Actively Recruiting
Enhanced Valves Interventions and Safe AI Generated End Results
Led by Montreal Heart Institute · Updated on 2025-10-09
21000
Participants Needed
15
Research Sites
52 weeks
Total Duration
On this page
Sponsors
M
Montreal Heart Institute
Lead Sponsor
C
Centre Hospitalier Universitaire de Bordeaux, FRANCE
Collaborating Sponsor
AI-Summary
What this Trial Is About
Researchers are evaluating an artificial intelligence (AI) system designed to improve the prediction of outcomes for patients undergoing transcatheter heart valve interventions for conditions affecting the aortic, mitral, and tricuspid valves. This non-interventional, retrospective study analyzes data from multiple specialized centers worldwide to validate AI algorithms that automatically analyze cardiac imaging and clinical data. The goal is to enhance patient selection, intervention planning, and outcome predictions while reducing human error and variability in image interpretation. The study collects medical imaging data, including multi-slice cardiac computed tomography (CT) and transesophageal echocardiography (TEE), along with preoperative clinical information from patients who have undergone various heart valve procedures. These include transcatheter aortic valve implantation (TAVI), transcatheter mitral valve implantation (TMVI), transcatheter tricuspid valve intervention (TTVI), and edge-to-edge repair procedures for the mitral (M-TEER) and tricuspid valves (T-TEER) using specific device generations. The AI framework applies deep learning, including convolutional neural networks, to segment anatomical structures and measure them accurately from images. Participants' data are analyzed retrospectively to compare AI-generated automated measurements with manual assessments and to evaluate the accuracy of AI predictions against actual patient outcomes at 30 days post-procedure. The study also monitors the performance of AI algorithms throughout an average of two years of retrospective data collection. This process aims to confirm the AI's ability to support clinical decision-making and improve intervention results in heart valve disease patients.
CONDITIONS
Brief Title
Enhanced Valves Interventions and Safe AI Generated End Results
Who Can Participate
Eligibility Criteria
You may qualify if you...
- Patients who have reached the age of legal majority under local laws.
- For TAVI group: Patients who have undergone TAVI with a third generation transcatheter heart valve and have a pre-procedural ECG-gated CT scan of optimal quality.
- For TMVI group: Patients who have had TMVI with a dedicated device and have an optimal quality CT scan.
- For TTVI group: Patients who have had TTVI with a dedicated device and have an optimal quality CT scan.
- For M-TEER group: Patients who have had M-TEER with G4 or newer MitraClip or G2 or newer Pascal, with pre-procedural TEE videos from Phillips or GE with clear mitral valve views at 40+ frames per second.
- For T-TEER group: Patients who have had T-TEER with G4 or newer TriClip or G2 or newer Pascal, with pre-procedural TEE videos from Phillips or GE with clear tricuspid valve views at 40+ frames per second and acceptable 3D reconstructions.
You will not qualify if you...
- For TAVI group: Patients who underwent valve-in-valve procedures.
- For TMVI group: Patients with valve-in-valve or valve-in-ring procedures.
- For TTVI group: Patients with valve-in-valve or valve-in-ring procedures.
- For M-TEER group: Patients with G3 or older MitraClip or G1 Pascal devices.
- For T-TEER group: Patients with G3 TriClip or G1 Pascal devices.
AI-Screening
AI-Powered Screening
Complete this quick 3-step screening to check your eligibility
Your Study Journey
Duration - 2 to 4 weeks
Participants are screened for eligibility to participate in the trial.
1 visit (in-person or remote) for eligibility assessment
Duration - Retrospective data collection over up to 2 years
Participants' pre-procedural cardiac imaging data such as CT and TEE are collected and analyzed using artificial intelligence algorithms to segment anatomical structures and perform measurements.
No additional visits; analysis based on existing imaging and clinical data
Duration - Up to 2 years retrospective follow-up
Participants' predicted transcatheter heart valve intervention outcomes from AI analysis are compared with actual clinical outcomes.
No participant visits; monitoring is based on existing clinical records
Trial Site Locations
Total: 15 locations
1
Montefiore Medical Center New York
New York, New York, United States, 10467
Actively Recruiting
2
Montreal Heart Institute, 5000 Rue Bélanger, Montréal
Montreal, Quebec, Canada, H1T 1C8
Actively Recruiting
3
St Michael's Hospital Toronto
Toronto, Canada
Actively Recruiting
4
St Paul's Hospital Vancouver
Vancouver, Canada
Actively Recruiting
5
Centre Hospitalier Universitaire (CHU) de Bordeaux, 12 rue Dubernat 33404 Talence cedex
Bourdeaux, France, 33404
Actively Recruiting
6
CHU Lille
Lille, France
Actively Recruiting
7
CHU Marseille
Marseille, France
Actively Recruiting
8
Centre Cardiologique du Nord Paris
Paris, France
Actively Recruiting
9
Institut Cardiovasculaire Paris-Sud Paris
Paris, France
Actively Recruiting
10
Centre Hospitalier Universitaire Rennes
Rennes, France
Actively Recruiting
11
Clinque Pasteur Toulouse - France
Toulouse, France
Actively Recruiting
12
University Medical Center Hamburg-Eppendorf
Hamburg, Germany
Actively Recruiting
13
Heart Valve Center Mainz
Mainz, Germany
Actively Recruiting
14
Istituto Clinico Città di Brescia
Brescia, Italy
Actively Recruiting
15
San Raffaele Heart Valve Center Milan
Milan, Italy
Actively Recruiting
Research Team
T
Thomas Modine, MD, PhD
W
Walid Ben Ali, MD, PhD
How is the study designed?
Study Type
OBSERVATIONAL
Masking
N/A
Allocation
N/A
Model
N/A
Primary Purpose
N/A
Number of Arms
5
Similar Trials
Frequently Asked Questions
Have more questions? Get in touch with our team for quick support
Not the Right Trial for You?
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here