DigitHeart Echo: predicting heart ultrasound findings from a smartphone photo of your ECG
Official title: Cardiac Anatomical and Mechanical Properties Prediction From Electrocardiography (ECG) With Multi-Modal Representation Learning
Run by The University of Hong Kong · NCT07558109 · Data updated 2026-11-12
This test checks whether a smartphone app can predict echocardiogram findings from a photo of a 12‑lead ECG in adults scheduled for echocardiography.
Quick facts
| Study type | Observational |
|---|---|
| Ages | 18 Years and up |
| Sex | All |
| Enrollment | 478 (estimated) |
| Sponsor | The University of Hong Kong Academic / other |
| Locations | 1 site (Hong Kong) |
| Status | Recruiting |
| Started | February 14, 2025 |
| Main results expected | February 14, 2027 |
| Study ends | February 14, 2027 |
| Registry record updated | April 30, 2026 |
| Trial ID | NCT07558109 on ClinicalTrials.gov |
Is this trial a fit for you?
✓Likely a good fit
Adults (≥18 years) who are scheduled for clinical echocardiography, can give informed consent, and meet exclusion criteria (no recent echo, no pacemaker rhythm, no dextrocardia, no complex adult congenital heart disease, no ventricular assist device).
✕Probably not a fit
Patients with pacemaker rhythm, dextrocardia, complex adult congenital heart disease, ventricular assist devices, or those who recently had an echocardiogram are unlikely to receive benefit from the model predictions.
A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.
What this trial studies
Adults who consent will have demographics and medical history recorded and a 12‑lead ECG taken if no recent ECG is available. Research staff will photograph the ECG using the DigitHeart‑2 smartphone application and record predictions from DigitHeart‑2/MERL‑ECHO and other machine‑learning models. A cardiologist blinded to the model outputs will perform echocardiography according to American Society of Echocardiography guidelines, and those echo results will serve as the gold standard for accuracy comparisons. The study is observational and does not change participants' clinical care.
Why it matters
Potential benefit: If successful, this could allow clinicians to screen or triage structural and mechanical heart problems using a smartphone photo of a standard ECG, speeding diagnosis and referral.
How similar studies have gone: Prior machine‑learning analyses of standard 12‑lead ECGs have shown promise for detecting left ventricular dysfunction and other structural abnormalities, though using smartphone photos and these exact models is relatively novel.
Who can join: the full rules
Show full inclusion / exclusion criteria
Inclusion Criteria: * aged ≥ 18 years old * planned to have echocardiography performed * voluntarily agree to participate in the trial. Exclusion Criteria: * had echocardiography performed within 1 month * pacemaker rhythm on ECG * dextrocardia * complex adult congenital heart disease * ventricular assist device implantation.
Where it's running
-
Department of Medicine Queen Marry Hospital, Hong Kong
Hong Kong, Hong Kong Recruiting
Study contacts
- Study coordinator: Chun Ka Dr Wong, Clinical Assistant Professor
- Email: wongeck@hku.hk
- Phone: +852-22553111
Similar studies
Other studies for the same condition or treatment. Recruiting studies are listed first.
-
Differentiating heart disease in autoimmune vs. type 2 diabetes
This study tests whether adult-onset autoimmune diabetes (AID/LADA) and type 2 diabetes (T2D) differ in how diabetic heart disease develops, using advanced heart imaging in people with AID, T2D, and healthy controls.
-
Single-beat TE-e' for estimating left atrial pressure in critically ill patients
A noninvasive test using a single-beat TE-e' echocardiography measurement is being tried to see if it can reliably indicate elevated left atrial pressure in critically ill adults.
-
Cardiac Function and Anesthesia Depth During Heart Surgery
This study tests whether deeper or lighter anesthesia affects heart function during elective heart surgery with bypass in adults.
-
Automated Echocardiography Measurements Using Deep-Learning Software
This study tests whether US2.AI can automatically measure heart structure and function from standard echocardiograms as accurately as expert clinicians in adults.
-
Cardiac and carotid ultrasound reference values in healthy children in Italy
This study tests heart and artery measurements to establish normal reference values for healthy kids in Italy.
-
COMA.NET: Targeted treatment for coronary microcirculation problems
This trial will test whether tailoring medicines based on intracoronary microvascular testing improves quality of life for people with ischemia but no large artery blockages.
How to take part
- Review the eligibility criteria above with your treating physician.
- Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
- Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.