Skip to content

Acorai Machine Learning Generalization (MLG) Study

A Multi-Site Observational Clinical Investigation to Collect Non-invasive Sensor Data During a Right Heart Catheterization and Train Machine Learning Models to Estimate Intracardiac Hemodynamic Parameters Evaluation of a Novel maChine leArning Model's Performance for Non-invasive inTracardiac pressURE Monitoring in Heart Failure - The CAPTURE-HF Trial

Status
Completed
Phases
Unknown
Study type
Observational
Source
ClinicalTrials.gov
Registry ID
NCT05835024
Acronym
CAPTURE-HF
Enrollment
1602
Registered
2023-04-28
Start date
2023-08-15
Completion date
2024-10-31
Last updated
2024-12-03

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

Conditions

Heart Failure

Brief summary

Acorai is developing a non-invasive monitoring system for the estimation of intracardiac hemodynamic parameters in patients with suspected or confirmed heart failure, and/or pulmonary hypertension, who require hemodynamic assessment. The device will be intended as a companion test or clinical decision support tool to be used and interpreted by qualified healthcare professionals to aid standard-of-care clinical assessment in identifying hemodynamic congestion and supporting personalized treatment of heart failure and pulmonary congestion. This study is part of the development of a non-invasive monitoring system for the estimation of intracardiac hemodynamic parameters. It will be conducted to collect the data needed to train the machine learning models retrospectively.

Interventions

DEVICEAcorai Sensor Data Collection (ASDC) system 1.0 - Phase 1

A supervised session that includes 10 minutes of patient information entry, 5 minutes of sensor recording time and 10 minutes of margin for setting up the system and patient, thus a total evaluation duration of approximately 25 minutes prior to the Right Heart Catheterization. No follow-up period for the subjects will be required for this clinical investigation. Patients will be followed as per standard of care, at the hospital. A single follow-up observational data collection will be conducted at 90 days, to collect the number of unplanned hospitalizations since the procedure visit. This does not involve active patient participation.

DEVICEAcorai Sensor Data Collection (ASDC) system 1.0 - Phase 2

A supervised session that includes 10 minutes of patient information entry, 5 minutes of sensor recording time and 10 minutes of margin for setting up the system and patient, thus a total evaluation duration of approximately 25 minutes prior to the Right Heart Catheterization. No follow-up period for the subjects will be required for this clinical investigation. Patients will be followed as per standard of care, at the hospital.

Sponsors

Acorai AB
Lead SponsorINDUSTRY

Study design

Observational model
COHORT
Time perspective
PROSPECTIVE

Eligibility

Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes

Inclusion criteria

* Subject is, at least, 18 years of age at the time of screening visit. * Subject is willing and physically able to comply with the specified evaluations as per the clinical investigation plan, as assessed by the investigator. * Subject is referred for invasive hemodynamic assessment with right heart cardiac catheterization. * Patient has provided written informed consent using the Ethics Committee/ Institutional Review Board approved consent form.

Exclusion criteria

* Discretionary exclusion when, in the opinion of the investigator, the inclusion of a potential subject is not in their best interest or not in the interest of compliant performance of the clinical investigation. * Subjects who are pregnant are excluded in the US

Design outcomes

Primary

MeasureTime frameDescription
Evaluation of the ML's performance to estimate pressureDay 0 to Day 90Performance of the ML model trained on data collected from the ASDC System to estimate left-sided filling pressure compared to right heart catheterization.

Secondary

MeasureTime frameDescription
Evaluation of the ML's performance to estimate other hemodynamic parametersDay 0 to Day 90The performance of the ML model trained on data collected from the ASDC System to estimate other hemodynamic parameters (such as right atrial pressure) compared to right heart catheterization.
Diagnostic accuracy of ML modelDay 0 to Day 90The diagnostic accuracy of the ML model trained on data collected from the ASDC System to detect clinically significant abnormal right heart catheterization measurements.

Countries

Belgium, Canada, Denmark, Sweden, United Kingdom, United States

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026