Skip to content

An International Multicenter, Multivendor Evaluation of the Free-Running Framework for Cardiac Function

Assessment and Validation of the Established Free-Running Framework for Cardiac Function by Magnetic Resonance Imaging (FAST-CMR): An International Multi-Center, Multi-Vendor Study at 1.5T

Status
Not yet recruiting
Phases
Unknown
Study type
Observational
Source
ClinicalTrials.gov
Registry ID
NCT07613398
Acronym
FAST-CMR
Enrollment
300
Registered
2026-05-29
Start date
2026-06-01
Completion date
2030-09-01
Last updated
2026-05-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Cardiac Diseases, Congenital Heart Diseases

Keywords

Free-running, Cine magnetic resonance imaging, Ventricular function, Multicenter, Multivendor, FISS (Fast- Interrupted Steady-State), Compressed sensing, Pragmatic clinical study

Brief summary

This project aims to evaluate a new cardiac MRI technique called the Free-Running Framework (FRF), which could simplify and accelerate the process of acquiring cardiac images. The investigators want to verify whether this method can provide functional heart measurements comparable to those obtained with traditional methods. More specifically, the goal of the study is to compare the measurements obtained with FRF to those obtained with standard sequences, to ensure they match and that this new approach can be reliably used in clinical practice. The FRF technique works differently from standard cardiac MRI. In standard exams, patients are asked to hold their breath several times and small electrodes (ECG) are used to monitor the heartbeat during the scan. These steps are needed to get clearer images of the heart as it moves. With FRF, these steps are no longer necessary: the scan is performed while patients' breath normally and without ECG monitoring. In addition, standard MRI takes multiple 2D slices of the heart, one after another. The FRF method instead captures a 3D image of the entire heart in one go, which can improve the consistency of the exam and reduce errors when doctors analyze the images later. This is all possible because the FRF method records data continuously and then organizes the images afterward based on how participants heart and breathing were moving during the exam. This helps the imager to get clear images of the heart, even without breath-holding or ECG monitoring. This project is aimed at individuals with heart disease who require cardiac MRI exams to monitor their health status (age ≥ 18 years) and are able to clearly understand the instructions provided by the research team. The investigators have already conducted small-scale technical and feasibility studies using FRF. These studies have shown that FRF is easy to use, faster than traditional methods, and provides image quality comparable to standard imaging techniques. The investigators now wish to evaluate its use in a clinical setting. More specifically, the investigators need to verify that FRF provides the same essential diagnostic information as standard techniques, so that it can be reliably used in future patient care. A maximum of 300 participants will be included in the investigation of this MRI technique between 2026 and 2031. This is a multi-center study, conducted internationally across 18 centers. This project is being carried out in compliance with Swiss legislation. The investigators follow all internationally recognized guidelines. The competent ethics committee has reviewed and approved this project.

