone scintigraphy testing consistent with ATTR-CM OR treatment with an approved transthyretin stabilizer or other ATTR-CM-specific therapy
  • 'Controls': any individuals not meeting the case definition. In these participants, we will consider all eligible ECG, POCUS, or TTE studies performed up to 12 months before diagnosis (first date of ICD code appearance, abnormal bone scintigraphy or treatment onset, whichever happened first) and any time after. 'Controls' will be drawn from ECGs, POCUS, or TTE studies performed in individuals not meeting the 'case' criteria above, including individuals who have never undergone dedicating testing or those who underwent e.g., bone scintigraphy, but with negative (or equivocal) findings.

Secondary Objective 2:

Additional inclusion criteria:

  • Having at least two years of follow-up time between the index test (ECG, POCUS, or TTE) and the date of analysis.
  • Having at least one healthcare encounter every two years across care settings from their first entry into the cohort through death or end of the follow-up period.

Study & Design

Study Type
OBSERVATIONAL
Study Design
Not specified
Primary Outcome Measures
NameTimeMethod
To describe the prevalence of probable AI-defined ATTR-CM in defined cohorts of individuals who have undergone standard cardiovascular investigations across a diverse network of US-based health care delivery systemsAt enrollment
Secondary Outcome Measures
NameTimeMethod
Validate the diagnostic performance of AI-enabled ECG, POCUS, and TTE algorithms for ATTR-CMAt enrollment
To examine the association between the AI-defined probability of ATTR-CM and the incidence of adverse cardiovascular eventsAt enrollment

Trial Locations

Locations (11)

University of California - San Francisco (UCSF) Health

🇺🇸

San Francisco, California, United States

Yale New Haven Health System

🇺🇸

New Haven, Connecticut, United States

Henry Ford Health

🇺🇸

Detroit, Michigan, United States

Mount Sinai

🇺🇸

New York, New York, United States

Duke Health

🇺🇸

Durham, North Carolina, United States

Providence Health

🇺🇸

Tigard, Oregon, United States

Medical University of South Carolina (MUSC) Health

🇺🇸

Charleston, South Carolina, United States

UT Southwestern Medical Center

🇺🇸

Dallas, Texas, United States

Houstin Methodist

🇺🇸

Houston, Texas, United States

University of Virginia School of Medicine

🇺🇸

Charlottesville, Virginia, United States

Scroll for more (1 remaining)
University of California - San Francisco (UCSF) Health
🇺🇸San Francisco, California, United States

MedPath

Empowering clinical research with data-driven insights and AI-powered tools.

© 2025 MedPath, Inc. All rights reserved.