AI in Outpatient Practice for Diagnosing Aortic Stenosis and Diastolic Dysfunction
Trial statusRecruiting
Trial phaseNot listed
Trial typeObservational
Biological sexAll
Age60+
SponsorMayo Clinic
About this trial
Two recently developed artificial intelligence-enabled electrocardiogram (AI-ECG) models have been developed to detect aortic stenosis (AS) and diastolic dysfunction (DD). AI-ECG for AS has a sensitivity of 78% and specificity of 74%, and AI-ECG for DD has a sensitivity of 83% and specificity of 80%. However, these models have never been prospectively applied to diagnose AS or DD, which may be useful for patients and providers from a diagnostic and prognostic perspective and especially in settings where access to higher- level medical care is limited. In this study, we aim to determine the clinical utility of these AI-ECG models by prospectively applying them to an outpatient cohort and then completing a focused point-of-care ultrasound to evaluate those who are AI-ECG positive for AS and DD.
Eligibility criteria
This trial does not accept healthy volunteersQualifiers
≥ 60 years of age must have a clinical scheduled ECG performed.
Disqualifiers
< 59 years of age
Is not scheduled for a clinical ECG
Unable to provide consent.
Trial population
Patients who are completing an outpatient electrocardiogram (ECG) at the Mayo Clinic.
Trial design
Design model
Cohort
Time perspective
Prospective
Treatments tested in this trial
AI-ECG Dashboard
DevicePatients standard of care ECG's will be processed through the AI-ECG Dashboard
Point of care ultrasound (POCUS)
Diagnostic testPatients will undergo a ultrasound to confirm diagnosis of atrial stenosis or diastolic dysfunction.
Treatment groups
2,000 Participants
are divided into 1 treatment groupGroup A: Patients who are completing an outpatient electrocardiogram (ECG) at the Mayo Clinic.2 interventions
Trial outcomes
Primary outcomes
1
Number of patients with positive AI-ECG
Positive AI-ECG will be determined by the sensitivity, specificity, positive predictive value, and negative predictive value.
Time frame
Baseline2
Number of studies with reasonable image quality in patients with positive AI-ECG
Image quality will be determined by sonographers at the time of imaging and will be scored on a scale from 1-4: 1. Excellent , sufficient for publication 2. Good, sufficient for data analysis 3. Fair, just enough for data analysis without complete views 4. Poor, not usable for data analysis
Time frame
BaselineSecondary outcomes
1
Number of times the AI ECG and TTE (transthoracic echocardiogram) are statistically comparative
Will be compared using parametric (2-sample t-test) and non-parametric tests (Wilcoxon rank sum test) for continuous variables, and the χ2 test or Fisher exact test for nominal variables. A p-value of \< 0.05 will be categorized as significant for the statistical analysis
Time frame
BaselineSponsors and contacts
Click on the lead sponsor to view all of their trials.
Source ClinicalTrials.gov