Detailed description

Cardiovascular disease remains the leading cause of death in industrialized nations. While a range of diagnostic tools exists for cardiovascular disease detection and monitoring, magnetic resonance imaging (MRI) remains the only modality that enables a safe, non-invasive assessment of the heart without exposure to ionizing radiation. MRI allows for a comprehensive evaluation of cardiac anatomy, function, myocardial tissue characterization, and blood flow quantification, making it a powerful tool for cardiovascular diagnostics. Despite strong clinical evidence supporting its utility, cardiac MRI (CMR) remains underutilized, primarily due to the length and complexity of a traditional CMR exam. Several factors contribute to the length and complexity of a standard CMR exam. First, a standard CMR protocol relies heavily on two-dimensional (2D) image acquisitions, each requiring manual slice planning by an experienced technologist-a process that is both time consuming and highly dependent on operator expertise. Second, traditional image acquisition requires electrocardiogram (ECG) triggering to synchronize with the cardiac cycle, requiring additional setup and potentially introducing errors if the ECG signal is suboptimal. Finally, most conventional sequences rely on repeated patient breath-holding to minimize respiratory motion artifacts, which can be particularly challenging for individuals with severe cardiovascular disease, congenital anomalies, or limited compliance. Even for highly skilled personnel, this process is timeconsuming and inefficient. Consequently, a significant portion of the patient's time in the scanner is spent on preparation and planning rather than actual image acquisition, leading to prolonged exam durations and increased healthcare costs. Given these challenges, there is a strong need for simplified, automated, and time-efficient CMR acquisitions. In response to this challenge, the investigators research group has developed an innovative "free-running framework" (FRF)-a set of MRI methods that continuously acquire threedimensional (3D) image data across the entire cardiac cycle and throughout free breathing, irrespective of cardiac or respiratory motion. Unlike conventional CMR sequences that require separate, prospectively planned acquisitions for each imaging plane and time point, FRF employs continuous, self-navigated data acquisition. This eliminates the need for complex slice planning and enables retrospective reconstruction of cardiac motion, ensuring that imaging is both standardized and independent of user expertise. By leveraging advanced motion-resolved reconstruction algorithms developed by the investigators group, the investigators can derive both cardiac and respiratory motion from a single dataset, providing a fully automated, 3D whole-heart imaging approach. Once diagnosed, patients with cardiac diseases often require lifelong monitoring and repeated imaging assessments to guide treatment decisions and evaluate disease progression. This makes a non-ionizing imaging modality like CMR admirable. Nonetheless, the prolonged scan durations and intricate manual planning associated with traditional CMR limit its accessibility and practical feasibility. By eliminating the need for slice planning and reducing scan complexity, FRF has the potential to significantly improve imaging for cardiac disease patients by: * Standardizing imaging: producing user-independent results robust to anatomical variations. * Reducing scan times: without compromising diagnostic information. * Improving accessibility: enabling easier adoption in centers with less experienced technologists. The feasibility of FRF has been demonstrated in experimental, pre-clinical, and small observational clinical studies, with no observed adverse effects. This study aims to evaluate its clinical feasibility and efficiency in real-world cardiac disease patients across multiple cardiac institutions, serving as a precursor for larger validation studies and eventual clinical implementation. This study will provide the first systematic clinical evaluation of 3D FRF across multiple cardiac institutions in cardiac disease patients, assessing both its technical feasibility and potential workflow benefits in a real-world setting. In particular, this study will: * Evaluate whether diagnostic information from FRF matches or exceeds standard CMR. * Assess the efficiency gains and the impact of automated, self-navigated imaging on scan duration and patient comfort. If successful, this study will lay the groundwork for future multi-center validation clinical trial studies and eventual clinical integration, addressing a critical gap in CMR accessibility and efficiency.

Interventions

None listed

Sponsors

Matthias Stuber
Lead SponsorOTHER
Mayo Clinic
CollaboratorOTHER
University of California, Los Angeles
CollaboratorOTHER

Study design

Observational model
COHORT
Time perspective
CROSS_SECTIONAL

Eligibility

Sex/Gender
ALL
Healthy volunteers
No

Inclusion criteria

* Diagnosed with cardiac disease of any complexity and scheduled for a clinical cardiac MRI on a clinical 1.5T MRI scanner * Able to understand and provide written informed consent or, for minors, ability to provide assent with written consent from a parent/legal representative, per local law and site policy prior to the study * Ability to undergo MRI without contraindications * Agrees to be informed in the event of incidental findings

Exclusion criteria

* Severe claustrophobia preventing MRI completion * Presence of non-MRI compatible implants (e.g., pacemakers, certain metallic implants) * Contraindications to the intravenous contrast agent * Movement disorders or inability to remain still during the scan

Design outcomes

Primary

MeasureTime frameDescription
mean paired difference (bias) in left ventricular ejection fraction (LVEF) between 5D FISS-FRF and conventional 2D cine CMRBaseline (during study MRI acquisition)estimated with a two-sided 95% confidence interval within each pre-defined patient cohort. The study is designed to ensure that the half-width of this confidence interval does not exceed 1.25% LVEF per cohort.

Secondary

MeasureTime frameDescription
Scan EfficiencyBaseline (during study MRI acquisition)* Sequence acquisition time: scanner-reported acquisition duration for 5D FISS-FRF and conventional 2D cine imaging. This will include the timing of the necessary localizers specific to the imaging plane for each acquisition mode. * Workflow time: elapsed time from initiation of cine planning to completion of cine acquisitions, including slice planning, patient instruction, rest between breath-holds, and repeated acquisitions when required.
Quantitative Image QualityBaseline (during study MRI acquisition)o Contrast ratio and endocardial-to-blood-pool interface sharpness between 5D FISS-FRF and conventional 2D cine.
Qualitative Image QualityBaseline (during study MRI acquisition)o Likert scale (0-4) consensus grading by ≥2 expert CMR readers blinded to images from both techniques.
Failure RateBaseline (during study MRI acquisition)o Percentage of non-diagnostic or excluded scans.
Anatomical CoverageBaseline (during study MRI acquisition)o Extent of volumetric coverage of the entire heart and related great vessels.
Multiparametric CapabilityBaseline (during study MRI acquisition)o Presence of additional diagnostic insights from respiratory motion tracking of FRF.
Clinical Acceptability of LVEF Measurements (±5% Threshold)Baseline (during study MRI acquisition)o Exploratory analyses will assess whether observed agreement metrics are compatible with commonly used clinical acceptability thresholds (e.g., ±5% LVEF). These analyses are hypothesis-generating and intended to inform future confirmatory studies.
LV massBaseline (during study MRI acquisition)Comparison between 5D FISS-FRF and 2D conventional cine
Regional wall motion abnormalitiesBaseline (during study MRI acquisition)assessed using a 4-point likert scale per AHA segment (normal, hypokinesia, akinesia, dyskinesia)
Left and right atrial volumesBaseline (during study MRI acquisition)maximum and minimum volumes and total atrial emptying fraction derived from conventional 2D cine long-axis views compared to 5D FISS-FRF
Agreement of Automated Ventricular Ejection Fraction Measurements (LVEF, RVEF)Baseline (during study MRI acquisition)Agreement of automatically post-processed ventricular ejection fraction measurements (LVEF, RVEF) derived from 5D FISS-FRF with (1) standard post-processing of 5D FISS-FRF and (2) standard post-processing of conventional 2D cine cardiac MRI. Agreement will be assessed using appropriate statistical metrics.
Agreement of Automated Ventricular Function Measurements (LVEDV, LVESV, LVSV, RVEDV, RVESV, RVSV)Baseline (during study MRI acquisition)Agreement of automatically post-processed ventricular function metrics (LVEDV, LVESV, LVSV, RVEDV, RVESV, RVSV) derived from 5D FISS-FRF with (1) standard post-processing of 5D FISS-FRF and (2) standard post-processing of conventional 2D cine cardiac MRI. Agreement will be assessed using appropriate statistical metrics.
Feasibility of AI-Based Image ReconstructionBaseline (during study MRI acquisition)Percentage of reconstructed datasets that are considered clinically usable based on predefined quality criteria.
Reconstruction TimeBaseline (during study MRI acquisition)Comparison of reconstruction time between AI-based reconstruction and standard compressed sensing methods.
Accuracy of Ventricular Ejection Fraction MeasurementsBaseline (during study MRI acquisition)Comparison of left and right ventricular ejection fraction (LVEF, RVEF) between AI-based and standard reconstructions.
Accuracy of Ventricular Volume MeasurementsBaseline (during study MRI acquisition)Comparison of left and right ventricular end-diastolic volume (EDV), end-systolic volume (ESV), and stroke volume (SV) between AI-based and standard reconstructions.
Accuracy of Left Ventricular Mass MeasurementsBaseline (during study MRI acquisition)Comparison of left ventricular mass between AI-based and standard reconstructions.
Voxel-wise Image Similarity Between Reconstruction MethodsBaseline (during study MRI acquisition)Image similarity assessed using structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), and root mean squared error (RMSE).

Countries

Australia, Colombia, Germany, Italy, Japan, Netherlands, Singapore, South Africa, United Kingdom, United States

Contacts

CONTACTKaterina Eyre, PhD
katerina.eyre@chuv.ch+41213147513
CONTACTMatthias Stuber, PhD
matthias.stuber@chuv.ch+41213147534
PRINCIPAL_INVESTIGATORMatthias Stuber, PhD

CHUV

PRINCIPAL_INVESTIGATORTim Leiner, MD, PhD

Mayo Clinic

PRINCIPAL_INVESTIGATORKim-Lien Nguyen, MD

University of California, Los Angeles

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: May 30, 2026